13.07.2013 Views

The IBM eServer BladeCenter JS20 - IBM Redbooks

The IBM eServer BladeCenter JS20 - IBM Redbooks

The IBM eServer BladeCenter JS20 - IBM Redbooks

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

<strong>The</strong> <strong>IBM</strong> Eserver<br />

<strong>BladeCenter</strong> <strong>JS20</strong><br />

Learn about <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong><br />

and blade server technology<br />

Understand the PowerPC 970<br />

microprocessor architecture<br />

Plan for, install, and<br />

configure <strong>BladeCenter</strong> <strong>JS20</strong><br />

ibm.com/redbooks<br />

Ben Gibbs<br />

Omkhar Arasaratnam<br />

Robert Arenburg<br />

Bradley Elkin<br />

Rogeli Grima<br />

James Kelly


International Technical Support Organization<br />

<strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong><br />

June 2005<br />

SG24-6342-01


Note: Before using this information and the product it supports, read the information in<br />

“Notices” on page xi.<br />

Second Edition (June 2005)<br />

This edition applies to the <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>, type 8842.<br />

© Copyright International Business Machines Corporation 2005. All rights reserved.<br />

Note to U.S. Government Users Restricted Rights -- Use, duplication or disclosure restricted by GSA ADP<br />

Schedule Contract with <strong>IBM</strong> Corp.


Contents<br />

Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii<br />

Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix<br />

Notices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi<br />

Trademarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii<br />

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii<br />

<strong>The</strong> team that wrote this redbook. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii<br />

Become a published author . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv<br />

Comments welcome. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv<br />

Part 1. Introduction to the <strong>BladeCenter</strong> <strong>JS20</strong>. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1<br />

Chapter 1. Introduction to <strong>BladeCenter</strong> and blade server technology. . . . 3<br />

Chapter 2. Hardware components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7<br />

2.1 Overview of the <strong>BladeCenter</strong> infrastructure . . . . . . . . . . . . . . . . . . . . . . . . 8<br />

2.1.1 <strong>BladeCenter</strong> chassis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8<br />

2.1.2 <strong>BladeCenter</strong> power options. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10<br />

2.1.3 Management Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13<br />

2.1.4 I/O modules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16<br />

2.1.5 Blade servers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21<br />

2.2 <strong>BladeCenter</strong> <strong>JS20</strong> . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22<br />

2.2.1 Base features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24<br />

2.2.2 Optional features. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27<br />

2.3 PowerPC 970 and PowerPC 970FX Microprocessors . . . . . . . . . . . . . . . 29<br />

2.3.1 Review of POWER and PowerPC Architecture . . . . . . . . . . . . . . . . 30<br />

2.3.2 Vector/SIMD Multimedia eXtension . . . . . . . . . . . . . . . . . . . . . . . . . 38<br />

2.3.3 Features of the PowerPC 970 and PowerPC 970FX . . . . . . . . . . . . 42<br />

Chapter 3. Software environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47<br />

3.1 Operating system support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48<br />

3.1.1 AIX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48<br />

3.1.2 Red Hat Enterprise Linux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48<br />

3.1.3 SUSE LINUX Enterprise Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49<br />

3.2 System management tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49<br />

3.2.1 <strong>BladeCenter</strong> Web interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49<br />

3.2.2 <strong>IBM</strong> Director . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50<br />

3.2.3 <strong>IBM</strong> Cluster Systems Management. . . . . . . . . . . . . . . . . . . . . . . . . . 50<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. iii


Chapter 4. Planning considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53<br />

4.1 Network planning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54<br />

4.1.1 Minimal network requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54<br />

4.1.2 High-performance, low-latency network requirements . . . . . . . . . . . 58<br />

4.1.3 Multiple <strong>BladeCenter</strong> environments . . . . . . . . . . . . . . . . . . . . . . . . . 59<br />

4.2 Operating system installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60<br />

4.2.1 Network installation planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60<br />

4.3 Systems management. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63<br />

4.3.1 <strong>IBM</strong> Director . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63<br />

4.3.2 <strong>IBM</strong> Cluster Systems Management. . . . . . . . . . . . . . . . . . . . . . . . . . 64<br />

Part 2. Implementing the <strong>BladeCenter</strong> <strong>JS20</strong>. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67<br />

Chapter 5. Hardware setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69<br />

5.1 Management Module configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70<br />

5.2 LAN Switch I/O Module configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73<br />

5.3 Blade server configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76<br />

5.3.1 Assigning names to blade servers . . . . . . . . . . . . . . . . . . . . . . . . . . 77<br />

5.3.2 Setting the boot sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78<br />

5.4 Firmware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79<br />

5.4.1 Management Module firmware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80<br />

5.4.2 LAN Switch I/O Module firmware . . . . . . . . . . . . . . . . . . . . . . . . . . . 82<br />

5.4.3 <strong>BladeCenter</strong> <strong>JS20</strong> firmware (BIOS) . . . . . . . . . . . . . . . . . . . . . . . . . 84<br />

5.4.4 Integrated Systems Management Processor firmware . . . . . . . . . . . 85<br />

5.5 Providing a console for the <strong>JS20</strong> blades . . . . . . . . . . . . . . . . . . . . . . . . . . 87<br />

5.5.1 Configuring Serial over LAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89<br />

5.5.2 Using Serial over LAN. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91<br />

5.5.3 Open Firmware interfaces. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94<br />

5.5.4 Specifying IP parameters to Open Firmware . . . . . . . . . . . . . . . . . . 96<br />

Chapter 6. Installing Linux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101<br />

6.1 Installing Linux. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102<br />

6.1.1 Configuring the sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102<br />

6.1.2 <strong>The</strong> zImage.initrd file. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107<br />

6.2 Configuring BOOTP and TFTP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109<br />

6.2.1 BOOTP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110<br />

6.2.2 Trivial File Transfer Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111<br />

6.3 Preparing an unattended install. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111<br />

6.3.1 Red Hat Kickstart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112<br />

6.3.2 SuSE AutoYaST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114<br />

6.4 Performing an unattended installation. . . . . . . . . . . . . . . . . . . . . . . . . . . 116<br />

6.4.1 Open Firmware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117<br />

6.4.2 mkzimage_cmdline: SuSE only. . . . . . . . . . . . . . . . . . . . . . . . . . . . 117<br />

6.5 Performing a network installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118<br />

iv <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


6.5.1 SUSE LINUX Enterprise Server 9. . . . . . . . . . . . . . . . . . . . . . . . . . 118<br />

6.5.2 Red Hat Enterprise Linux 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121<br />

Chapter 7. Installing AIX on the <strong>JS20</strong> . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123<br />

7.1 Minimal NIM installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124<br />

7.2 Adding resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127<br />

7.3 Adding machine information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130<br />

7.4 Preparing the NIM master . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134<br />

7.5 Preparing the client . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135<br />

Chapter 8. System management scenarios . . . . . . . . . . . . . . . . . . . . . . . 139<br />

8.1 <strong>IBM</strong> Director . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140<br />

8.1.1 Setting up an <strong>IBM</strong> Director Server . . . . . . . . . . . . . . . . . . . . . . . . . 142<br />

8.1.2 Installing an <strong>IBM</strong> Director Agent . . . . . . . . . . . . . . . . . . . . . . . . . . . 145<br />

8.1.3 Using Director to manage <strong>JS20</strong>s . . . . . . . . . . . . . . . . . . . . . . . . . . 147<br />

8.2 <strong>IBM</strong> Cluster Systems Management. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147<br />

8.2.1 Setting up a CSM management node . . . . . . . . . . . . . . . . . . . . . . . 148<br />

8.2.2 Installing and managing <strong>BladeCenter</strong> <strong>JS20</strong>s using CSM . . . . . . . . 154<br />

Abbreviations and acronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159<br />

Related publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161<br />

<strong>IBM</strong> <strong>Redbooks</strong> . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161<br />

Other publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161<br />

Online resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162<br />

How to get <strong>IBM</strong> <strong>Redbooks</strong> . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164<br />

Help from <strong>IBM</strong> . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164<br />

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165<br />

Contents v


vi <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Figures<br />

2-1 <strong>BladeCenter</strong> chassis front view . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8<br />

2-2 <strong>BladeCenter</strong> chassis rear view . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9<br />

2-3 <strong>BladeCenter</strong> power domains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11<br />

2-4 KVM Management Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14<br />

2-5 Management Module connectors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15<br />

2-6 <strong>BladeCenter</strong> <strong>JS20</strong> physical layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22<br />

2-7 <strong>BladeCenter</strong> <strong>JS20</strong> block diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23<br />

2-8 PowerPC logical processing model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34<br />

2-9 PowerPC registers used by application programs . . . . . . . . . . . . . . . . . 35<br />

2-10 Scalar and vector operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39<br />

2-11 PowerPC with VMX logical processing model . . . . . . . . . . . . . . . . . . . . 40<br />

2-12 VMX registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41<br />

2-13 PowerPC 970 internal organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43<br />

4-1 Network logical view . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55<br />

4-2 Myrinet network infrastructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59<br />

5-1 Management Module controls and indicators . . . . . . . . . . . . . . . . . . . . 71<br />

5-2 IP address configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72<br />

5-3 LAN Switch I/O Module IP address . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74<br />

5-4 LAN Switch I/O Module setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75<br />

5-5 Blade server naming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77<br />

5-6 Boot sequence of blades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78<br />

5-7 Viewing firmware levels on <strong>BladeCenter</strong> . . . . . . . . . . . . . . . . . . . . . . . . 80<br />

5-8 Updating the Management Module firmware . . . . . . . . . . . . . . . . . . . . . 81<br />

5-9 Firmware upgrade of 4-Port Gb Ethernet Switch Module . . . . . . . . . . . 83<br />

5-10 Blade firmware update . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86<br />

5-11 Serial over LAN components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88<br />

5-12 Serial over LAN configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90<br />

5-13 Serial over LAN status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91<br />

5-14 Management Module’s command line interface . . . . . . . . . . . . . . . . . . 92<br />

5-15 Setting CRLF in Windows 2000 Telnet . . . . . . . . . . . . . . . . . . . . . . . . . 94<br />

5-16 POST checkpoint codes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95<br />

6-1 YaST installation server configuration . . . . . . . . . . . . . . . . . . . . . . . . . 104<br />

6-2 Selecting the distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105<br />

6-3 Source Configuration pane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106<br />

6-4 Completing the setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107<br />

6-5 redhat-config-kickstart main window . . . . . . . . . . . . . . . . . . . . . . . . . . 112<br />

6-6 AutoYaST module within YaST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114<br />

6-7 Main AutoYaST window. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. vii


6-8 Language selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119<br />

6-9 Main menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119<br />

6-10 Installation menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120<br />

6-11 Source menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120<br />

6-12 Protocol menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120<br />

6-13 Installation location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121<br />

7-1 NIM configuration window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124<br />

7-2 Configuring as a NIM Master . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125<br />

7-3 Selecting the NIM interface and network name . . . . . . . . . . . . . . . . . . 125<br />

7-4 Choosing to initialize minimal NIM environment . . . . . . . . . . . . . . . . . 126<br />

7-5 Finalizing the NIM environment settings . . . . . . . . . . . . . . . . . . . . . . . 126<br />

7-6 lpp_source and SPOT configuration . . . . . . . . . . . . . . . . . . . . . . . . . . 127<br />

7-7 AIX source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128<br />

7-8 Default resource settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128<br />

7-9 Completing the resource installation . . . . . . . . . . . . . . . . . . . . . . . . . . 129<br />

7-10 Adding new machine information to NIM . . . . . . . . . . . . . . . . . . . . . . . 130<br />

7-11 Host name of new machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131<br />

7-12 Entering machine-specific information . . . . . . . . . . . . . . . . . . . . . . . . . 132<br />

7-13 Entering MAC address information . . . . . . . . . . . . . . . . . . . . . . . . . . . 133<br />

7-14 Installing an operating system from NIM . . . . . . . . . . . . . . . . . . . . . . . 134<br />

7-15 Setting BOS preferences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135<br />

7-16 Installation language panel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136<br />

7-17 Operating system selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136<br />

7-18 Installation and language options . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137<br />

7-19 Kernel options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138<br />

8-1 Selecting the type of database. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143<br />

8-2 Database connection information . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144<br />

8-3 Setting Director server encryption . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145<br />

8-4 Installing the Director Agent. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146<br />

8-5 Configuring Director Agent security . . . . . . . . . . . . . . . . . . . . . . . . . . . 146<br />

viii <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Tables<br />

2-1 Power domain A minimum power requirements . . . . . . . . . . . . . . . . . . 12<br />

2-2 Power domain B minimum power requirements . . . . . . . . . . . . . . . . . . 13<br />

2-3 Physical attributes of PowerPC 970 and PowerPC 970FX . . . . . . . . . . 42<br />

5-1 <strong>BladeCenter</strong> and <strong>BladeCenter</strong> <strong>JS20</strong> firmware components . . . . . . . . . 79<br />

5-2 Management Module commands for SoL . . . . . . . . . . . . . . . . . . . . . . . 93<br />

5-3 Network configuration information worksheet . . . . . . . . . . . . . . . . . . . . 96<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. ix


x <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Notices<br />

This information was developed for products and services offered in the U.S.A.<br />

<strong>IBM</strong> may not offer the products, services, or features discussed in this document in other countries. Consult<br />

your local <strong>IBM</strong> representative for information on the products and services currently available in your area.<br />

Any reference to an <strong>IBM</strong> product, program, or service is not intended to state or imply that only that <strong>IBM</strong><br />

product, program, or service may be used. Any functionally equivalent product, program, or service that<br />

does not infringe any <strong>IBM</strong> intellectual property right may be used instead. However, it is the user's<br />

responsibility to evaluate and verify the operation of any non-<strong>IBM</strong> product, program, or service.<br />

<strong>IBM</strong> may have patents or pending patent applications covering subject matter described in this document.<br />

<strong>The</strong> furnishing of this document does not give you any license to these patents. You can send license<br />

inquiries, in writing, to:<br />

<strong>IBM</strong> Director of Licensing, <strong>IBM</strong> Corporation, North Castle Drive Armonk, NY 10504-1785 U.S.A.<br />

<strong>The</strong> following paragraph does not apply to the United Kingdom or any other country where such provisions<br />

are inconsistent with local law: INTERNATIONAL BUSINESS MACHINES CORPORATION PROVIDES<br />

THIS PUBLICATION "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR IMPLIED,<br />

INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF NON-INFRINGEMENT,<br />

MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Some states do not allow disclaimer<br />

of express or implied warranties in certain transactions, therefore, this statement may not apply to you.<br />

This information could include technical inaccuracies or typographical errors. Changes are periodically made<br />

to the information herein; these changes will be incorporated in new editions of the publication. <strong>IBM</strong> may<br />

make improvements and/or changes in the product(s) and/or the program(s) described in this publication at<br />

any time without notice.<br />

Any references in this information to non-<strong>IBM</strong> Web sites are provided for convenience only and do not in any<br />

manner serve as an endorsement of those Web sites. <strong>The</strong> materials at those Web sites are not part of the<br />

materials for this <strong>IBM</strong> product and use of those Web sites is at your own risk.<br />

<strong>IBM</strong> may use or distribute any of the information you supply in any way it believes appropriate without<br />

incurring any obligation to you.<br />

Information concerning non-<strong>IBM</strong> products was obtained from the suppliers of those products, their published<br />

announcements or other publicly available sources. <strong>IBM</strong> has not tested those products and cannot confirm<br />

the accuracy of performance, compatibility or any other claims related to non-<strong>IBM</strong> products. Questions on<br />

the capabilities of non-<strong>IBM</strong> products should be addressed to the suppliers of those products.<br />

This information contains examples of data and reports used in daily business operations. To illustrate them<br />

as completely as possible, the examples include the names of individuals, companies, brands, and products.<br />

All of these names are fictitious and any similarity to the names and addresses used by an actual business<br />

enterprise is entirely coincidental.<br />

COPYRIGHT LICENSE:<br />

This information contains sample application programs in source language, which illustrates programming<br />

techniques on various operating platforms. You may copy, modify, and distribute these sample programs in<br />

any form without payment to <strong>IBM</strong>, for the purposes of developing, using, marketing or distributing application<br />

programs conforming to the application programming interface for the operating platform for which the<br />

sample programs are written. <strong>The</strong>se examples have not been thoroughly tested under all conditions. <strong>IBM</strong>,<br />

therefore, cannot guarantee or imply reliability, serviceability, or function of these programs. You may copy,<br />

modify, and distribute these sample programs in any form without payment to <strong>IBM</strong> for the purposes of<br />

developing, using, marketing, or distributing application programs conforming to <strong>IBM</strong>'s application<br />

programming interfaces.<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. xi


Trademarks<br />

<strong>The</strong> following terms are trademarks of the International Business Machines Corporation in the United States,<br />

other countries, or both:<br />

Eserver®<br />

Eserver®<br />

developerWorks®<br />

<strong>eServer</strong><br />

ibm.com®<br />

iSeries<br />

pSeries®<br />

xSeries®<br />

AIX 5L<br />

AIX®<br />

AS/400®<br />

xii <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong><br />

<strong>BladeCenter</strong><br />

Domino®<br />

<strong>IBM</strong>®<br />

Lotus®<br />

PowerPC Architecture<br />

PowerPC 601®<br />

PowerPC 603<br />

PowerPC 604<br />

PowerPC®<br />

POWER<br />

POWER2<br />

<strong>The</strong> following terms are trademarks of other companies:<br />

POWER3<br />

POWER4<br />

POWER4+<br />

POWER5<br />

<strong>Redbooks</strong><br />

<strong>Redbooks</strong> (logo) <br />

RETAIN®<br />

RS/6000®<br />

ServeRAID<br />

System/360<br />

3090<br />

Intel, Intel Inside (logos), MMX, and Pentium are trademarks of Intel Corporation in the United States, other<br />

countries, or both.<br />

Microsoft, Windows, Windows NT, and the Windows logo are trademarks of Microsoft Corporation in the<br />

United States, other countries, or both.<br />

UNIX is a registered trademark of <strong>The</strong> Open Group in the United States and other countries.<br />

Other company, product, and service names may be trademarks or service marks of others.


Preface<br />

Blade servers are a relatively new technology. <strong>The</strong>y have captured industry focus<br />

because of their modular design, which can reduce cost with a more efficient use<br />

of valuable floor space. <strong>The</strong>y offer simplified management, which can help to<br />

speed such tasks as installing, reprovisioning, updating, and troubleshooting<br />

hundreds of blade servers. You can do all of this remotely using one graphical<br />

console with <strong>IBM</strong>® Director systems management tools.<br />

In addition, blade servers provide improved performance by doubling current rack<br />

density. By integrating resources and sharing key components, costs decrease<br />

and availability increases.<br />

<strong>The</strong> <strong>IBM</strong> Eserver® <strong>BladeCenter</strong> boasts innovative modular technology,<br />

leadership density, and availability. It was designed to help solve a multitude of<br />

real-world problems.<br />

This <strong>IBM</strong> Redbook takes an in-depth look at the <strong>IBM</strong> Eserver <strong>BladeCenter</strong><br />

<strong>JS20</strong>. This is a two-way blade server for applications requiring 64-bit computing.<br />

It is ideal for computer-intensive applications and transactional Internet servers.<br />

This <strong>IBM</strong> Redbook helps you to install, tailor, and configure the <strong>IBM</strong> Eserver<br />

<strong>BladeCenter</strong> <strong>JS20</strong>.<br />

<strong>The</strong> team that wrote this redbook<br />

This redbook was produced by a team of specialists from around the world<br />

working at the International Technical Support Organization (ITSO), Austin<br />

Center.<br />

Ben Gibbs is a Senior Consulting Engineer with Technonics, Inc.<br />

(http://www.technonics.com) in Austin, Texas. He has over 20 years of<br />

experience with UNIX®-based operating systems. He started working with the<br />

AIX® operating system in November of 1989. His areas of expertise include<br />

performance analysis and tuning, operating system internals, and device driver<br />

development for the AIX operating system. He is also an <strong>IBM</strong> Learning Services<br />

instructor for advanced AIX classes. He was the project leader for this <strong>IBM</strong><br />

Redbook.<br />

Omkhar Arasaraknum is a Team Leader with <strong>IBM</strong> Global Services in Canada.<br />

He has worked with Linux since 1998 and uses SuSE, Red Hat, and Gentoo.<br />

Omkhar is the Linux technical lead within his Service Delivery Center and has<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. xiii


worked actively with Linux adoption within <strong>IBM</strong>. Omkhar has also written <strong>IBM</strong><br />

Redpapers about Microsoft® Windows® to Linux migrations for <strong>IBM</strong>.<br />

Robert Arenburg is a Senior Technical Consultant in the Solutions Enablement<br />

organization in the <strong>IBM</strong> Systems and Technology Group and is located in Austin,<br />

Texas. He has a PhD in Engineering Mechanics from Virginia Tech. He has<br />

worked at <strong>IBM</strong> for 13 years. His areas of expertise include high performance<br />

computing, performance, capacity planning, 3D graphics, solid mechanics,<br />

computational, and finite element methods.<br />

Bradley Elkin is a Senior Software Engineer for <strong>IBM</strong> U.S. He holds a PhD in<br />

Chemical Engineering from the University of Pennsylvania and has 17 years of<br />

experience in high performance computing. His areas of expertise include<br />

applications in computational fluid mechanics, computational chemistry, and<br />

bioinformatics. He has written several articles for <strong>IBM</strong> ^ Development<br />

Domain.<br />

Rogeli Grima is a research staff member at the CEPBA <strong>IBM</strong> Research Institute<br />

in Spain. He has three years of experience in applied mathematics. He has also<br />

collaborated on the development of <strong>BladeCenter</strong> <strong>JS20</strong> solution for<br />

bioinformatics.<br />

James Kelly is a Systems Architect in <strong>IBM</strong> Australia specializing in deep<br />

computing. He has 17 years of experience in system engineering and technical<br />

sales support. His current areas of expertise include the solution design,<br />

benchmarking, and installation of server clusters running either AIX or Linux, to<br />

support the installation of high performance computing applications.<br />

Thanks to the following people for their contributions to this project:<br />

ITSO, Austin Center<br />

Trina Bunting<br />

<strong>IBM</strong> Dallas<br />

Luciano Chavez<br />

Matt Davis<br />

Martin Gramlich<br />

Jon Mason<br />

Doug W Oliver<br />

Brian Rzycki<br />

<strong>IBM</strong> Austin<br />

Anton Blanchard<br />

<strong>IBM</strong> OzLabs, Canberra, Australia<br />

xiv <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Janet Ellsworth<br />

<strong>IBM</strong> Poughkeepsie, New York<br />

Become a published author<br />

Join us for a two- to six-week residency program! Help write an <strong>IBM</strong> Redbook<br />

dealing with specific products or solutions, while getting hands-on experience<br />

with leading-edge technologies. You'll team with <strong>IBM</strong> technical professionals,<br />

Business Partners and/or customers.<br />

Your efforts will help increase product acceptance and customer satisfaction. As<br />

a bonus, you'll develop a network of contacts in <strong>IBM</strong> development labs, and<br />

increase your productivity and marketability.<br />

Find out more about the residency program, browse the residency index, and<br />

apply online at:<br />

Comments welcome<br />

ibm.com/redbooks/residencies.html<br />

Your comments are important to us!<br />

We want our <strong>Redbooks</strong> to be as helpful as possible. Send us your comments<br />

about this or other <strong>Redbooks</strong> in one of the following ways:<br />

► Use the online Contact us review redbook form found at:<br />

ibm.com/redbooks<br />

► Send your comments in an Internet note to:<br />

redbook@us.ibm.com<br />

► Mail your comments to:<br />

<strong>IBM</strong> Corporation, International Technical Support Organization<br />

Dept. JN9B Building 905 Internal Zip 9053D003<br />

11501 Burnet Road<br />

Austin, Texas 78758-3493<br />

Preface xv


xvi <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Part 1 Introduction<br />

Part 1<br />

to the<br />

<strong>BladeCenter</strong> <strong>JS20</strong><br />

This part introduces and describes the hardware and software components of<br />

the <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>. It also presents planning considerations.<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. 1


2 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


1<br />

Chapter 1. Introduction to <strong>BladeCenter</strong><br />

and blade server technology<br />

This chapter presents an overview of <strong>IBM</strong> Eserver <strong>BladeCenter</strong> and blade<br />

server technology. It also covers the hardware for the chassis and the blades.<br />

<strong>The</strong> term blade server refers to a chassis that can hold a number of<br />

hot-swappable devices called blades. Blades come in two varieties: server blades<br />

and option blades.<br />

A server blade is an independent server that contains one or more processors<br />

and associated memory, disk storage, and network controllers. It runs its own<br />

operating system and applications. Each server blade within a system chassis<br />

slides into a blade bay2. It plugs into a midplane or backplane to share common<br />

infrastructure components. <strong>The</strong>se components may include power supplies, fans,<br />

CD-ROM, and floppy drives, Ethernet and Fibre Channel switches, and system<br />

ports.<br />

Option blades may be sharable by server blades. Option blades provide<br />

additional features, such as controllers for external I/O or disk arrays, power<br />

supplies, and so on.<br />

For organizations that seek server consolidation, <strong>BladeCenter</strong> centralizes<br />

servers for increased flexibility, ease of maintenance, reduced cost, and<br />

streamlined human resources. Companies that need to install new applications<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. 3


for e-commerce and On Demand Business can achieve speed while ensuring<br />

flexibility, scalability, and availability. For such enterprise requirements as<br />

file-and-print and collaboration, <strong>BladeCenter</strong> is designed to offer reliability,<br />

flexibility for growth, and cost effectiveness. And clients with compute-intensive<br />

applications that need highly available clustering can use the <strong>BladeCenter</strong> to<br />

help achieve high degrees of scalability and performance.<br />

Benefits of the <strong>IBM</strong> Eserver <strong>BladeCenter</strong> family<br />

<strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> family of products features a modular design that<br />

integrates multiple computing resources into a cost-effective, high-density<br />

enclosure for a platform that provides:<br />

► Lower cost in certain configurations: Due to efficiencies in power usage, heat<br />

emissions, and data center floor space utilization, when a configuration of<br />

other servers exceeds one rack, it is often less expensive over a multi-year<br />

period to use a single rack of <strong>BladeCenter</strong> chassis and <strong>BladeCenter</strong> <strong>JS20</strong>s.<br />

► Physical server consolidation: A <strong>BladeCenter</strong> server is an ideal way to<br />

replace many uniprocessor or 2-way servers to save space. 14U of rack<br />

space for 14 1U servers can be replaced with one 7U <strong>BladeCenter</strong> chassis.<br />

Additional rack space that is normally taken for Ethernet, Fibre Channel, and<br />

other switches is eliminated by the integrated switches in the <strong>BladeCenter</strong><br />

chassis.<br />

► High availability: <strong>The</strong> <strong>BladeCenter</strong> chassis offers redundant and hot-swap<br />

components that can prevent failure of the chassis or blade servers when one<br />

of those components fail. <strong>The</strong> chassis has redundant, hot-swap cooling,<br />

power, midplane features, management, and I/O switches. It also has a<br />

hot-swap media tray, with CD-ROM and floppy drive, that can be removed<br />

and serviced while all blade servers are operating.<br />

► Integrated switch technology: All service is from the front and rear of<br />

<strong>BladeCenter</strong>. <strong>The</strong>re is no need to slide the chassis out of the rack and remove<br />

the top cover for service. Also, numerous cables are eliminated, reducing<br />

both cabling cost and servicer or administrator time.<br />

► More integrated systems management features: <strong>The</strong> <strong>BladeCenter</strong> chassis<br />

includes a management module. This module eliminates the need for<br />

individual management adapters, such as the Remote Supervisor Adapter<br />

I or II, in each blade server for remote control. It also eliminates the need for<br />

RS485 interconnect cabling between the blade servers.<br />

► SAN optimization: A Fibre Channel switch module installed in a <strong>BladeCenter</strong><br />

chassis provides storage area network (SAN) access to all blade servers in<br />

the chassis without internal cabling.<br />

4 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


<strong>BladeCenter</strong> versus an <strong>IBM</strong> rack server<br />

<strong>BladeCenter</strong> is an ideal solution for certain environments. In other environments,<br />

an <strong>IBM</strong> rack server may be a better fit:<br />

► Need for a small number of servers: A <strong>BladeCenter</strong> chassis is required for one<br />

blade to be cost effective compared to some stand-alone servers. <strong>The</strong>refore,<br />

a chassis should be full or nearly full.<br />

► Need for existing Peripheral Component Interconnect (PCI) adapters:<br />

<strong>BladeCenter</strong> does not include adapter slots as shipped. An optional I/O<br />

Expansion Card feature offers one PCI slot per blade, limiting the blade to one<br />

integrated development environment (IDE) drive. <strong>The</strong> expansion feature<br />

supports PCI cards designed for <strong>BladeCenter</strong>. <strong>The</strong>refore, a mix of<br />

<strong>BladeCenter</strong> and traditional rack-optimized servers may be appropriate.<br />

► Need for large internal storage or external SCSI storage: <strong>BladeCenter</strong><br />

supports a maximum of 80 GB of IDE or 146.8 GB of SCSI storage internally.<br />

While there is no provision for external SCSI storage, there is an external<br />

Fibre Channel SAN storage option. Using the optional internal SCSI storage<br />

feature doubles the space requirement of the blade, cutting in half the number<br />

of blade servers that can be installed in a chassis.<br />

For large internal storage, the <strong>IBM</strong> Eserver xSeries® 345 is capable of<br />

holding up to 880.8 GB of hot-swap SCSI storage. All rack-mounted xSeries<br />

servers can support, either natively or via an optional ServeRAID adapter,<br />

external SCSI arrays.<br />

► Need for re-installation or repurposing at end of original: Stand-alone<br />

uniprocessor and 2-way servers can be distributed individually for use as<br />

departmental file/print servers and other low-horsepower uses. Because<br />

blade servers cannot be used without a chassis, the entire<br />

chassis-and-blades combo needs to stay together. <strong>The</strong> original chassis may<br />

be useful if older blade servers are discarded and newer, faster ones take<br />

their places in the chassis.<br />

► Power requirements: <strong>The</strong> data center is wired for 110-120 V power.<br />

<strong>BladeCenter</strong> requires 220-240 V power.<br />

Learn more about <strong>IBM</strong> Eserver <strong>BladeCenter</strong> and its components in the <strong>IBM</strong><br />

Redpaper <strong>The</strong> Cutting Edge: <strong>IBM</strong> Eserver <strong>BladeCenter</strong>, REDP-3581.<br />

<strong>The</strong> following chapters address the descriptions and considerations of the<br />

hardware and software components.<br />

Chapter 1. Introduction to <strong>BladeCenter</strong> and blade server technology 5


6 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Chapter 2. Hardware components<br />

2<br />

This chapter describes the major hardware components associated with the <strong>IBM</strong><br />

^ <strong>BladeCenter</strong> <strong>JS20</strong>. It begins by reviewing the core <strong>BladeCenter</strong><br />

infrastructure. <strong>The</strong>n it provides a detailed description of the <strong>BladeCenter</strong> <strong>JS20</strong><br />

and the range of options that you can order from <strong>IBM</strong>. Finally, this chapter<br />

reviews the <strong>IBM</strong> PowerPC® 970 RISC Microprocessor and <strong>IBM</strong> PowerPC 970FX<br />

RISC Microprocessor that are used as main processors on the <strong>BladeCenter</strong><br />

<strong>JS20</strong>.<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. 7


2.1 Overview of the <strong>BladeCenter</strong> infrastructure<br />

<strong>The</strong> <strong>BladeCenter</strong> <strong>JS20</strong> is designed to be installed in the <strong>IBM</strong> ^<br />

<strong>BladeCenter</strong> Type 8677. This section reviews the <strong>BladeCenter</strong> infrastructure that<br />

is relevant to the installation of the <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

2.1.1 <strong>BladeCenter</strong> chassis<br />

<strong>The</strong> core component of the <strong>BladeCenter</strong> infrastructure is the <strong>BladeCenter</strong><br />

chassis. Figure 2-1 illustrates the front view of the <strong>BladeCenter</strong> chassis. <strong>The</strong><br />

<strong>BladeCenter</strong> chassis is designed to be installed in an industry standard 19-inch<br />

rack. Each <strong>BladeCenter</strong> chassis occupies seven rack units. You can install up to<br />

six <strong>BladeCenter</strong> chassis in a single 42U rack such as the <strong>IBM</strong> NetBAY42<br />

Enterprise Rack.<br />

In the front view of the <strong>BladeCenter</strong> chassis, you see these standard features:<br />

► Fourteen bays where you can install blade servers<br />

► A media bay containing one CD-ROM drive, one diskette drive, and a<br />

Universal Serial Bus (USB) port that can be dynamically assigned to any<br />

single blade server in the <strong>BladeCenter</strong> chassis<br />

Figure 2-1 <strong>BladeCenter</strong> chassis front view<br />

8 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Figure 2-2 illustrates the rear view of the <strong>BladeCenter</strong> chassis. In the rear of the<br />

<strong>BladeCenter</strong> chassis, you see these standard features:<br />

► One management module<br />

► One bay where you can install an optional redundant Management Module<br />

► Four bays where you can install optional input/output (I/O) modules<br />

► A redundant pair of power supply modules<br />

► Two bays where you can install an additional pair of redundant power supply<br />

modules<br />

► A redundant pair of blowers<br />

Figure 2-2 <strong>BladeCenter</strong> chassis rear view<br />

Chapter 2. Hardware components 9


Attention: Nonredundant power is not supported in <strong>BladeCenter</strong> products.<br />

Power modules must always be present in power bays 1 and 2. When any<br />

blade server or option is in blade bays 7 through 14, power modules must be<br />

present in all four power-module bays.<br />

If a power module fails or an ac power failure occurs, <strong>BladeCenter</strong> units<br />

configured for redundant power operation operate in a nonredundant mode, and<br />

the blower modules runs at full speed. You must replace the failing power module<br />

or restore ac power as soon as possible to regain redundant power operation and<br />

to reset the blower modules to their normal operating speed.<br />

As of the date of this publication, these four <strong>BladeCenter</strong> power-module options<br />

are supported:<br />

► <strong>IBM</strong> <strong>BladeCenter</strong> 1200W Power Supply Module (part number 48P7052)<br />

► <strong>IBM</strong> <strong>BladeCenter</strong> 1200W to 1400W Power Supply Upgrade Kit (part number<br />

90P0197)<br />

► <strong>IBM</strong> <strong>BladeCenter</strong> 1800W Power Supply Module (part number 13N0570)<br />

► <strong>IBM</strong> <strong>BladeCenter</strong> 2000W Power Supply Module (part number 26K4816)<br />

Not visible in either view of the <strong>BladeCenter</strong> chassis are the redundant pair of<br />

midplanes in the center of the <strong>BladeCenter</strong> chassis that interconnect the blade<br />

servers installed in the front, and other modules that are installed in the rear of<br />

the <strong>BladeCenter</strong> chassis. <strong>The</strong> midplanes support the hot-plugging of blade<br />

servers and the other <strong>BladeCenter</strong> modules.<br />

Note: <strong>IBM</strong> offers a variant of the <strong>BladeCenter</strong> called the <strong>BladeCenter</strong> T. <strong>The</strong><br />

<strong>BladeCenter</strong> T is designed to meet the special needs of the<br />

telecommunications industry and has been tested for NEBS compliance.<br />

Since <strong>BladeCenter</strong> <strong>JS20</strong> is not supported in the <strong>BladeCenter</strong> T at this time, we<br />

do not describe <strong>BladeCenter</strong> T and its associated features in this book.<br />

2.1.2 <strong>BladeCenter</strong> power options<br />

<strong>The</strong> standard redundant power supplies are installed in power bays 1 and 2 of<br />

the <strong>BladeCenter</strong> chassis, and provide power to the first six blade server bays.<br />

This is known as power domain A.<br />

To install blade servers in the remaining bays 7 through 14, you must install an<br />

additional pair of redundant power supply modules in power bays 3 and 4. This is<br />

known as power domain B. Figure 2-3 shows these two power domains.<br />

10 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Figure 2-3 <strong>BladeCenter</strong> power domains<br />

Chapter 2. Hardware components 11


Attention: Power supply modules must always be installed in pairs. You can<br />

have either two power supply modules (in power bays 1 and 2) or four power<br />

supply modules (in power bays 1 through 4). Both power supply modules in a<br />

redundant pair should have the same capacity.<br />

When setting up the power requirements for the <strong>BladeCenter</strong>, remember the<br />

following criteria:<br />

► Install the power modules in pairs in a domain. <strong>The</strong>y must match each other in<br />

capacity (wattage, amperage, and so on).<br />

► A power domain operating above the capacity of a single power module<br />

results in a nonredundant power condition.<br />

► In a pair of power modules, a power module that is not connected to a<br />

200–240 volt ac power source results in a nonredundant power condition.<br />

► To provide true redundant power, connect <strong>BladeCenter</strong> power modules 1<br />

and 3 to a different 200–240 volt ac power source than to power modules 2<br />

and 4.<br />

► Connect an installed power module to an ac power source. <strong>The</strong> module must<br />

not be used as a filler.<br />

Attention: To maintain proper system cooling, each unoccupied blade bay<br />

must contain a filler blade as well as each unoccupied I/O,<br />

management-module, or power-module bay. An installed power module must<br />

be powered, but must not be used as a filler.<br />

Table 2-1 shows the minimum power requirements for power domain A.<br />

Table 2-1 Power domain A minimum power requirements<br />

Sum of <strong>JS20</strong> power units Minimum power module required<br />

Less than 7.4 1200 W (labeled 7.5A)<br />

7.4 up to less than 9.0 1400 W (labeled 9A)<br />

9.0 up to less than 10.0 1800 W (labeled 12A)<br />

10.0 or greater 2000 W (labeled 13.5A)<br />

12 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Table 2-2 shows the minimum power requirements for power domain B.<br />

Table 2-2 Power domain B minimum power requirements<br />

Sum of <strong>JS20</strong> power units Minimum power module required<br />

Less than 9.9 1200 W (labeled 7.5A)<br />

9.9 up to less than 11.5 1400 W (labeled 9A)<br />

11.5 up to less than 13.4 1800 W (labeled 12A)<br />

13.4 or greater 2000 W (labeled 13.5A)<br />

Each <strong>BladeCenter</strong> <strong>JS20</strong> (1.6 GHz/800 MHz 8842-21X) requires 1.5 power units<br />

for two installed microprocessors. For example, if you fully populate all six bays,<br />

the total power units is 9.0 (6 bays times 1.5). According to Table 2-1, you need a<br />

minimum of the 1800 W power module.<br />

For other configurations, refer to the latest version of <strong>IBM</strong> ^ <strong>BladeCenter</strong><br />

Power Module Upgrade Guidelines Technical Update, found at:<br />

http://www.ibm.com/pc/support/site.wss/document.do?lndocid=MIGR-53353<br />

Also, you can use the following steps to obtain the <strong>BladeCenter</strong> Planning and<br />

Installation Guide:<br />

1. Go to:<br />

http://www.ibm.com/pc/support/<br />

2. In the Browse by product section, click Servers, and then select the brand<br />

Servers and the family <strong>BladeCenter</strong>.<br />

3. Click Continue.<br />

4. From the View by document type menu, click Online Publications.<br />

2.1.3 Management Module<br />

Each <strong>BladeCenter</strong> chassis comes standard with a Management Module, as<br />

shown in Figure 2-4. You can also install a redundant Management Module by<br />

ordering the <strong>IBM</strong> KVM/Redundant Management Module Option (P/N 48P7055).<br />

When you have two Management Modules, only one is active. <strong>The</strong> other module<br />

is in standby mode until it is switched to become the active Management Module.<br />

Chapter 2. Hardware components 13


Figure 2-4 KVM Management Module<br />

<strong>The</strong> Management Module provides the following three major functions:<br />

► Console switching and remote graphical console for blade servers that use a<br />

keyboard, video monitor, and mouse (KVM) for their console<br />

► Serial over local area network (LAN) (SoL) remote text consoles for blade<br />

servers that support SoL consoles (includes the <strong>BladeCenter</strong> <strong>JS20</strong>)<br />

► Service processor for the entire <strong>BladeCenter</strong><br />

Important: <strong>The</strong> console switching and remote graphical console functionality<br />

of the Management Module is not used by the <strong>BladeCenter</strong> <strong>JS20</strong> because it<br />

does not support KVM consoles. You can still use the console switching and<br />

remote graphical console functions if you have other blade servers installed in<br />

the same <strong>BladeCenter</strong> chassis that support KVM consoles.<br />

You use the service processor functionality of the Management Module to<br />

perform many different tasks to:<br />

► Initially configure <strong>BladeCenter</strong><br />

► Define and manage the users who can remotely access the Management<br />

Module<br />

► Monitor the <strong>BladeCenter</strong> and reporting abnormal conditions to system<br />

managers either via e-mail or Simple Network Management Protocol (SNMP)<br />

events sent to management systems<br />

► Remotely power on and off individual blade servers and I/O modules<br />

14 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


► Remotely switch the CD-ROM, diskette, and USB interfaces in the media bay<br />

of the <strong>BladeCenter</strong> chassis to a particular blade server<br />

► Maintain the firmware embedded in various <strong>BladeCenter</strong> components<br />

<strong>The</strong> Management Module is connected to each blade server and I/O module via<br />

management buses embedded in the <strong>BladeCenter</strong> chassis midplanes. <strong>The</strong><br />

Management Module uses these connections to provide service processor<br />

functionality in conjunction with the Blade Systems Management Processor<br />

(BSMP) on each blade server.<br />

When using the SoL remote text console function, the Management Module also<br />

uses a virtual LAN (VLAN) provided by the LAN switch module installed in I/O<br />

module bay 1 to communicate with each blade server. This VLAN is not<br />

accessible externally.<br />

Each Management Module is equipped with several connectors, as illustrated in<br />

Figure 2-5.<br />

Figure 2-5 Management Module connectors<br />

Chapter 2. Hardware components 15


2.1.4 I/O modules<br />

You use the keyboard, video, and mouse connectors to attach a local KVM<br />

console that can be switched to any blade server that supports KVM consoles.<br />

You use the 10/100BaseT Ethernet connector to attach the Management Module<br />

to an external Ethernet network. You use this interface to access the remote<br />

graphical console, SoL consoles, and the service processor functionality of the<br />

Management Module.<br />

You can access the various functions of the Management Module via the<br />

Ethernet interface using several different methods:<br />

► Using a graphical interface through a Web browser (HTTP or HTTPS)<br />

► Using a command line interface through a Telnet or SSH client<br />

► Using system management software such as <strong>IBM</strong> Director or <strong>IBM</strong> Cluster<br />

Systems Management<br />

We demonstrate these methods in Chapter 5, “Hardware setup” on page 69, and<br />

in Chapter 8, “System management scenarios” on page 139.<br />

I/O modules enable connectivity between blade servers within the same chassis,<br />

and between blade servers and the outside world. You can install up to four I/O<br />

modules in a <strong>BladeCenter</strong> chassis, one in each I/O module bay.<br />

<strong>The</strong> available I/O modules can be used to provide connectivity to three different<br />

types of network:<br />

► Local area networks<br />

► Storage area networks (SANs)<br />

► High-bandwidth low-latency networks used in clustering (Myrinet)<br />

Most I/O modules support connectivity to only one type of network. <strong>The</strong><br />

exception is the pass-through module, which can be used with all three types of<br />

networks.<br />

Each I/O module is connected to every blade server via the <strong>BladeCenter</strong> chassis<br />

midplanes. Each single width blade server can have up to four interfaces, one to<br />

each I/O module bay. Double-width blade servers can have up to eight interfaces,<br />

two to each I/O module bay.<br />

Each blade server must have a compatible interface to communicate with an I/O<br />

module. Every blade server that is currently available has standard Gigabit<br />

Ethernet interfaces that are connected to I/O module bays 1 and 2. <strong>The</strong>refore,<br />

you can use only a LAN Switch or Pass-Thru I/O Module in these I/O module<br />

bays.<br />

16 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Restriction: You must install one of the LAN switch modules in I/O module<br />

bay 1 when using the SoL remote text console function. This function depends<br />

on a private VLAN provided by the LAN switch to function correctly.<br />

Every single-width blade server currently available supports the installation of an<br />

optional expansion card that connects to I/O module bay 3 and 4. Double-width<br />

blade servers support two optional expansion cards. If you install an expansion<br />

card, it must be compatible with any I/O modules installed in I/O module bays<br />

3 and 4.<br />

You can order the following LAN Switch I/O Modules:<br />

► P/N 13N0568: 4 port Gigabit Ethernet Switch Module<br />

► P/N 13N2281: CISCO Systems Intelligent Gigabit Ethernet Switch Module<br />

► P/N 73P9057: Nortel Networks Layer 2-7 Gigabit Ethernet Switch Module<br />

You can order the following SAN switch I/O modules:<br />

► P/N 48P7062: 2-Port Fibre Channel Switch Module<br />

► P/N 26K5601: Brocade Entry SAN Switch Module<br />

► P/N 90P0165: Brocade Enterprise SAN Switch Module<br />

For the Pass-Thru I/O Module, you can order Optical Pass-thru Module (P/N<br />

02R9080).<br />

We describe each of these modules in more detail in the sections that follow.<br />

<strong>IBM</strong> 4-Port Gigabit Ethernet Switch Module<br />

<strong>The</strong> <strong>IBM</strong> 4-Port Gigabit Ethernet Switch Module is a layer 2 LAN switch that<br />

provides four external 10/100/1000BaseT interfaces for connecting to LANs that<br />

are external to the <strong>BladeCenter</strong> chassis. Internally the switch module is designed<br />

to connect to a single Gigabit Ethernet interface on each installed single-width<br />

blade server, or two such interfaces on each installed double-width blade server.<br />

<strong>The</strong> <strong>IBM</strong> 4-Port Gigabit Ethernet Switch Module is also connected via a Fast<br />

Ethernet interface to each installed Management Module. This is used to<br />

remotely access the switch module via the Management Module’s 10/100BaseT<br />

Ethernet interface. You can use this capability to support initial configuration and<br />

management of the switch module.<br />

Major features of the 4-Port Gigabit Ethernet Switch Module include:<br />

► VLAN support based on either port or 802.1Q tagging<br />

► Link aggregation of multiple external interfaces to provide increased<br />

bandwidth to external networks<br />

Chapter 2. Hardware components 17


Note: <strong>The</strong> <strong>IBM</strong> 4-Port Gb Ethernet Switch Module for <strong>BladeCenter</strong> may be<br />

installed in your <strong>BladeCenter</strong> unit and you must replace the CD-ROM media<br />

tray in your <strong>BladeCenter</strong> unit. In this case, you must also update the<br />

microcode in the Ethernet switch module.<br />

CISCO Systems Intelligent Gigabit Ethernet Switch Module<br />

<strong>The</strong> CISCO Systems Intelligent Gigabit Ethernet Switch Module (IGESM) is a<br />

layer 2-3 LAN switch that provides four external 10/100/1000BaseT interfaces for<br />

connecting to LANs external to the <strong>BladeCenter</strong> chassis. Internally the switch<br />

module is designed to connect to a single Gigabit Ethernet interface on each<br />

installed single width blade server, or two such interfaces on each installed<br />

double-width blade server.<br />

<strong>The</strong> IGESM is also connected via a Fast Ethernet interface to each installed<br />

Management Module. This is used to remotely access the switch module via the<br />

Management Module’s 10/100BaseT Ethernet interface. You can use this<br />

capability to support initial configuration and management of the switch module.<br />

<strong>The</strong> IGESM uses CISCO IOS. Use the IGESM when you require the advanced<br />

functionality of CISCO IOS or maximum compatibility with external networks that<br />

constructed using switches running CISCO IOS.<br />

Note: To use the SoL remote text console function with the IGESM, you must<br />

manually create the required VLAN using the procedures in the CISCO<br />

Systems Intelligent Gigabit Ethernet Switch Module - Installation Guide.<br />

Nortel Networks Layer 2-7 Gigabit Ethernet Switch Module<br />

<strong>The</strong> Nortel Networks Layer 2-7 Gigabit Ethernet Switch Module is a layer 2-7<br />

LAN switch that provides four external 10/100/1000BaseT interfaces for<br />

connecting to LANs external to the <strong>BladeCenter</strong> chassis. Internally the switch<br />

module is designed to connect to a single Gigabit Ethernet interface on each<br />

installed single width blade server, or two such interfaces on each installed<br />

double-width blade server.<br />

<strong>The</strong> Nortel Networks Layer 2-7 Gigabit Ethernet Switch Module is also<br />

connected via a Fast Ethernet interface to each installed Management Module.<br />

This is used to remotely access the switch module via the Management Module’s<br />

10/100BaseT Ethernet interface. You can use this capability to support initial<br />

configuration and management of the switch module.<br />

18 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


<strong>The</strong> Nortel Networks Layer 2-7 Gigabit Ethernet Switch Module provides several<br />

advanced networking features including:<br />

► Rich layer 2 and layer 3 switching capabilities<br />

► Virtual server load balancing (layer 4 switching)<br />

► Content aware load balancing (layer 7 switching)<br />

► High availability clustering of switch modules in the same or different chassis<br />

You can learn more about the Nortel Networks Layer 2-7 Gigabit Ethernet Switch<br />

Module in the <strong>IBM</strong> Redpaper <strong>IBM</strong> Eserver <strong>BladeCenter</strong> Layer 2-7 Network<br />

Switching, REDP-3755.<br />

2-Port Fibre Channel Switch Module<br />

<strong>The</strong> 2-Port Fibre Channel Switch Module is a Fibre Channel SAN switch that<br />

provides two external 2 Gbps Fibre Channel interfaces for connecting to SANs<br />

external to the <strong>BladeCenter</strong> chassis. Internally the switch module is designed to<br />

connect to a single 2 Gbps Fibre Channel interface on each installed blade<br />

server that is equipped with a Fibre Channel expansion card.<br />

<strong>The</strong> 2-Port Fibre Channel Switch Module is also connected via a Fast Ethernet<br />

interface to each installed Management Module. This is used to remotely access<br />

the switch module via the Management Module’s 10/100BaseT Ethernet<br />

interface. You can use this capability to support initial configuration and<br />

management of the switch module.<br />

<strong>The</strong> 2-Port Fibre Channel Switch Module can be installed only in I/O module<br />

bays 3 or 4. If you install a 2-Port Fibre Channel Switch Module, any blade<br />

servers that have expansion cards must use a Fibre Channel expansion card.<br />

Brocade Enterprise SAN Switch Module<br />

<strong>The</strong> Brocade Enterprise SAN Switch Module is a Fibre Channel SAN switch that<br />

provides two external 1 Gbps or 2 Gbps (autosense) Fibre Channel interfaces for<br />

connecting to SANs external to the <strong>BladeCenter</strong> chassis. Internally the switch<br />

module is designed to connect to a single 2 Gbps Fibre Channel interface on<br />

each installed blade server that is equipped with a Fibre Channel expansion<br />

card.<br />

<strong>The</strong> Brocade Enterprise SAN Switch Module is also connected via a Fast<br />

Ethernet interface to each installed Management Module. This is used to<br />

remotely access the switch module via the Management Module’s 10/100BaseT<br />

Ethernet interface. You can use this capability to support initial configuration and<br />

management of the switch module.<br />

Chapter 2. Hardware components 19


<strong>The</strong> Brocade Enterprise SAN Module can be installed only in I/O module bays<br />

3 or 4. If you install a 2-Port Fibre Channel Switch Module, any blade servers that<br />

have expansion cards must use a Fibre Channel expansion card.<br />

Use the Brocade Enterprise SAN Switch Module when you require maximum<br />

compatibility with external SAN fabrics constructed using Brocade-based<br />

switches. <strong>The</strong> Brocade Enterprise SAN Switch Module is functionality equivalent<br />

to the Brocade Silkworm 3900 SAN switch.<br />

Brocade Entry SAN Switch Module<br />

<strong>The</strong> Brocade Entry SAN Switch Module is almost identical to the Brocade<br />

Enterprise SAN Switch Module described earlier. However, it is limited to<br />

participating in SAN fabrics that contain at most two SAN switches.<br />

Optical Pass-thru Module<br />

<strong>The</strong> Optical Pass-thru Module provides the capability to access one of the<br />

network interfaces on each blade server via an externally accessible optical<br />

interface. You can use this optical interface to connect the blade servers to<br />

external switches that support compatible optical interfaces, including:<br />

► LAN switches<br />

► SAN switches<br />

► Myrinet switches<br />

Each Optical Pass-thru Module has four multi-port optical connectors. <strong>The</strong><br />

connectors are designed for special break-out cables that provide either four SC<br />

connectors (P/N 73P5992) or four LC connectors (P/N 73P6033) per multi-port<br />

optical connector. You can connect to a maximum of 14 blade server network<br />

interfaces using four of these cables (two of the connectors are unused on the<br />

fourth cable).<br />

<strong>The</strong> type of external switch that you connect to the Optical Pass-thru Module<br />

must be compatible with the network interface on the blade servers that are<br />

connected to the I/O module bay where the Optical Pass-thru Module is installed.<br />

If you install an Optical Pass-thru Module in I/O module bay 1 or 2, you must<br />

connect the optical interfaces to a Gigabit Ethernet switch with 1000Base-SX<br />

interfaces.<br />

Restriction: When a <strong>BladeCenter</strong> chassis contains a <strong>BladeCenter</strong> <strong>JS20</strong>,<br />

always use a LAN switch module in I/O module bay 1 to support the SoL<br />

remote text console function. This is because there is no other mechanism for<br />

providing a console on a <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

20 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


2.1.5 Blade servers<br />

If you install an Optical Pass-thru Module in I/O module bay 3 or 4, you must<br />

connect the optical interfaces to an external switch that is compatible with the<br />

type of expansion cards installed on blade servers.<br />

If you install a Myrinet expansion card on any blade server, you must install an<br />

Optical Pass-thru Module in I/O module bay 4 to enable the Myrinet interface on<br />

the expansion card to be connected to an external Myrinet switch. This is<br />

because currently no Myrinet switch modules are available for the <strong>BladeCenter</strong>.<br />

<strong>IBM</strong> offers a variety of blade servers that you can install in a <strong>BladeCenter</strong><br />

chassis.<br />

► <strong>IBM</strong> ^ <strong>BladeCenter</strong> <strong>JS20</strong> Type 8842 Blade Server<br />

► <strong>IBM</strong> ^ <strong>BladeCenter</strong> HS20 Type 8678 Blade Server<br />

► <strong>IBM</strong> ^ <strong>BladeCenter</strong> HS20 Type 8832 Blade Server<br />

► <strong>IBM</strong> ^ <strong>BladeCenter</strong> HS40 Type 8839 Blade Server<br />

Because the <strong>BladeCenter</strong> <strong>JS20</strong> is the subject of this book, we describe it in<br />

greater detail in 2.2, “<strong>BladeCenter</strong> <strong>JS20</strong>” on page 22.<br />

<strong>The</strong> following sections briefly introduce the other blade servers. You can install<br />

any combination of blade servers in the same <strong>BladeCenter</strong> chassis subject to<br />

available power supply module capacity.<br />

HS20 Blade Server<br />

<strong>The</strong> HS20 Blade Server is a single-width blade server that contains either one or<br />

two Intel® Xeon DP processors and up to 8 GB of memory. Up to 14 HS20 Blade<br />

Servers can be installed in each <strong>BladeCenter</strong> chassis.<br />

<strong>The</strong> original HS20 Type 8678 Blade Server used a 400 MHz front side bus (FSB)<br />

and supports a range of processor speeds from 2.0 GHz to 2.8 GHz. <strong>The</strong> HS20<br />

Type 8678 Blade Server is now withdrawn from marketing and has been replaced<br />

by the HS20 Type 8832 Blade Server.<br />

<strong>The</strong> HS20 Type 8832 Blade Server uses a 533 MHz FSB and supports a range<br />

of processor speeds from 2.4 Ghz to 3.2 GHz.<br />

HS40 Blade Server<br />

<strong>The</strong> HS40 Blade Server is a double-width blade server that contains from one to<br />

four Intel Xeon MP processors and up to 16 GB of memory. Up to seven HS40<br />

Blade Servers can be installed in each <strong>BladeCenter</strong> chassis.<br />

Chapter 2. Hardware components 21


2.2 <strong>BladeCenter</strong> <strong>JS20</strong><br />

<strong>The</strong> <strong>BladeCenter</strong> <strong>JS20</strong> is a single-width blade server that offers these features:<br />

► Two PowerPC 970 or PowerPC 970FX microprocessors<br />

► Up to 4 GB of RAM<br />

► Up to two integrated development environment (IDE) hard disk drives<br />

mounted on the blade server<br />

► Two integrated Gigabit Ethernet interfaces<br />

► Expansion option connectors for an expansion card option to provide either:<br />

– Two additional Gigabit Ethernet interfaces<br />

– Two 2 Gbps Fibre Channel interfaces<br />

– One Myrinet interface<br />

► Blade Systems Management Processor<br />

Figure 2-6 illustrates the physical layout of the <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

I/O expansion option<br />

I/O expansion option<br />

connector (J18)<br />

connector (J22)<br />

Four Memory DIMMS<br />

Secondary IDE (J2) Primary IDE (J1) Battery<br />

(shown with attached disk)<br />

Figure 2-6 <strong>BladeCenter</strong> <strong>JS20</strong> physical layout<br />

22 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong><br />

Microprocessor 0<br />

and heat sink (U87)<br />

Microprocessor 1<br />

and heat sink (U41)


<strong>The</strong> architecture of the <strong>BladeCenter</strong> <strong>JS20</strong> is illustrated by the diagram in<br />

Figure 2-7.<br />

PowerPC<br />

970<br />

or<br />

970FX<br />

6.4 GBps<br />

or<br />

8.8 GBps<br />

PowerPC<br />

970<br />

or<br />

970FX<br />

Memory Controller and Host I/O Bridge<br />

(Northbridge)<br />

HyperTransport PCI-X Tunnel<br />

HyperTransport I/O Hub<br />

IDE Controller<br />

IDE<br />

Disk<br />

IDE<br />

Disk<br />

1.6 GBps<br />

800 MBps<br />

6.4 GBps<br />

or<br />

8.8 GBps<br />

5.3 GBps<br />

Real Time<br />

Clock<br />

1.06 GBps<br />

1.06 GBps<br />

USB<br />

Controllers<br />

Super I/O<br />

Figure 2-7 <strong>BladeCenter</strong> <strong>JS20</strong> block diagram<br />

UART<br />

BSMP<br />

PC2700 ECC<br />

Memory Modules<br />

Gigabit<br />

Ethernet<br />

Controller<br />

Expansion Card<br />

Boot<br />

FLASH<br />

NVRAM<br />

Chassis<br />

Midplanes<br />

<strong>The</strong> <strong>BladeCenter</strong> <strong>JS20</strong> includes both base features and a range of optional<br />

features. We describe each of these in the sections that follow.<br />

Chapter 2. Hardware components 23


2.2.1 Base features<br />

<strong>The</strong>re are two models of the <strong>BladeCenter</strong> <strong>JS20</strong>. Both models have the same<br />

base features with the exception of the microprocessors used in the processor<br />

subsystem. <strong>The</strong> <strong>JS20</strong> Type 8842-21X Blade Server uses the PowerPC 970<br />

microprocessor. <strong>The</strong> <strong>JS20</strong> Type 8842-41X Blade Server uses the PowerPC<br />

970FX microprocessor.<br />

Processor subsystem<br />

<strong>The</strong> processor subsystem of the <strong>BladeCenter</strong> <strong>JS20</strong> is comprised of either two<br />

PowerPC 970 or two PowerPC 970FX microprocessors. <strong>The</strong>y operate at internal<br />

clock frequencies of either 1.6 GHz or 2.2 GHz, respectively.<br />

Each processor includes a 32 KB level 1 data cache, a 64 KB level 1 instruction<br />

cache, and a 512 KB ECC level 2 cache.<br />

Each processor is also interconnected with the rest of the system via a dedicated<br />

interface to the memory controller and host I/O bridge (Northbridge). <strong>The</strong><br />

interface to each processor supports a maximum aggregate bandwidth of<br />

6.4 GBps (for processors operating at 1.6 GHz) or 8.8 GBps (for processors<br />

operating at 2.2 GHz).<br />

You can learn more about the features and architecture of the processors in 2.3,<br />

“PowerPC 970 and PowerPC 970FX Microprocessors” on page 29.<br />

Memory subsystem<br />

<strong>The</strong> memory subsystem is comprised of the memory controller and up to four<br />

DDR1 333 MHz (PC2700) ECC memory modules. <strong>The</strong> memory controller uses<br />

two-way interleaving to increase the available peak memory bandwidth to 5.3<br />

GBps (333 MHz x 16 bytes).<br />

<strong>The</strong> <strong>BladeCenter</strong> <strong>JS20</strong> includes two 256 MB memory modules that provide a<br />

total memory capacity of 512 MB. <strong>The</strong>se memory modules are installed in slots 3<br />

and 4. <strong>The</strong> remaining memory module slots (1 and 2) are empty in the base<br />

<strong>BladeCenter</strong> <strong>JS20</strong>.<br />

You can expand the available memory by installing optional memory modules.<br />

<strong>The</strong> available memory modules are described in “Memory modules” on page 27.<br />

I/O subsystem<br />

<strong>The</strong> I/O subsystem is connected to the processor and memory subsystems via a<br />

hyper transport link that supports an aggregate bandwidth of 1.6 GBps.<br />

Downstream of this link is a range of I/O devices and expansion interfaces as<br />

explained in the following sections.<br />

24 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Gigabit Ethernet controller<br />

Each <strong>BladeCenter</strong> <strong>JS20</strong> includes a Broadcom BCM5704s Gigabit Ethernet<br />

controller, which provides the primary mechanism for connecting the<br />

<strong>BladeCenter</strong> <strong>JS20</strong> to other systems. This controller provides two independent<br />

Gigabit Ethernet interfaces.<br />

Each Gigabit Ethernet interface is connected to a separate <strong>BladeCenter</strong> I/O<br />

module bay via the <strong>BladeCenter</strong> midplanes. <strong>The</strong> first interface is connected to I/O<br />

module bay 1, and the second interface is connected to I/O module bay 2.<br />

To use a Gigabit Ethernet interface, install a compatible I/O module in the<br />

corresponding I/O module bay. Install a LAN Switch I/O Module in I/O module<br />

bay 1 to support remote access to the <strong>BladeCenter</strong> <strong>JS20</strong> text console via SoL.<br />

I/O expansion option connectors<br />

Each <strong>BladeCenter</strong> <strong>JS20</strong> contains I/O expansion option connectors, which you<br />

can use to attach one of several different expansion cards. This capability<br />

enables you to expand the external I/O connectivity of the <strong>BladeCenter</strong> <strong>JS20</strong> to<br />

include Fibre Channel SANs, Myrinet high-bandwidth low-latency networks, and<br />

additional Gigabit Ethernet LAN interfaces.<br />

<strong>The</strong> expansion card installed in the I/O expansion option connectors can access<br />

I/O modules installed in <strong>BladeCenter</strong> I/O module slots 3 and 4 via the<br />

<strong>BladeCenter</strong> midplanes.<br />

Note: <strong>The</strong> main I/O expansion option connector provides a PCI-X electrical<br />

interface. However, a special form factor is required to enable installation<br />

within the confined dimensions of a blade server. <strong>The</strong>refore, standard PCI-X<br />

adapter cards cannot be used with the <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

You can learn more about these expansion cards in “Expansion cards” on<br />

page 28.<br />

IDE controller<br />

Each <strong>BladeCenter</strong> <strong>JS20</strong> includes a dual-channel IDE controller and two<br />

connectors for 2.5-inch IDE hard disk drives. You can use this capability to install<br />

one or two optional 2.5-inch IDE hard disk drives on the <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

Note: You must install at least one hard disk drive for use as a boot device<br />

unless you plan to install a Fibre Channel Expansion Card and configure the<br />

<strong>BladeCenter</strong> <strong>JS20</strong> to boot from a SAN. You can also boot from a network<br />

interface to support the network installation of an operating system.<br />

Chapter 2. Hardware components 25


You can learn more about the available hard disk drives in “Expansion cards” on<br />

page 28.<br />

USB controllers<br />

Each <strong>BladeCenter</strong> <strong>JS20</strong> includes two USB 1.1 host controllers. <strong>The</strong>y are<br />

connected via the <strong>BladeCenter</strong> midplanes to the USB peripherals located in the<br />

media bay of the <strong>BladeCenter</strong> chassis.<br />

You use these controllers when accessing either the CD-ROM or diskette drive<br />

that are located in the media bay of the <strong>BladeCenter</strong> chassis. You can also<br />

access other USB devices that may be attached to the USB connector located in<br />

the media bay of the <strong>BladeCenter</strong> chassis.<br />

Note: Verify that the USB device is supported by the operating system<br />

installed on the blade server before you attach it to the USB connector located<br />

in the media bay of the <strong>BladeCenter</strong> chassis.<br />

You can assign only the USB peripherals to one blade server at a time. Either<br />

use a button on the front of the blade server or one of the remote interfaces to the<br />

Management Module to switch the USB peripherals to a specific blade server.<br />

Miscellaneous I/O<br />

Each <strong>BladeCenter</strong> <strong>JS20</strong> incorporates various miscellaneous I/O devices that are<br />

necessary to implement a functional server. <strong>The</strong>se are illustrated in the<br />

<strong>BladeCenter</strong> <strong>JS20</strong> diagram in Figure 2-7 on page 23 and include:<br />

► A real-time clock (RTC) for maintaining the date and time<br />

► Non-volatile memory (NVRAM) for storing configuration settings<br />

► Boot FLASH memory for storing the firmware (BIOS) for the blade server<br />

► A serial port (UART)<br />

<strong>The</strong> serial port is connected within the blade server to the BSMP. <strong>The</strong> BSMP<br />

participates in the implementation of the SoL remote text console function that<br />

provides remote connectivity to the serial port via the <strong>BladeCenter</strong> Management<br />

Module.<br />

Blade Systems Management Processor<br />

Each <strong>BladeCenter</strong> <strong>JS20</strong> has a BSMP. <strong>The</strong> BSMP provides the interface that<br />

enables the <strong>BladeCenter</strong> Management Module to manage the <strong>BladeCenter</strong> <strong>JS20</strong><br />

via the <strong>BladeCenter</strong> chassis midplanes.<br />

<strong>The</strong> BSMP is responsible for powering up the <strong>BladeCenter</strong> <strong>JS20</strong>, configuring the<br />

various hardware subsystems, monitoring temperatures, and driving the status<br />

indicators on the blade server front panel.<br />

26 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


<strong>The</strong> BSMP also participates in the implementation of the SoL remote text<br />

console function on the <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

2.2.2 Optional features<br />

You can install several optional features on a <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

Memory modules<br />

Each <strong>BladeCenter</strong> <strong>JS20</strong> is equipped with 512 MB of memory. This is provided<br />

using a pair of 256 MB memory modules.<br />

You can also install the following optional memory modules on a <strong>BladeCenter</strong><br />

<strong>JS20</strong>:<br />

► P/N 73P2275: <strong>IBM</strong> 512 MB 333 MHz PC2700 CL2.5 ECC DDR RDIMM<br />

► P/N 73P2276: <strong>IBM</strong> 1 GB 333 MHz PC2700 CL2.5 ECC DDR RDIMM<br />

You must install the optional memory module features in identical pairs. This is<br />

necessary because the memory controller on the <strong>BladeCenter</strong> <strong>JS20</strong> uses<br />

two-way interleaving.<br />

Two spare memory module slots (1 and 2) are available on the <strong>BladeCenter</strong><br />

<strong>JS20</strong>. You can expand the available memory to either 1.5 GB or 2.5 GB, using<br />

either a pair of 512 MB or 1 GB memory modules respectively.<br />

You can also remove the standard pair of 256 MB memory modules and replace<br />

them by a pair of either 512 MB or 1 GB memory modules if further memory<br />

expansion is required. To obtain the maximum supported memory configuration<br />

of 4 GB, remove both of the base 256 MB memory modules and install four 1 GB<br />

memory module features.<br />

Hard disk drives<br />

<strong>The</strong> base <strong>BladeCenter</strong> <strong>JS20</strong> is not equipped with any hard disk drives. You can<br />

install one or two of the 2.5-inch IDE hard disk drive option 40 GB 5400 rpm<br />

ATA-100 Hard Disk Drive (P/N 48P7063).<br />

When you install two hard disk drives, each hard disk drive is connected to its<br />

own dedicated IDE bus.<br />

Restriction: If you install an expansion card on a <strong>BladeCenter</strong> <strong>JS20</strong>, only one<br />

hard disk drive can be installed on the same blade server.<br />

You do not have to install a hard disk drive if you have installed a Fibre Channel<br />

Expansion Card and have configured the <strong>BladeCenter</strong> <strong>JS20</strong> to boot from a SAN.<br />

Chapter 2. Hardware components 27


Expansion cards<br />

You can install any one of the following expansion cards in the I/O expansion<br />

option connectors of the <strong>BladeCenter</strong> <strong>JS20</strong>:<br />

► P/N 73P9030: Gigabit Ethernet Expansion Card<br />

► P/N 13N2203: Fibre Channel Expansion Card<br />

► P/N 73P6000: Myrinet Cluster Expansion Card<br />

We describe each of these expansion cards in the sections that follow.<br />

Fibre Channel Expansion Card<br />

You can use the Fibre Channel Expansion Card to connect the <strong>BladeCenter</strong><br />

<strong>JS20</strong> to SANs. <strong>The</strong> Fibre Channel Expansion Card provides two 2 Gbps Fibre<br />

Channel interfaces that are connected to I/O module bays 3 and 4. You must<br />

install either a SAN switch I/O module or the Optical Pass-Thru I/O Module in one<br />

or both of these I/O module bays to connect the blade server Fibre Channel<br />

interfaces to an external SAN.<br />

Note: <strong>The</strong>re have been two different versions of the Fibre Channel Expansion<br />

Card. Both versions are functionally equivalent. <strong>The</strong> earlier version (P/N<br />

48P7061) is now withdrawn from marketing. However, you may still find it in<br />

existing <strong>BladeCenter</strong> installations.<br />

Gigabit Ethernet Expansion Card<br />

You can use the Gigabit Ethernet Expansion Card to increase the number of<br />

Gigabit Ethernet network interfaces on the <strong>BladeCenter</strong> <strong>JS20</strong> from two to four.<br />

<strong>The</strong> Gigabit Ethernet Expansion Card provides two Gigabit Ethernet interfaces<br />

that are connected to I/O module bays 3 and 4. You must install either a LAN<br />

Switch I/O Module or the Optical Pass-Thru I/O Module in one or both of these<br />

I/O module bays to connect the Gigabit Ethernet interfaces on the expansion<br />

card to an external LAN.<br />

Myrinet Cluster Expansion Card<br />

You can use the Myrinet Cluster Expansion Card to connect the <strong>BladeCenter</strong><br />

<strong>JS20</strong> to a Myrinet network. Myrinet was developed by Myricom Inc., and<br />

conforms to an ANSI standard (ANSI/VITA 26-1998).<br />

Myrinet networks are typically used to support certain types of high performance<br />

computing applications that distribute computation across a cluster of multiple<br />

servers. Such applications are said to exploit distributed memory parallelism.<br />

Distributed memory parallel applications often require the servers in the cluster<br />

to exchange messages with each other through some form of interconnection<br />

28 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


network. <strong>The</strong> performance of such applications often depends on how quickly<br />

these messages can be exchanged. <strong>The</strong> highest performance is usually<br />

obtained by using a high-bandwidth, low-latency interconnection network.<br />

Myrinet provides a high-bandwidth, low-latency interconnection network that can<br />

be used to optimize the performance of distributed memory parallel applications.<br />

Consider using Myrinet if your application requires higher performance than is<br />

achievable using Gigabit Ethernet technology.<br />

You can learn more about Myrinet technology at the Myrinet Web site at:<br />

http://www.myri.com<br />

Note: <strong>The</strong> Myrinet Cluster Expansion Card is a repackaging of the Myricom<br />

M3F-PCIXD-2 PCI-X adapter to fit the special form factor required for blade<br />

server expansion cards.<br />

When you install a Myrinet Cluster Expansion Card on a blade server, you must<br />

install an Optical Pass-thru Module in I/O module bay 4 of the <strong>BladeCenter</strong><br />

chassis. You must also connect the external interfaces of the Optical Pass-thru<br />

Module to an external Myrinet switch via appropriate optical fiber cables.<br />

We elaborate on the planning implications associated with using the Myrinet<br />

Cluster Expansion Card in 4.1.2, “High-performance, low-latency network<br />

requirements” on page 58.<br />

You can order Myrinet switches from <strong>IBM</strong> as a component of the <strong>IBM</strong> ^<br />

Cluster 1350 product. You can learn more about Cluster 1350 at:<br />

http://www.ibm.com/eserver/cluster<br />

2.3 PowerPC 970 and PowerPC 970FX Microprocessors<br />

This section provides more detailed information about the <strong>IBM</strong> PowerPC 970<br />

Microprocessor and <strong>IBM</strong> PowerPC 970FX RISC Microprocessor that are used for<br />

the main processors on the <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

This section reviews the POWER and PowerPC Architecture upon which the<br />

PowerPC 970 and PowerPC 970FX microprocessors are based. It also<br />

introduces the Vector/SIMD Multimedia eXtension (VMX) to the PowerPC<br />

Architecture. And it describes the major features of the PowerPC 970 and<br />

PowerPC 970FX microprocessors.<br />

Chapter 2. Hardware components 29


2.3.1 Review of POWER and PowerPC Architecture<br />

In 1990, <strong>IBM</strong> announced a new processor architecture called performance<br />

optimization with enhanced RISC (POWER). This architecture was first<br />

implemented in the <strong>IBM</strong> RS/6000®.<br />

<strong>The</strong> POWER architecture was based on earlier <strong>IBM</strong> Research efforts to develop<br />

reduced instruction set computer (RISC) technology. <strong>The</strong>se efforts began in the<br />

1970s and continued throughout the 1980s.<br />

RISC and CISC concepts<br />

In the 1960s and 1970s, most computers were of a design now known as a<br />

complex instruction set computer (CISC). This type of design featured a large<br />

and highly functional instruction set. Each instruction in a CISC processor design<br />

typically needs several processor clock cycles to execute. An example of a CISC<br />

design is the <strong>IBM</strong> System/360 and its successors.<br />

<strong>The</strong> need for complex and highly functional instructions primarily existed<br />

because computers were equipped with small quantities of slow memory.<br />

Complex instructions resulted in fewer instructions per program, so less memory<br />

was needed for program storage. Complex instructions were also easier for<br />

human programmers to use when developing applications in assembly language<br />

as single instructions could be rich in functionality.<br />

However, studies showed that only a small percentage of CISC instructions were<br />

commonly used by programs. Also, the benefit of the CISC approach diminished<br />

over time as memory became larger and faster, and high-level languages<br />

replaced assembly language.<br />

<strong>The</strong> RISC concept was originally developed by <strong>IBM</strong> Fellow John Cocke in the<br />

early 1970s and includes the following characteristics:<br />

► A simple architecture with an optimized set of machine instructions<br />

<strong>The</strong> instructions in a RISC architecture are typically simple in function and<br />

regular in size. This simplifies the design of the instruction decode hardware.<br />

<strong>The</strong> large number of infrequently used instructions typically found in CISC<br />

architectures are eliminated and replaced by sequences of simple<br />

instructions in RISC architectures.<br />

► A high instruction execution rate<br />

A common objective of early RISC processor designs was to execute an<br />

average of one instruction per processor clock cycle. This has subsequently<br />

been improved upon in superscalar RISC processor designs that can execute<br />

multiple instructions per processor clock cycle.<br />

30 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


► Optimizing compiler availability<br />

RISC architectures assume the availability of good optimizing compilers, and<br />

that most users will be developing applications in high-level languages. This<br />

has simplified the hardware design by moving some of the complexity that<br />

was previously in the hardware into the compiler. However, the simplified<br />

instruction set of RISC architectures often makes it easier for compilers to<br />

choose optimal instruction sequences.<br />

► Load and store architecture<br />

RISC architectures typically incorporate large numbers of processor registers,<br />

and computational instructions manipulate values only in these registers.<br />

Separate instructions are provided for loading data from memory and storing<br />

data to memory. Compilers can take advantage of this clean separation and<br />

schedule instructions in a way that keeps the processor busy despite the long<br />

latency associated with memory access.<br />

Today the difference between CISC and RISC is less clear. Many CISC<br />

architectures have adopted many RISC concepts. Meanwhile, the instruction<br />

sets of many RISC architectures have grown to include hundreds of instructions.<br />

<strong>The</strong> original POWER and POWER2 processors<br />

<strong>The</strong> first implementation of POWER architecture released by <strong>IBM</strong> was the<br />

POWER processor as used in the first <strong>IBM</strong> RS/6000 servers and workstations.<br />

This processor was a multiple-chip implementation of the POWER architecture<br />

comprised of either seven or nine separate chips. It featured an 8 KB instruction<br />

cache and either a 32 KB or 64 KB data cache. Over the life of the POWER<br />

processor, it was shipped with clock speeds ranging from 20 MHz to 62.5 MHz.<br />

<strong>The</strong> original POWER processor introduced the concept of a superscalar RISC<br />

processor that could perform multiple operations in a single clock cycle by<br />

exploiting multiple execution units that operated concurrently. At the time, this<br />

represented a significant improvement over contemporary processor<br />

implementations available from other manufacturers.<br />

<strong>The</strong> floating-point performance of the POWER processor was further enhanced<br />

by the presence of the fused multiply-add (FMA) instruction. It enabled two<br />

floating-point operations to be executed by the floating-point execution unit each<br />

processor clock cycle. <strong>The</strong> peak floating-point performance of the POWER<br />

processor was therefore twice the processor clock rate. For example, a POWER<br />

processor operating at 25 MHz had a peak floating-point performance of 50<br />

MFLOPS.<br />

In subsequent years, <strong>IBM</strong> introduced a single chip implementation of the<br />

POWER processor called the RISC Single Chip (RSC), which was used in the<br />

<strong>IBM</strong> RS/6000® Model 220 and Model 230.<br />

Chapter 2. Hardware components 31


In 1993, <strong>IBM</strong> announced the POWER2 processor, which expanded upon the<br />

POWER processor superscalar concept by adding additional execution units to<br />

increase the number of operations that could be performed concurrently in a<br />

single clock cycle. This processor was a multiple chip implementation of a<br />

super-set of the original POWER architecture.<br />

<strong>The</strong> POWER2 processor had two floating-point execution units, each able to<br />

execute a FMA instruction every clock cycle. <strong>The</strong>refore, the peak floating-point<br />

performance of the POWER2 processor was four times the processor clock rate.<br />

For example, a POWER2 processor operating at 55 MHz had a peak<br />

floating-point performance of 220 MFLOPS.<br />

<strong>The</strong> memory subsystem of the POWER2 was also enhanced in order to keep up<br />

with the increased level of processor performance. This was achieved using<br />

larger caches and wider paths to memory.<br />

As a super-set of the POWER architecture, the POWER2 supported several<br />

instructions that are not present on any other POWER or PowerPC Architecture<br />

processors.<br />

<strong>IBM</strong> subsequently delivered a single chip implementation of the POWER2 called<br />

the POWER2 Super Chip (P2SC), which was offered at processor clock speeds<br />

as high as 160 MHz, and delivered peak floating-point performance of 640<br />

MFLOPS.<br />

PowerPC Architecture<br />

In 1991, <strong>IBM</strong> teamed with Apple Computer and Motorola to define the PowerPC<br />

Architecture. <strong>The</strong> goals of this new architecture were to:<br />

► Permit a broad range of implementations, from low-cost controllers to<br />

high-performance processors<br />

► Be sufficiently simple to permit the design of processors that have a very<br />

short cycle time<br />

► Minimize the effects that hinder the design of aggressive superscalar<br />

implementations.<br />

► Include multiprocessor features.<br />

► Define a 64-bit architecture that is a super-set of the 32-bit POWER<br />

architecture, providing application binary compatibility for 32-bit applications<br />

While based on the POWER architecture, the PowerPC Architecture<br />

incorporated several modifications to enable it to be more widely applied in a<br />

variety of application scenarios. This vision has been subsequently realized with<br />

processors implementing PowerPC Architecture now having been installed in<br />

32 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


desktop, server, and embedded systems across commercial, consumer,<br />

industrial, and scientific settings.<br />

To achieve the design goals for the PowerPC Architecture, some features of the<br />

original POWER architecture were removed. <strong>The</strong>se were mostly features that<br />

were infrequently used. None of the POWER2 extensions to the original POWER<br />

architecture were included in the PowerPC Architecture.<br />

<strong>The</strong> PowerPC Architecture defines both 32-bit and 64-bit modes of operation.<br />

<strong>The</strong> primary differences in these two modes of operation are in the effective<br />

length of addresses used by the processor, and the availability of extra<br />

capabilities to manipulate double word (64-bit) fixed-point operands in 64-bit<br />

mode. Floating-point capabilities are the same in both 32-bit and 64-bit modes.<br />

<strong>The</strong> 32-bit PowerPC Architecture implementations only support the 32-bit mode<br />

of operation. <strong>The</strong> 64-bit PowerPC Architecture implementations support both the<br />

32-bit and 64-bit modes of operation. This enables the 64-bit PowerPC<br />

Architecture implementations to support the full-speed execution of existing<br />

32-bit applications, alongside 64-bit applications, in the same operating<br />

environment.<br />

<strong>The</strong> PowerPC Architecture is documented in a series of books. You can<br />

download the most current version of the PowerPC Architecture books from the<br />

<strong>IBM</strong> developerWorks® Web site at:<br />

http://www.ibm.com/developerworks/eserver/articles/archguide.html<br />

<strong>The</strong> architecture is divided into three parts, each one documented in a separate<br />

book:<br />

► Book I: PowerPC User Instruction Set Architecture<br />

► Book II: PowerPC Virtual Environment Architecture<br />

► Book III: PowerPC Operating Environment Architecture<br />

Book IV is also available for each specific PowerPC processor implementation. It<br />

describes extensions to the architecture that are specific to a particular<br />

implementation.<br />

We briefly review the PowerPC User Instruction Set Architecture, since this is the<br />

part of the architecture that is of most interest to application programmers.<br />

PowerPC User Instruction Set Architecture<br />

<strong>The</strong> logical components of a PowerPC processor from the perspective of a<br />

application program are:<br />

► Fixed-point processor (FXU)<br />

► Floating-point processor (FPU)<br />

► Branch processor (BPU)<br />

Chapter 2. Hardware components 33


<strong>The</strong>se components all interact with storage as illustrated in Figure 2-8.<br />

Instructions<br />

from Storage<br />

Fixed-Point<br />

Instructions<br />

Fixed-Point<br />

Processing<br />

(FXU)<br />

Figure 2-8 PowerPC logical processing model<br />

A PowerPC Architecture groups the instructions that an application program uses<br />

on a PowerPC processor along the same lines into:<br />

► Fixed-point instructions<br />

► Floating-point instructions<br />

► Branch instructions<br />

All instructions in PowerPC Architecture are four bytes (one word) in size and are<br />

aligned on word boundaries. This simplifies the decoding of instructions and is<br />

characteristic of a RISC architecture.<br />

<strong>The</strong> fixed-point instructions operate on a set of 32 general purpose registers<br />

(GPRs), that are each 8 bytes (64 bits) in size on 64-bit PowerPC Architecture<br />

processors. <strong>The</strong>re is also a fixed-point exception register (XER). In 32-bit mode,<br />

only the lower 32 bits of the GPRs are considered significant, and the double<br />

word load and store instructions are not available.<br />

<strong>The</strong> floating-point instructions operate on a set of 32 floating-point registers<br />

(FPRs), that are each 8 bytes (64 bits) in size. <strong>The</strong>re is also a floating-point<br />

status and control register (FPSCR). <strong>The</strong> floating point capabilities are the same<br />

in both 32-bit and 64-bit mode. <strong>The</strong> large number of GPRs and FPRs is also<br />

typical of a RISC architecture.<br />

34 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong><br />

Branch<br />

Processing<br />

(BPU)<br />

Data to/from<br />

Storage<br />

Storage<br />

Floating-Point<br />

Instructions<br />

Floating-<br />

Point<br />

Processing<br />

(FPU)


<strong>The</strong> architecture also features a condition register that is set by the result of fixed<br />

or floating point computational instructions and that can be used to control<br />

branch processing. Branch instructions also make use of a count and a link<br />

register.<br />

Figure 2-9 illustrates the complete set of architected PowerPC registers used by<br />

application programs.<br />

CR<br />

0 31<br />

LR<br />

0 63<br />

CTR<br />

0 63<br />

GPR0<br />

GPR1<br />

...<br />

GPR31<br />

0 63<br />

XER<br />

0 63<br />

FPR0<br />

FPR1<br />

...<br />

FPR31<br />

0 63<br />

FPSCR<br />

0 31<br />

Condition Register<br />

Link Register<br />

Count Register<br />

General Purpose Registers<br />

Figure 2-9 PowerPC registers used by application programs<br />

Fixed-Point Exception Register<br />

Floating-Point Registers<br />

Floating-Point Status<br />

and Control Register<br />

In the PowerPC Architecture, fixed and floating-point computational instructions<br />

can operate on operands stored in registers. Data must be loaded from memory<br />

into a register using a fixed or floating-point load instruction before computations<br />

can be performed. Similarly, data must be stored from a register into memory<br />

using an explicit fixed or floating-point store instruction once computation is<br />

completed. This behavior is also characteristic of a RISC architecture.<br />

Chapter 2. Hardware components 35


<strong>IBM</strong> processors implementing PowerPC Architecture<br />

Since <strong>IBM</strong> teamed with Motorola and Apple Computer to define PowerPC<br />

Architecture, <strong>IBM</strong> has developed many different processors that implement this<br />

architecture. <strong>The</strong>se processors fall into two main categories:<br />

► Processors that are used as the main processor within <strong>IBM</strong> workstations and<br />

servers<br />

► Processors that <strong>IBM</strong> produces for embedding in a wide variety of computing<br />

and non-computing devices<br />

We focus our review of <strong>IBM</strong> processors implementing PowerPC Architecture on<br />

those processors that are used as the main processors within <strong>IBM</strong> workstations<br />

and servers.<br />

32-bit PowerPC processors<br />

In the early 1990s, <strong>IBM</strong> jointly designed with Motorola a series of 32-bit PowerPC<br />

processors with the designation 60X that were used in <strong>IBM</strong> RS/6000<br />

workstations and servers. <strong>The</strong>y were also used in computer systems from other<br />

computer manufacturers such as Apple Computer.<br />

<strong>The</strong> first of these was the PowerPC 601®. It debuted in the <strong>IBM</strong> RS/6000 Model<br />

250 in 1993. <strong>The</strong> PowerPC 601 was an evolution of the RISC Single Chip<br />

implementation of the POWER processor and did not implement the full<br />

PowerPC Architecture. It represented a transition between the earlier POWER<br />

architecture and the PowerPC Architecture.<br />

<strong>The</strong> PowerPC 601 was subsequently followed by the PowerPC 603 and<br />

PowerPC 604. <strong>The</strong>se represented full implementations of the 32-bit PowerPC<br />

Architecture. Enhanced versions of these processors known as the PowerPC<br />

603e and PowerPC 604e were subsequently introduced.<br />

<strong>The</strong> PowerPC 603 was an entry-level processor. <strong>The</strong> PowerPC 604 was a<br />

high-end processor and introduced support for symmetrical multi-processor<br />

(SMP) configurations.<br />

RS64 processors<br />

In 1997, <strong>IBM</strong> introduced its first servers that incorporated processors based on<br />

the 64-bit PowerPC Architecture. <strong>The</strong>se processors were called the RS64. <strong>The</strong><br />

RS64 was subsequently enhanced with the introduction of the RS64-II, RS64-III,<br />

and RS64-IV in later years. <strong>The</strong> RS64-IV operated at processor clock speeds as<br />

high as 750 MHz.<br />

<strong>The</strong> RS64 processor and its successors were optimized for use in commercial<br />

applications and de-emphasized the floating point capabilities of the earlier<br />

POWER2 processors. <strong>The</strong> RS64 processors were used in both <strong>IBM</strong> RS/6000<br />

36 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


(later <strong>IBM</strong> Eserver pSeries®) and <strong>IBM</strong> AS/400® (later <strong>IBM</strong> Eserver iSeries)<br />

servers, most frequently in SMP configurations, with some as large as 24<br />

processors.<br />

POWER3 and POWER3-II processors<br />

<strong>The</strong> POWER3 processor was first used in <strong>IBM</strong> RS/6000 servers and<br />

workstations introduced in 1998. <strong>The</strong> POWER3 processor continued the<br />

evolution of the excellent floating-point performance provided by the POWER2<br />

Super Chip (P2SC) processor, while implementing 64-bit PowerPC Architecture<br />

and supporting SMP configurations of up to 16 processors.<br />

Upon introduction, the POWER3 processor operated at 200 MHz, and delivered<br />

peak floating-point performance of 800 MFLOPS per processor. An improved<br />

version of the POWER3 processor, known as the POWER3-II, was subsequently<br />

introduced. <strong>The</strong> POWER3-II featured processor clock speeds as high as 450<br />

MHz, and delivered peak floating point performance of 1.8 GFLOPS per<br />

processor.<br />

You can learn more about the POWER3 processor in the <strong>IBM</strong> white paper<br />

POWER3: Next Generation 64-bit PowerPC Processor Design, which is available<br />

on the Web at:<br />

http://www.ibm.com/servers/eserver/pseries/hardware/whitepapers/power3wp.html<br />

POWER4 and POWER4+ processors<br />

<strong>The</strong> POWER4 processor combined the best attributes of both the RS64 and<br />

POWER3 processors. It delivers a 64-bit PowerPC Architecture processor that is<br />

capable of delivering high levels of performance in both commercial and scientific<br />

applications.<br />

<strong>The</strong> POWER4 processor is the first PowerPC Architecture processor to feature<br />

multiple processor cores on a single chip. Each POWER4 processor chip has<br />

two independent processor cores and a shared level-2 cache.<br />

<strong>The</strong> original POWER4 processor operated at processor clock speeds as high as<br />

1.3 GHz and has featured in SMP configurations as large as 32 processor cores.<br />

An enhanced version of the POWER4 processor, known as the POWER4+, was<br />

subsequently introduced and installed at processor clock speeds as high as<br />

1.9 GHz in the 32 processor core <strong>IBM</strong> Eserver pSeries Model 690.<br />

You can learn more about the POWER4 processor in the <strong>IBM</strong> white paper<br />

entitled POWER4 System Microarchitecture, which is available on the Web at:<br />

http://www.ibm.com/servers/eserver/pseries/hardware/whitepapers/power4.html<br />

Chapter 2. Hardware components 37


PowerPC 970 and PowerPC 970FX processors<br />

<strong>The</strong>se processors were first used by <strong>IBM</strong> in the <strong>BladeCenter</strong> <strong>JS20</strong> that is the<br />

subject of this book. We explore these processors in more detail in 2.3,<br />

“PowerPC 970 and PowerPC 970FX Microprocessors” on page 29.<br />

POWER5 processor<br />

<strong>IBM</strong> recently introduced the POWER5 processor in the <strong>IBM</strong> Eserver i5 and p5<br />

families of servers.<br />

<strong>The</strong> POWER5 processor represents an evolution of the POWER4 processor.<br />

Like the POWER4, the POWER5 processor implements two processor cores on<br />

a single chip with a shared level-2 cache.<br />

<strong>The</strong> major innovation in the POWER5 processor is the introduction of<br />

symmetrical multi-threading (SMT) capabilities. SMT enables two threads of<br />

execution (contexts) to execute simultaneously on the same processor core. This<br />

can improve the performance of many applications through the exploitation of<br />

otherwise idle execution units within the processor core.<br />

<strong>The</strong> other significant innovation in the POWER5 processor design is the<br />

integration of a memory controller onto the processor chip to decrease the<br />

latency to memory and improve system reliability.<br />

2.3.2 Vector/SIMD Multimedia eXtension<br />

<strong>The</strong> Vector/SIMD Multimedia eXtension (VMX) is an extension to the PowerPC<br />

Architecture. It defines additional registers and instructions to support<br />

single-instruction multiple-data (SIMD) operations that accelerate data-intensive<br />

tasks.<br />

<strong>The</strong> VMX extensions to the PowerPC Architecture were developed jointly by<br />

Apple Computer, <strong>IBM</strong>, and Motorola. Apple Computer and Motorola use different<br />

terminology to refer to the VMX extensions of the PowerPC Architecture.<br />

Specifically, Motorola uses the term Altivec, and Apple uses the term Velocity<br />

Engine.<br />

A short vector processing history<br />

<strong>The</strong> basic concept behind vector processing is to enhance the performance of<br />

data-intensive applications by providing hardware support for operations that can<br />

manipulate an entire vector (or array) of data in a single operation. <strong>The</strong> number<br />

of data elements operated upon at a time is called the vector length.<br />

38 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Scalar processors perform operations that manipulate single data elements such<br />

as fixed-point or floating-point numbers. For example, scalar processors usually<br />

have an instruction that adds two integers to produce a single-integer result.<br />

Vector processors perform operations on multiple data elements arranged in<br />

groups called vectors (or arrays). For example, a vector add operation to add two<br />

vectors performs a pairwise addition of each element of one source vector with<br />

the corresponding element of the other source vector. It places the result in the<br />

corresponding element of the destination vector. Typically a single vector<br />

operation on vectors of length n is equivalent to performing n scalar operations.<br />

Figure 2-10 illustrates the difference between scalar and vector operations.<br />

Scalar Add Operation<br />

7<br />

1<br />

4 + 7 11<br />

6<br />

3<br />

3<br />

10<br />

+<br />

Figure 2-10 Scalar and vector operations<br />

Vector Add Operation<br />

Processor designers are continually looking for ways to improve application<br />

performance. <strong>The</strong> addition of vector operations to a processor architecture is one<br />

method that a processor designer can use to make it easier to improve the peak<br />

performance of a processor. However, the actual performance improvements that<br />

can be obtained for a specific application depend on how well the application can<br />

exploit vector operations.<br />

<strong>The</strong> concept of vector processing has existed since the 1950s. Early<br />

implementations of vector processing (known as array processing) were installed<br />

in the 1960s. <strong>The</strong>y used special purpose peripherals attached to general<br />

purpose computers. An example is the <strong>IBM</strong> 2938 Array Processor, which could<br />

be attached to some models of the <strong>IBM</strong> System/360. This was followed by the<br />

<strong>IBM</strong> 3838 Array Processor in later years.<br />

By the mid-1970s, vector processing became an integral part of the main<br />

processor in large supercomputers manufactured by companies such as Cray<br />

Research. By the mid-1980s, vector processing became available as an optional<br />

feature on large general-purpose computers such as the <strong>IBM</strong> 3090.<br />

In the 1990s, developers of microprocessors used in desktop computers adapted<br />

the concept of vector processing to enhance the capability of their<br />

microprocessors when running desktop multimedia applications. <strong>The</strong>se<br />

5<br />

6<br />

11<br />

4<br />

5<br />

2<br />

12<br />

7<br />

17<br />

7<br />

8<br />

12<br />

Chapter 2. Hardware components 39


capabilities were usually referred to as Single Instruction Multiple Data (SIMD)<br />

extensions and operated on short vectors. Examples of SIMD extensions in<br />

widespread use today include:<br />

► Intel Multimedia Extensions (MMX)<br />

► Intel Streaming SIMD Extensions (SSE)<br />

► AMD 3DNow!<br />

► Motorola Altivec and <strong>IBM</strong> VMX<br />

<strong>The</strong> SIMD extensions found in microprocessors used in desktop computers<br />

operate on short vectors of length 2, 4, 8, or 16. This is in contrast to the classic<br />

vector supercomputers that can often exploit long vectors of length 64 or more.<br />

VMX extensions to PowerPC Architecture<br />

<strong>The</strong> VMX extensions to PowerPC Architecture add a vector processor (VXU) to<br />

the PowerPC logical processing model that was described earlier in “PowerPC<br />

User Instruction Set Architecture” on page 33. This is illustrated in Figure 2-11.<br />

Fixed-Point<br />

Instructions<br />

Fixed-Point<br />

Processing<br />

(FXU)<br />

Instructions<br />

from Storage<br />

Figure 2-11 PowerPC with VMX logical processing model<br />

<strong>The</strong> VXU operates on vectors that are a total of 128-bits long. <strong>The</strong>se can be<br />

interpreted by the VXU as either:<br />

► A vector of sixteen 8-bit bytes<br />

► A vector of eight 16-bit half words<br />

► A vector of four 32-bit words<br />

40 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong><br />

Branch<br />

Processing<br />

(BPU)<br />

Floating-Point<br />

Instructions<br />

Floating-<br />

Point<br />

Processing<br />

(FPU)<br />

Storage<br />

Data to/from<br />

Storage<br />

Vector<br />

Processing<br />

(VXU)<br />

Vector<br />

Instructions


<strong>The</strong> VMX extensions to PowerPC Architecture define 32 vector registers that<br />

form the vector register file (VRF). <strong>The</strong> VRF is architecturally distinct from the<br />

standard PowerPC FPRs and GPRs.<br />

<strong>The</strong> VMX extensions to PowerPC also define two additional registers:<br />

► <strong>The</strong> VMX status and control register (VSCR), which is used to control the<br />

operation of the VXU and report the status of some VMX operations<br />

► <strong>The</strong> VRSAVE register, which can be used to assist the operating system save<br />

state across context switches by providing a mechanism for software to<br />

indicate what vector registers are in use<br />

<strong>The</strong> additional registers provided by VMX are illustrated in Figure 2-12.<br />

VRSAVE<br />

0 31<br />

VSCR<br />

0 31<br />

Figure 2-12 VMX registers<br />

VRSAVE Register<br />

Vector Status and Control Register<br />

Vector Register File (VRF)<br />

VR0<br />

VR1<br />

VR31<br />

0 127<br />

<strong>The</strong> VMX extensions to PowerPC Architecture define new instructions that use<br />

the VXU to manipulate vectors stored in the VRF. <strong>The</strong>se instructions fall into<br />

these categories:<br />

► Vector integer arithmetic instructions (on 8-bit, 16-bit, or 32-bit integers)<br />

► Vector floating-point arithmetic instructions (32-bit only)<br />

► Vector load and store instructions<br />

► Vector permutation and formatting instructions<br />

► Processor control instructions used to read and write from the VMX status<br />

and control register<br />

► Memory control instructions used to manage caches<br />

...<br />

Chapter 2. Hardware components 41


For additional information about the topics presented in this chapter, we<br />

recommend that you read the PowerPC Microprocessor Family: AltiVec<br />

Technology Programming Environments Manual. You can find this manual on the<br />

Web at:<br />

http://www-306.ibm.com/chips/techlib/techlib.nsf/techdocs/FBFA164F824370F987256<br />

D6A006F424D<br />

2.3.3 Features of the PowerPC 970 and PowerPC 970FX<br />

<strong>The</strong> <strong>IBM</strong> PowerPC 970 Microprocessor is a recent implementation of 64-bit<br />

PowerPC Architecture designed for use in workstations and small servers<br />

requiring from one to four processors. <strong>The</strong> PowerPC 970 is based on the<br />

processor core used in the POWER4 with the addition of a VXU to support the<br />

VMX extensions to the PowerPC Architecture.<br />

<strong>The</strong> <strong>IBM</strong> PowerPC 970FX RISC Microprocessor is an enhanced version of the<br />

PowerPC 970 that takes advantage of improvements in semiconductor<br />

manufacturing processes to deliver higher performance and lower power<br />

consumption. In most respects, the PowerPC 970FX is identical to the PowerPC<br />

970.<br />

Table 2-3 lists the physical attributes of the PowerPC 970 and PowerPC 970FX.<br />

Table 2-3 Physical attributes of PowerPC 970 and PowerPC 970FX<br />

Physical attribute PowerPC 970 PowerPC 970FX<br />

Process geometry 130 nm 90 nm<br />

Physical size 125 mm 2<br />

Unless otherwise noted, the following discussion applies to both the PowerPC<br />

970 and PowerPC 970FX. References to PowerPC 970 also refer to the<br />

PowerPC 970FX.<br />

Figure 2-13 illustrates the internal structure of the PowerPC 970.<br />

42 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong><br />

66 mm 2<br />

Transistor count 58 million 58 million<br />

Package pins 576 576


Figure 2-13 PowerPC 970 internal organization<br />

Chapter 2. Hardware components 43


<strong>The</strong> major features of the PowerPC 970 include:<br />

► Speculative, superscalar RISC processor core with vector extensions<br />

► On-chip level-1 and level-2 caches<br />

► Deeply pipelined design up to 25 stages long<br />

► Aggressive branch prediction<br />

► In order dispatch of up to five operations per cycle<br />

► Out of order issue of up to ten operations per cycle<br />

► In order completion of up to five operations per cycle<br />

► Register renaming<br />

► Up to 215 instructions in-flight<br />

► Fast selective flush of incorrect speculative instructions and results<br />

► Hardware prefetching of up to eight streams including four VMX streams<br />

We elaborate on some of these features in the following sections.<br />

Cache structure<br />

<strong>The</strong> PowerPC 970 includes several on-chip caches to reduce memory latency<br />

when fetching instructions and performing data load and store operations.<br />

Understanding the structure of these caches is sometimes useful when doing<br />

low-level assembly language programming.<br />

<strong>The</strong> on-chip caches include:<br />

► 64 KB, direct-mapped instruction cache<br />

► 32 KB, 2-way set associative data cache<br />

► 128-entry, instruction effective to real address translation (ERAT) cache<br />

instructions<br />

► 128-entry, data ERAT cache<br />

► 64-entry, fully associative segment lookaside buffer (SLB)<br />

► 1024-entry, 4-way set associative translation lookaside buffer (TLB)<br />

► 512 KB, 8-way set associative, level-2 cache<br />

Speculative superscalar features<br />

<strong>The</strong> design of the PowerPC 970 uses a variety of techniques to enable<br />

superscalar operation, where multiple instructions are executed during each<br />

processor clock cycle. This capability is enabled by the exploitation of multiple<br />

execution units within the processor core.<br />

44 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


<strong>The</strong> speculative superscalar techniques used in the PowerPC 970 include:<br />

► Aggressive branch prediction<br />

– Prediction for up to two branches per cycle<br />

– Support for up to 16 predicted branches in flight<br />

– Prediction support for both branch direction and branch addresses<br />

► In order dispatch of up to five operations into a distributed issue queue<br />

structure per cycle<br />

► Out of order issue of up to ten operations into ten execution pipelines per<br />

cycle, comprised of:<br />

– Two load or store operations<br />

– Two fixed-point register-to-register operations<br />

– Two floating-point operations<br />

– One branch operation<br />

– One condition register operation<br />

– One VMX permute operation<br />

– One VMX arithmetic operation<br />

► Register renaming on most architected PowerPC registers including the<br />

GPRs, FPRs, VRFs, condition register fields, FPSCR, VSCR, the link register<br />

(LR), and the count register (CTR)<br />

<strong>The</strong> total number of in-flight instructions within the processor core at any point in<br />

time can be as high as 215.<br />

<strong>The</strong> VMX execution units operate on multiple data elements concurrently.<br />

<strong>The</strong>refore, they provide a performance equivalent to multiple scalar execution<br />

units.<br />

Bus Interface Unit<br />

<strong>The</strong> interface between the PowerPC 970 and external devices is provided by the<br />

Bus Interface Unit (BIU). <strong>The</strong> interface is comprised of two unidirectional paths<br />

(one in, one out) that are each four bytes wide.<br />

<strong>The</strong> internal processor clock operates at an integral multiple of the interface<br />

speed. On the <strong>BladeCenter</strong> <strong>JS20</strong>, the processor clock operates at double the<br />

interface speed.<br />

<strong>The</strong> total aggregate transfer rate across the interface in bytes per second is four<br />

times of the processor clock speed (1/2 x 2 x 4). This equates to 6.4 GBps for<br />

processors operating at 1.6 GHz, and 8.8 GBps for processors operating at<br />

2.2 GHz.<br />

<strong>The</strong> interface is multiplexed, with both addresses and data being transferred<br />

across the same wires. <strong>The</strong>refore the peak theoretical data throughput across<br />

Chapter 2. Hardware components 45


the bus interface is slightly lower than the total aggregate bus transfer rate<br />

derived above.<br />

For more detailed information about the PowerPC 970 and PowerPC 970FX<br />

microprocessors, refer to the <strong>IBM</strong> Microelectronics Division Web site at:<br />

http://www.chips.ibm.com<br />

To access the documentation, follow these steps:<br />

1. Point your Internet browser to:<br />

http://www.chips.ibm.com<br />

2. Click Support.<br />

3. Click Documentation.<br />

4. Under Browse a product family, click PowerPC.<br />

5. Under Categories, click PowerPC 9XX Microprocessors.<br />

6. Under Categories, click PowerPC 970 and 970FX Microprocessors.<br />

Various documents describe the PowerPC Architecture, programming, and 64-bit<br />

computing.<br />

46 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Chapter 3. Software environment<br />

3<br />

This chapter discusses the software environments that are available for the <strong>IBM</strong><br />

Eserver <strong>BladeCenter</strong> <strong>JS20</strong>. It covers the operating systems and <strong>IBM</strong><br />

middleware supported on the <strong>BladeCenter</strong> <strong>JS20</strong>, as well as the system<br />

management tools for the <strong>JS20</strong>.<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. 47


3.1 Operating system support<br />

3.1.1 AIX<br />

<strong>The</strong> <strong>BladeCenter</strong> <strong>JS20</strong> is supported by the AIX and Linux operating systems. At<br />

the time of this writing, AIX 5.2H, AIX 5.3, Red Hat Enterprise Linux 3.0, and<br />

SUSE LINUX Enterprise Server (SLES) are the supported operating systems on<br />

this platform.<br />

Both AIX and Linux operating systems work with a 64-bit kernel and mostly 32-bit<br />

userland programs. Some of the included programs were compiled for 64 bit<br />

when the code fully exploited the 64-bit address space.<br />

AIX has evolved from its beginnings on the <strong>IBM</strong> RT to becoming the operating<br />

system of choice for <strong>IBM</strong>’s largest UNIX servers. AIX is an enterprise operating<br />

system that scales from workstations all the way up to massively parallel super<br />

computers.<br />

Certified as C2 security compliant by the US government, AIX also supports<br />

industry-standard security features such as Pluggable Authentication Modules<br />

(PAM), authenication by x.509 certificates, OpenSSH, Kerberos, and LDAP. AIX<br />

fully exploits the features of the <strong>JS20</strong> with its 64-bit kernel and support for over<br />

16 TB of disk space.<br />

With AIX installed, the <strong>JS20</strong> can run thousands of software titles that were<br />

written for the AIX platform, taking full advantage of AIX’s rich capabilities. At the<br />

time of the writing of this book, AIX Version 5.2H and AIX 5.3 support the<br />

<strong>BladeCenter</strong> <strong>JS20</strong>.<br />

3.1.2 Red Hat Enterprise Linux<br />

Red Hat has made its distribution of Linux available since 1994. Originally, Red<br />

Hat offered its distribution for free download, and sold support contracts. In 2002,<br />

Red Hat began marketing Red Hat Enterprise Linux (RHEL).<br />

Unlike the original Red Hat distribution, RHEL was available only with a<br />

maintenance contract from Red Hat. RHEL runs on x86_64, i386, ia64, ppc/64,<br />

s390, and s390x platforms.<br />

By early 2003, Red Hat stopped developing its non-enterprise distribution and<br />

focused its efforts on RHEL. Red Hat elected to turn development of its free<br />

software over to the community. This free community-supported distribution<br />

became known as Fedora.<br />

48 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Red Hat Enterprise Linux 3 - Update 2 and later is supported on the <strong>JS20</strong>.<br />

Although VMX exploitation was not supported at the time of this book’s<br />

development in the version of GNU C Compiler (gcc) that ships with RHEL 3,<br />

kernel support is available for programs that have been written to exploit VMX<br />

instructions.<br />

3.1.3 SUSE LINUX Enterprise Server<br />

SuSE has been a Linux distributor since 1992 and was acquired by Novell in<br />

early 2004. Originally, SuSE provided a free downloadable distribution, for which<br />

support contracts could be purchased as desired. Eventually, their product line<br />

was split. SuSE now markets SUSE LINUX Desktop, as well as SUSE LINUX<br />

Enterprise Server.<br />

SUSE LINUX Desktop is free to download, where SLES is available only with the<br />

purchase of a maintenance contract. <strong>The</strong> desktop product is also available only<br />

for x86_64 and IA32, where SLES supports IA32, ia64, ppc/64, s390, s390x, and<br />

x86_64.<br />

SLES Version 8 and Version 9 are supported on the <strong>BladeCenter</strong> <strong>JS20</strong>. SLES 9,<br />

the latest version at the time of writing, ships with the 2.6 Linux kernel. This<br />

brings significant gains to performance and scalability.<br />

Both versions of SLES are VMX savvy and have versions of gcc that allow<br />

manual exploitation of the vector engine.<br />

3.2 System management tools<br />

<strong>IBM</strong> has devised a strategy for systems management called the Universal<br />

Manageability Initiative. <strong>The</strong>re are several technologies that have been<br />

incorporated under this umbrella. <strong>The</strong>y all work toward the same goal: simple,<br />

effective solutions for managing heterogeneous environments.<br />

One of the main building blocks in this strategy is the use of simple yet powerful<br />

software to oversee the management of many different systems. This section<br />

investigates some of the possible options that are available for managing such<br />

systems. It reviews the <strong>BladeCenter</strong> Web interfaces, <strong>IBM</strong> Director, Cluster<br />

Systems Management, and xCAT.<br />

3.2.1 <strong>BladeCenter</strong> Web interfaces<br />

<strong>The</strong> <strong>BladeCenter</strong> Web interface allows system administrators to easily and<br />

effectively manage up to 14 blades from an integrated interface. From trivial tasks<br />

such and powering blades on or off, to more complex tasks such as firmware<br />

Chapter 3. Software environment 49


3.2.2 <strong>IBM</strong> Director<br />

management, the Web interface allows powerful control over all blades and<br />

input/output (I/O) modules that are attached to the <strong>BladeCenter</strong> chassis.<br />

Management of other <strong>BladeCenter</strong> resources, such as I/O modules, is also<br />

possible from here, as well as the retrieval of system health information.<br />

<strong>BladeCenter</strong>-specific features are also configured from here such as the Serial<br />

over LAN (SoL).<br />

<strong>IBM</strong> Director assists with the remote management of many <strong>IBM</strong> and non-<strong>IBM</strong><br />

machines including the <strong>BladeCenter</strong> <strong>JS20</strong>. <strong>The</strong> <strong>IBM</strong> Director console allows<br />

system administrators to manage multiple <strong>BladeCenter</strong> chassis from a common<br />

interface. It is the ideal solution for administering <strong>BladeCenter</strong> chassis in<br />

heterogeneous environments or environments where a Director infrastructure<br />

exists.<br />

From the <strong>IBM</strong> Director console, system administrators can monitor resource<br />

utilization. This key feature can be used for performance and capacity planning,<br />

as well as to alert support staff of critical errors such as hardware failure.<br />

<strong>IBM</strong> Director also allows the remote management of software. You can remotely<br />

reconcile and install software from the console interface. This can be useful in<br />

large environments for such resource-intensive tasks as patch management. By<br />

using the Software Distribution Premium Edition, you can remotely install<br />

patches to several servers at the same time with the click of a button.<br />

You can learn more about <strong>IBM</strong> Director in 8.1, “<strong>IBM</strong> Director” on page 140.<br />

3.2.3 <strong>IBM</strong> Cluster Systems Management<br />

<strong>IBM</strong> Cluster Systems Management (CSM) provides many useful functions to<br />

manage a cluster from a single point of control. <strong>The</strong>se include resource<br />

monitoring, automated monitoring and operation, remote hardware control,<br />

remote command execution, security, configuration file management, parallel<br />

network installation, and diagnostics.<br />

By consolidating these capabilities, CSM helps to increase efficiency of an<br />

administrator’s time and reduce their expenses. It helps administrators install<br />

their clusters rapidly by automating many configuration tasks and by leveraging<br />

existing open source products.<br />

CSM also provides efficient monitoring of cluster resources without<br />

overwhelming network bandwidth. <strong>The</strong> automated error detection CSM provides<br />

helps catch problems before they impact the environment, and assists with rapid<br />

50 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


esolution and recovery of problems that occur. CSM’s architecture and modular<br />

construction maximizes flexibility so cluster solutions can evolve and grow as<br />

needs change.<br />

CSM is a collection of components that have been integrated to provide the basis<br />

to construct and manage a cluster. Each of these components provides specific<br />

capabilities related to the management of the cluster. This component-based<br />

architecture provides flexibility for future expansion of the capabilities provided<br />

by CSM. Each of the CSM components can be easily personalized to help meet<br />

specific needs. For example, a cluster administrator can set up monitoring of<br />

application processes and take actions if those processes disappear.<br />

In short, CSM has been designed for large-scale cluster environments that<br />

require simple control over multiple homonymous servers.<br />

This book covers CSM in more detail in 8.2, “<strong>IBM</strong> Cluster Systems Management”<br />

on page 147.<br />

Chapter 3. Software environment 51


52 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Chapter 4. Planning considerations<br />

4<br />

This chapter discusses planning considerations associated with the<br />

implementation of the <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>. Specifically it covers<br />

network planning, operating system installation, and systems management.<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. 53


4.1 Network planning<br />

Successful installation of the <strong>BladeCenter</strong> <strong>JS20</strong> requires that you have a clear<br />

plan for how you will use the various networking capabilities of the <strong>BladeCenter</strong><br />

infrastructure. This plan should address the following questions:<br />

► What network connectivity is needed for the blade servers to support the<br />

applications installed on them?<br />

► What network connectivity is needed to manage the <strong>BladeCenter</strong>,<br />

input/output (I/O) modules, and blade servers?<br />

► What Virtual local area networks (VLANs) are required for each LAN Switch<br />

I/O Module to provide the network connectivity established previously?<br />

► What IP subnet will be used for each VLAN and how will IP addresses be<br />

allocated to each device on the VLAN?<br />

► Will IP addresses be assigned statically or dynamically using DHCP?<br />

► What host names will be used for each network interface?<br />

► How will host names be resolved to IP addresses?<br />

► Are there requirements for a high-performance, low-latency interconnection<br />

network?<br />

► Where multiple <strong>BladeCenter</strong> chassis are installed, how will they be<br />

interconnected?<br />

<strong>The</strong> following sections explore common network planning requirements that we<br />

expect to arise in the installation of <strong>BladeCenter</strong> <strong>JS20</strong>s.<br />

4.1.1 Minimal network requirements<br />

At a minimum, most <strong>BladeCenter</strong> <strong>JS20</strong> environments have the following network<br />

requirements:<br />

► A dedicated hardware management subnet: It is used to access both the<br />

Management Module and management interfaces of I/O modules that are<br />

installed in each <strong>BladeCenter</strong> chassis.<br />

► A Serial over LAN (SoL) subnet internal to each <strong>BladeCenter</strong> chassis that<br />

supports the SoL remote text console function: This is always implemented<br />

using a LAN Switch I/O Module installed in I/O module bay 1.<br />

► A subnet connected to each <strong>BladeCenter</strong> <strong>JS20</strong>: It is used to install and<br />

manage the operating system on the blade server. Where you run AIX on<br />

some blade servers, and Linux on other blade servers, you may need to use<br />

multiple subnets for this purpose.<br />

54 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


► One or more subnets connected to each <strong>BladeCenter</strong> <strong>JS20</strong>: <strong>The</strong>y are used by<br />

applications installed on the blade server to communicate other systems.<br />

Figure 4-1 illustrates how these requirements can be provided in a logical<br />

network view.<br />

I/O<br />

Modules<br />

I/O<br />

Module I/O<br />

Module I/O<br />

Module<br />

<strong>JS20</strong><br />

Blade<br />

Server<br />

Application<br />

Subnet<br />

Figure 4-1 Network logical view<br />

Hardware<br />

Management<br />

Subnet<br />

<strong>JS20</strong><br />

Blade<br />

Server<br />

Management<br />

Module<br />

SOL<br />

Subnet<br />

Console Console Console Console Console Console<br />

<strong>JS20</strong><br />

Blade<br />

Server<br />

<strong>JS20</strong><br />

Blade<br />

Server<br />

<strong>JS20</strong><br />

Blade<br />

Server<br />

OS<br />

Management<br />

Subnet<br />

<strong>JS20</strong><br />

Blade<br />

Server<br />

eth0 eth1 eth0 eth1 eth0 eth1 eth0 eth1 eth0 eth1 eth0 eth1<br />

<strong>The</strong> sections that follow describe each of the logical networks illustrated in<br />

Figure 4-1.<br />

Chapter 4. Planning considerations 55


Hardware management subnet<br />

We recommend that you establish a dedicated hardware management subnet.<br />

<strong>The</strong> 10/100BaseT Ethernet interface on the Management Modules installed in<br />

each <strong>BladeCenter</strong> chassis provides the gateway to this subnet for each<br />

<strong>BladeCenter</strong> chassis. You need to install external LAN switches or hubs to<br />

interconnect the 10/100BaseT Ethernet interface on the Management Modules<br />

of each <strong>BladeCenter</strong> chassis with external management systems.<br />

You use this subnet to access the Management Module Web interface and<br />

command line interface (CLI). You also use this subnet to access the Web<br />

interface and CLI of I/O modules.<br />

System management applications such as <strong>IBM</strong> Director or <strong>IBM</strong> Cluster Systems<br />

Management (CSM) also use this subnet to communicate with the hardware<br />

management functions of the <strong>BladeCenter</strong> infrastructure.<br />

Restrict access to this subnet to those management systems, system<br />

administrators, and operators who have responsibility for managing the<br />

<strong>BladeCenter</strong> infrastructure.<br />

You need to allocate multiple IP addresses to each <strong>BladeCenter</strong> chassis on this<br />

subnet, including:<br />

► One IP address for the external interface of the Management Module in each<br />

<strong>BladeCenter</strong> chassis<br />

► One IP address for the internal interface of the Management Module in each<br />

<strong>BladeCenter</strong> chassis<br />

► One IP address for the management interface of each I/O module in each<br />

<strong>BladeCenter</strong> chassis<br />

Note: Although the logical network view illustrated in Figure 4-1 shows the I/O<br />

Management Module interfaces connecting directly to the hardware<br />

management subnet, they are physically connected via the Management<br />

Module, which acts as gateway to those interfaces. <strong>The</strong> Management Module<br />

performs a proxy Address Resolution Protocol (ARP) function to make it<br />

appear as though the I/O module management interfaces are attached to the<br />

hardware management subnet.<br />

It is possible to use a different subnet for the Management Module internal<br />

network interface and I/O module management interfaces. However, we do not<br />

recommend this configuration.<br />

Learn how to configure these IP addresses in Chapter 5, “Hardware setup” on<br />

page 69.<br />

56 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Serial over LAN subnet<br />

<strong>The</strong> SoL remote text console function requires a subnet and underlying VLAN<br />

that is implemented by a LAN Switch I/O Module installed in I/O module bay 1 of<br />

the <strong>BladeCenter</strong> chassis (see Figure 2-2 on page 9). This subnet and VLAN is<br />

entirely internal to each <strong>BladeCenter</strong> chassis and should not be externally<br />

accessible.<br />

If you use the 4-Port Gigabit Ethernet Switch Module or the Nortel Networks<br />

Layer 2-7 Gigabit Ethernet Switch Module, the VLAN uses VLAN ID 4095. If you<br />

use the Cisco Systems Intelligent Gigabit Ethernet Switch Module, you can<br />

choose the VLAN ID.<br />

You need to assign a unique range of IP addresses to this subnet for use by the<br />

SoL remote text console function.<br />

Important: One IP address is required for each blade server.<br />

You need to specify only the starting IP address within the range of IP addresses<br />

that you assign into the Management Module. <strong>The</strong> Management Module then<br />

automatically assigns consecutive IP addresses from the starting address that<br />

you provide to each blade server that you have installed.<br />

You can learn how to configure the SoL subnet and VLAN in 5.5.1, “Configuring<br />

Serial over LAN” on page 89.<br />

Operating system management subnet<br />

We expect most environments that use the <strong>BladeCenter</strong> <strong>JS20</strong> to rely on the<br />

network installation procedure to install the operating systems. We discuss this<br />

further in 4.2, “Operating system installation” on page 60.<br />

<strong>The</strong> operating system management subnet is used to support both the initial<br />

installation and subsequent management of the operating systems installed on<br />

<strong>BladeCenter</strong> <strong>JS20</strong>s. This subnet is implemented using a VLAN provided by the<br />

LAN Switch I/O Modules installed in I/O module bay 2 of each <strong>BladeCenter</strong><br />

chassis. Alternatively you can implement it using a Pass-Thru I/O Module<br />

installed in I/O module bay 2 that is connected to an external LAN switch.<br />

You may want to install both AIX and Linux operating systems on different<br />

<strong>BladeCenter</strong> <strong>JS20</strong>s in the same environment. In this case, you may need to set<br />

up multiple operating system management subnets and underlying VLANs, one<br />

for blade servers running AIX and the other for blade servers running Linux.<br />

Chapter 4. Planning considerations 57


Application subnet<br />

<strong>The</strong> primary reason you install <strong>BladeCenter</strong> <strong>JS20</strong>s is to support applications.<br />

Many applications have requirements to communicate with other systems. You<br />

use one or more application subnets for this purpose.<br />

<strong>The</strong> primary application subnet is implemented using a VLAN provided by the<br />

LAN Switch I/O Modules installed in I/O module bay 1 of each <strong>BladeCenter</strong><br />

chassis. <strong>The</strong> same LAN Switch I/O Module is used to support the SoL subnet<br />

and VLAN.<br />

Where different <strong>BladeCenter</strong> <strong>JS20</strong>s participate in different applications and there<br />

is a requirement to separate application traffic, you may need to define multiple<br />

application subnets and VLANs. Each <strong>BladeCenter</strong> <strong>JS20</strong> is connected to the<br />

appropriate application subnet based on the applications that are installed on the<br />

blade server.<br />

If application communication requirements with other systems are complex, you<br />

can install an additional pair of Gigabit Ethernet interfaces on each <strong>BladeCenter</strong><br />

<strong>JS20</strong>. You do this using the Gigabit Ethernet Expansion Card in conjunction with<br />

compatible I/O modules installed in I/O module bays 3 and 4.<br />

4.1.2 High-performance, low-latency network requirements<br />

Distributed memory parallel applications may require the installation of a<br />

high-performance, low-latency interconnection network between <strong>BladeCenter</strong><br />

<strong>JS20</strong>s. This requirement can be supported through the use of an optional<br />

Myrinet network.<br />

To install a Myrinet network, you must install a Myrinet Expansion Card on each<br />

<strong>BladeCenter</strong> <strong>JS20</strong> that requires connectivity to the high-performance,<br />

low-latency interconnection network. You must also install an Optical Pass-Thru<br />

I/O Module in I/O module bay 4 of each <strong>BladeCenter</strong> chassis that contains blade<br />

servers equipped with Myrinet Expansion Cards.<br />

<strong>The</strong>n connect the Optical Pass-Thru I/O Module to external Myrinet switches to<br />

complete the Myrinet network infrastructure. This is illustrated in Figure 4-2.<br />

58 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Blade<br />

Server<br />

Myrinet<br />

Expansion<br />

Card<br />

Blade<br />

Server<br />

Myrinet<br />

Expansion<br />

Card<br />

Break-out<br />

Cables<br />

Blade<br />

Server<br />

Myrinet<br />

Expansion<br />

Card<br />

...<br />

Optical Pass-Thru Module<br />

I/O Module Bay 4<br />

External<br />

Myrinet<br />

Switch<br />

Figure 4-2 Myrinet network infrastructure<br />

<strong>The</strong> Myrinet network infrastructure can be used to support application<br />

programming interfaces such as the message passing interface (MPI) that are<br />

often used by distributed memory parallel applications.<br />

You can define IP addresses for each Myrinet network interface. You can also<br />

use the Myrinet network infrastructure to support any application communication<br />

based on IP protocols. For example, you can use this capability to support a<br />

clustered file system such as <strong>IBM</strong> General Parallel File System (GPFS).<br />

You can define a dedicated IP subnet for use with the Myrinet network<br />

infrastructure that is distinct from the other IP subnets that are identified in 4.1.1,<br />

“Minimal network requirements” on page 54.<br />

4.1.3 Multiple <strong>BladeCenter</strong> environments<br />

Blade<br />

Server<br />

Myrinet<br />

Expansion<br />

Card<br />

Blade<br />

Server<br />

Myrinet<br />

Expansion<br />

Card<br />

Blade<br />

Server<br />

Myrinet<br />

Expansion<br />

Card<br />

Large configurations of <strong>BladeCenter</strong> <strong>JS20</strong>s require the installation of multiple<br />

<strong>BladeCenter</strong> chassis. To support the network requirements identified previously,<br />

you need to install external LAN switches to interconnect the <strong>BladeCenter</strong><br />

chassis with each other.<br />

Chapter 4. Planning considerations 59


4.2 Operating system installation<br />

<strong>The</strong>re are two methods to install the <strong>JS20</strong> in a <strong>BladeCenter</strong>. <strong>The</strong> first is an<br />

attended method, using CD media. <strong>The</strong> second is a remote method through a<br />

network installation. Using the CD installation media is similar to other pSeries<br />

systems. You assign the CD-ROM to a blade, follow the installation steps<br />

answering any prompts, and then repeat the process on the next blade.<br />

Refer to the following Web site for instructions to install SuSE from CDs:<br />

http://www.ibm.com/pc/support/site.wss/document.do?sitestyle=ibm&lndocid=<br />

MIGR-54819<br />

With network installation, you can perform several installations at the same time.<br />

<strong>The</strong> method is designed to reduce the installation time required when a large<br />

number of blades requires operating system installation. This method was<br />

chosen as the focus for the following sections.<br />

4.2.1 Network installation planning<br />

You can use two approaches to set up a network installation environment for the<br />

<strong>BladeCenter</strong> <strong>JS20</strong>:<br />

► Establish one or more network installation servers and manually initiate<br />

network installation tasks.<br />

► Use a systems management tool, such as <strong>IBM</strong> Cluster Systems Management<br />

or xCAT, which can be used to automate much of the setup of network<br />

installation servers and initiation of network installation tasks.<br />

This book provides examples of both approaches. <strong>The</strong> remainder of this section<br />

focuses on the planning considerations for the first approach. You can find<br />

planning considerations for the second approach in 4.3, “Systems management”<br />

on page 63.<br />

Required network services<br />

Network installation depends on several different network services:<br />

► A bootstrap protocol (BOOTP) server or dynamic host configuration protocol<br />

(DHCP) server<br />

► A Trivial File Transfer Protocol (TFTP) server<br />

► A Network File System (NFS) server<br />

You can use one physical server to provide all three of these network services. If<br />

you want to install AIX on <strong>BladeCenter</strong> <strong>JS20</strong>s, this server must run AIX. If you<br />

want to install Linux on the <strong>BladeCenter</strong> <strong>JS20</strong>s, we recommend that you use the<br />

60 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


same Linux distribution in the network installation server that you are planning to<br />

install on the <strong>BladeCenter</strong> <strong>JS20</strong>s.<br />

In some situations, you may want to install AIX on some <strong>BladeCenter</strong> <strong>JS20</strong>s and<br />

Linux on other <strong>BladeCenter</strong> <strong>JS20</strong>s. While it is possible to use a single AIX server<br />

to do this, we recommend that you establish two network install servers, one for<br />

installing AIX and the other for installing your chosen Linux distribution.<br />

Important: A potential problem may arise when you have multiple servers<br />

acting as BOOTP or DHCP servers. Be careful when planning your network so<br />

that they do not interfere with one another.<br />

One approach is to place the <strong>BladeCenter</strong> <strong>JS20</strong>s running AIX and the AIX<br />

network install server on a different VLAN from the <strong>BladeCenter</strong> <strong>JS20</strong>s<br />

running Linux and the Linux install server. If you use this approach, you must<br />

also disable the relay of BOOTP or DHCP requests in your network routers.<br />

To learn how to set up AIX network install servers, see Chapter 7, “Installing AIX<br />

on the <strong>JS20</strong>” on page 123. Or to set up Linux network install servers, see<br />

Chapter 6, “Installing Linux” on page 101.<br />

Setting up network installation<br />

<strong>The</strong> <strong>BladeCenter</strong> <strong>JS20</strong> firmware has the capability to boot an operating system<br />

over a network using the BOOTP. You use this capability to initiate network<br />

installation of the <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

<strong>The</strong> BOOTP protocol is a client-server protocol. When initiating a network<br />

installation on a <strong>BladeCenter</strong> <strong>JS20</strong>, it behaves as a BOOTP client. <strong>The</strong>refore,<br />

you need to set up a BOOTP server to support initiating a network installation.<br />

When you instruct the <strong>BladeCenter</strong> <strong>JS20</strong> firmware to boot over a network using<br />

BOOTP, it sends a request to the BOOTP server. <strong>The</strong> BOOTP server should<br />

generate a response that contains the following information:<br />

► <strong>The</strong> IP address, subnet mask, and default gateway that the <strong>BladeCenter</strong> <strong>JS20</strong><br />

should use for the network interface that sent the BOOTP request<br />

► <strong>The</strong> IP address of a TFTP server that the <strong>BladeCenter</strong> <strong>JS20</strong> should contact,<br />

and the name of the file on that server that the <strong>BladeCenter</strong> <strong>JS20</strong> should<br />

request, to get the operating system installation boot image<br />

<strong>The</strong> BOOTP protocol has been around since the mid1980s. In many<br />

environments, BOOTP has now been replaced by the newer DHCP protocol.<br />

DHCP was designed to interoperate with BOOTP, and most DHCP servers can<br />

serve BOOTP clients.<br />

Chapter 4. Planning considerations 61


When you perform network installations of AIX, you use the AIX BOOTP server.<br />

When you perform network installations of Linux, you use a DHCP server that is<br />

configured to support BOOTP clients.<br />

When the <strong>BladeCenter</strong> <strong>JS20</strong> firmware has received a response from a BOOTP<br />

server, it contacts the TFTP server specified in the BOOTP response to load the<br />

operating system’s installation boot image.<br />

<strong>The</strong> operating system’s installation boot image then starts running on the<br />

<strong>BladeCenter</strong> <strong>JS20</strong>. Eventually it contacts the Network File System (NFS) server<br />

to obtain the files that it needs to perform the operating system installation.<br />

Learn the details about the actual process of setting up network installation in<br />

5.5.4, “Specifying IP parameters to Open Firmware” on page 96, and 7.1,<br />

“Minimal NIM installation” on page 124.<br />

Network interface selection<br />

<strong>The</strong> <strong>BladeCenter</strong> <strong>JS20</strong> has two standard Gigabit Ethernet interfaces. <strong>The</strong> first is<br />

connected to I/O module bay 1, and the second is connected to I/O module bay<br />

2. Refer to Figure 2-2 on page 9.<br />

Restriction: <strong>The</strong> <strong>BladeCenter</strong> <strong>JS20</strong> does not support the keyboard, video and<br />

mouse (KVM) console supported by other blade servers. Instead the<br />

<strong>BladeCenter</strong> <strong>JS20</strong> only supports SoL remote text consoles via the<br />

<strong>BladeCenter</strong> Management Module.<br />

<strong>The</strong> SoL remote text console function uses the first Gigabit Ethernet interface on<br />

each <strong>BladeCenter</strong> <strong>JS20</strong> for communication between the Management Module<br />

and the service processor found in each <strong>BladeCenter</strong> <strong>JS20</strong>. This interface is<br />

known as eth0 under Linux and en0 under AIX.<br />

Under normal conditions, using the Gigabit Ethernet interface by the SoL remote<br />

text console function is entirely transparent. It does not impact usage of the<br />

Ethernet interface by the operating system running on the <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

Unfortunately, the initiation of network installation using the BOOTP protocol from<br />

the <strong>BladeCenter</strong> <strong>JS20</strong> firmware temporarily disrupts the operation of the SoL<br />

remote text console when it occurs on the same physical Gigabit Ethernet<br />

interface. This disruption makes it difficult to diagnose problems with the network<br />

installation process. <strong>The</strong>refore, we recommend that you use the second Gigabit<br />

Ethernet interface on each <strong>BladeCenter</strong> <strong>JS20</strong> during network installation of the<br />

operating system. <strong>The</strong> second interface is known as eth1 under Linux and en1<br />

under AIX.<br />

62 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


To use the second interface, you must have a LAN Switch I/O Module, or<br />

Pass-Thru I/O Module connected to an external LAN switch, installed in I/O<br />

module bay 2. You can learn more about why you should use the second Gigabit<br />

Ethernet interface for network installation in the RETAIN® tip H181655 on the<br />

Web at:<br />

http://www-307.ibm.com/pc/support/site.wss/document.do?lndocid=MIGR-55282<br />

4.3 Systems management<br />

4.3.1 <strong>IBM</strong> Director<br />

<strong>The</strong> <strong>BladeCenter</strong> <strong>JS20</strong> supports a rich set of systems management capabilities.<br />

This section discusses planning considerations associated with exploiting some<br />

of those systems management capabilities.<br />

<strong>IBM</strong> Director is a systems management tool. It was originally developed to<br />

provide a comprehensive management solution for servers and workstations<br />

based on Intel microprocessors. Recently its capabilities have been expanded to<br />

support management of other platforms such as the <strong>BladeCenter</strong> <strong>JS20</strong>. We<br />

introduce <strong>IBM</strong> Director in 3.2.2, “<strong>IBM</strong> Director” on page 50.<br />

This section reviews the major planning considerations associated with the<br />

installation of <strong>IBM</strong> Director. For a comprehensive treatment of planning for <strong>IBM</strong><br />

Director, refer to the <strong>IBM</strong> Redbook Implementing System Management Solutions<br />

using <strong>IBM</strong> Director, SG24-6188.<br />

<strong>IBM</strong> Director has three components:<br />

► <strong>IBM</strong> Director Server: This is the core of the <strong>IBM</strong> Director system<br />

management solution.<br />

► <strong>IBM</strong> Director Console: This provides the user interface that system<br />

administrators and operators use to interact with <strong>IBM</strong> Director.<br />

► <strong>IBM</strong> Director Agents: <strong>The</strong>se provide the mechanism by which the <strong>IBM</strong><br />

Director can monitor and control a specific server and operating system<br />

environment.<br />

You can only install the <strong>IBM</strong> Director Agent on the <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

Chapter 4. Planning considerations 63


<strong>IBM</strong> Director Agent for the <strong>BladeCenter</strong> <strong>JS20</strong><br />

Support for the <strong>BladeCenter</strong> <strong>JS20</strong> was introduced in <strong>IBM</strong> Director Version 4.12.<br />

You can learn how to install the <strong>IBM</strong> Director Agent in 8.1.2, “Installing an <strong>IBM</strong><br />

Director Agent” on page 145.<br />

We recommend that you install the <strong>IBM</strong> Director Server on an xSeries server.<br />

For maximum functionality, this server needs IP connectivity to the Management<br />

Module in each <strong>BladeCenter</strong> chassis and to every blade server. This connectivity<br />

can be provided by connecting the <strong>IBM</strong> Director Server to both the hardware<br />

management subnet and operating system management subnets described in<br />

4.1.1, “Minimal network requirements” on page 54.<br />

If your environment is primarily comprised of servers running Linux, consider<br />

installation of the <strong>IBM</strong> Director Server under Linux.<br />

4.3.2 <strong>IBM</strong> Cluster Systems Management<br />

<strong>IBM</strong> Cluster Systems Management is a systems management tool. It simplifies<br />

the management of clusters of servers running the AIX and Linux operating<br />

systems.<br />

This section focuses on the planning considerations that are associated with<br />

using CSM to manage <strong>BladeCenter</strong> <strong>JS20</strong>s. For more planning information, refer<br />

to the <strong>IBM</strong> Cluster Systems Management for Linux Planning and Installation<br />

Guide, SA22-7873, and the <strong>IBM</strong> Cluster Systems Management for AIX5L<br />

Planning and Installation Guide, SA22-7919.<br />

Supported CSM and operating system releases<br />

Preliminary support for the <strong>BladeCenter</strong> <strong>JS20</strong> was introduced in CSM for Linux<br />

Version 1.3.3.1.<br />

CSM requires specific support for each operating system version that is running<br />

on servers that it manages. Currently (at the time of publication) CSM Version 1.4<br />

supports the following operating systems on the <strong>BladeCenter</strong> <strong>JS20</strong>:<br />

► Red Hat Enterprise Linux Version 3 (QU3)<br />

► AIX 5L Version 5.2 and 5.3<br />

► SUSE LINUX Enterprise Server (SLES) Version 8.1and 9<br />

To check for updates to this list of supported operating systems, refer to the CSM<br />

publication site at:<br />

http://publib.boulder.ibm.com/clresctr/windows/public/clusterbooks.html<br />

64 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Management node selection<br />

Each CSM cluster requires a management node, which provides the central point<br />

of administration and control for the entire cluster. CSM has complex rules<br />

concerning the compatibility of management nodes, non-management nodes<br />

(cluster nodes), and the operating system they run. We attempt to simplify these<br />

rules by focusing exclusively on the requirements for a management node that<br />

only manages <strong>BladeCenter</strong> <strong>JS20</strong>s. If your environment is more complex, refer to<br />

the CSM planning documents.<br />

You can use a CSM management node to both install and manage <strong>BladeCenter</strong><br />

<strong>JS20</strong>s, or you can use a management node to manage <strong>BladeCenter</strong> <strong>JS20</strong>s that<br />

were installed using other mechanisms. Most environments use CSM to both<br />

install and manage <strong>BladeCenter</strong> <strong>JS20</strong>s, so we focus on that scenario.<br />

If you use a management node to both install and manage <strong>BladeCenter</strong> <strong>JS20</strong>s<br />

running Linux, you must use the same type of Linux distribution on both the<br />

management node and the <strong>BladeCenter</strong> <strong>JS20</strong>s. However, the management node<br />

does not need the same processor architecture that is used in the <strong>BladeCenter</strong><br />

<strong>JS20</strong>.<br />

For example, if you plan to install SLES on <strong>BladeCenter</strong> <strong>JS20</strong>s, you can use the<br />

following types of management node and operating system combinations to both<br />

install and manage the <strong>BladeCenter</strong> <strong>JS20</strong>s:<br />

► A supported xSeries server running SLES<br />

► A supported pSeries server running SLES<br />

► A <strong>IBM</strong> HS20 Blade Server running SLES<br />

► A <strong>IBM</strong> <strong>BladeCenter</strong> <strong>JS20</strong> running SLES<br />

In large installations, we recommend that you use either an xSeries or pSeries<br />

server as a management node. If your environment also contains<br />

non-<strong>BladeCenter</strong> <strong>JS20</strong>s, such as the HS20 Blade Server, or xSeries cluster<br />

nodes, we recommend that you use an xSeries management node such as the<br />

xSeries 345.<br />

We performed our testing of CSM using an xSeries management node, since<br />

many users of the <strong>BladeCenter</strong> <strong>JS20</strong> may have environments that includes HS20<br />

Blade Servers.<br />

Chapter 4. Planning considerations 65


Network considerations and summary<br />

<strong>IBM</strong> CSM is designed to work in the type of network environment that is outlined<br />

in 4.1, “Network planning” on page 54, with multiple subnets that segregate<br />

different types of network traffic. <strong>The</strong> CSM management node is normally<br />

equipped with multiple Ethernet network interfaces that are connected as follows:<br />

► One interface connects to the hardware management subnet to support<br />

CSM’s hardware management functions.<br />

► One interface connects to the operating system management subnet to<br />

support CSM’s capabilities to install operating systems, maintain software<br />

and configuration files, and provide ongoing monitoring of cluster nodes.<br />

► One interface connects to an external network to enable remote access to the<br />

management node by system administrators and operators.<br />

For further information about the <strong>IBM</strong> Cluster Systems Management facility, refer<br />

to the online documentation that is available.<br />

66 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Part 2 Implementing<br />

Part 2<br />

the<br />

<strong>BladeCenter</strong> <strong>JS20</strong><br />

This part discusses implementation of the <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>. It<br />

covers planning considerations and setting up hardware. It explains how to install<br />

AIX 5L and Linux. Plus it presents systems management scenarios.<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. 67


68 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Chapter 5. Hardware setup<br />

5<br />

This chapter explains the various activities that you must perform before you can<br />

install an operating system on the <strong>BladeCenter</strong> <strong>JS20</strong>. Such activities include<br />

configuring the Management Module and LAN switch modules and updating the<br />

firmware. <strong>The</strong>y also include configuring blade servers and the Serial over LAN<br />

(SoL) remote text console function and using the Open Firmware interfaces.<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. 69


5.1 Management Module configuration<br />

This section describes the steps to set up the <strong>BladeCenter</strong> Management Module<br />

so that you can work with <strong>BladeCenter</strong> <strong>JS20</strong>s. Advanced configuration of the<br />

Management Module is beyond the scope of this book.<br />

<strong>The</strong> primary setup task for the Management Module is to assign IP addresses,<br />

which are necessary to communicate with the Management Module Web and<br />

command line interfaces. To learn more about selecting these IP addresses, see<br />

Chapter 4, “Planning considerations” on page 53.<br />

<strong>The</strong> Management Module has two network interfaces, an external interface<br />

(eth0) and an internal interface (eth1). <strong>The</strong> external interface is accessible via the<br />

10/100BaseT connector on the Management Module. <strong>The</strong> internal interface is<br />

connected to the management interfaces of all the installed I/O modules that<br />

support such interfaces. This includes all switch I/O modules.<br />

<strong>The</strong> default behavior of a new Management Module is to request an IP address<br />

for the external interface via DHCP. If the Management Module does not receive<br />

a valid response from a DHCP server within two minutes of powering on, it uses<br />

a default static IP address of 192.168.70.125 with the default subnet mask of<br />

255.255.255.0.<br />

Note: <strong>The</strong> default host name is MMxxxxxxxxxxxx, where xxxxxxxxxxxx is the<br />

burned-in medium access control (MAC) address.<br />

<strong>The</strong> default IP address for the internal interface is statically assigned to be<br />

192.168.70.126 with a subnet mask of 255.255.255.0.<br />

You can reset the IP addresses of a Management Module that was previously<br />

configured back to the factory defaults by using the IP reset button shown in<br />

Figure 5-1 on the Management Module. You can find the procedure for doing this<br />

in the <strong>IBM</strong> Eserver <strong>BladeCenter</strong> Management Module User’s Guide.<br />

70 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Figure 5-1 Management Module controls and indicators<br />

Configuring the Management Module<br />

Use the following procedure to set the IP addresses for the Management Module<br />

external and internal interfaces:<br />

1. Connect the Management Module to an isolated private Ethernet network.<br />

Also connect a workstation that has a Web browser to the same isolated<br />

private Ethernet network.<br />

2. Configure a static IP address for the workstation Ethernet interface that is in<br />

the same subnet as the Management Module default IP addresses.<br />

For example, we used a static address of 192.168.70.100 with a subnet mask<br />

of 255.255.255.0 for the workstation Ethernet interface. Do not use addresses<br />

in the range of 192.168.70.125 through 192.168.70.130. <strong>The</strong>se addresses<br />

conflict with the default addresses assigned by the Management Module.<br />

3. Connect to the Management Module Web interface by pointing your Web<br />

browser on the workstation to:<br />

http://192.168.70.125<br />

4. Enter a valid user ID and password. <strong>The</strong> factory default configuration of a<br />

Management Module defines a user ID named USERID with a password of<br />

Chapter 5. Hardware setup 71


PASSW0RD. <strong>The</strong> 0 in the password is a zero. In production environments,<br />

consider changing these defaults.<br />

5. From the Management Module Web interface, select MM Control → Network<br />

Interfaces.<br />

6. <strong>The</strong> form window shown in Figure 5-2 is displayed. Enter the desired external<br />

and internal IP addresses, subnet masks, and default gateway for the<br />

Management Module. <strong>The</strong>n click Save to store the new IP addresses.<br />

Figure 5-2 IP address configuration<br />

72 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong><br />

Note: <strong>The</strong> default user ID is USERID, and the default password is<br />

PASSW0RD. <strong>The</strong> number zero is between the W and the R.


7. Restart the Management Module. In the Management Module Web interface,<br />

select MM Control → Restart MM. <strong>The</strong>n click the Restart button on the<br />

displayed form. You are prompted to confirm the restart before it occurs.<br />

8. Remove the Management Module from the isolated private Ethernet network<br />

and connect it to the network you will use to manage the <strong>BladeCenter</strong>.<br />

You can now connect to the Management Module Web and command line<br />

interfaces using the IP address that you assigned to the Management Module<br />

external network interface.<br />

Now consider performing other Management Module setup tasks, such as:<br />

► Setting the Management Module date and time so that log entries have useful<br />

timestamps<br />

► Defining user IDs and passwords for the system administrators and operators<br />

who will manage the <strong>BladeCenter</strong><br />

Alternatively, you can configure the Management Module to use a Lightweight<br />

Directory Access Protocol (LDAP) directory for this purpose.<br />

► Configuring the Management Module to send alerts to management systems<br />

via SNMP, or system administrators using e-mail via Simple Mail Transfer<br />

Protocol (SMTP)<br />

► Enabling the use of Secure Sockets Layer (SSL) to securely access the<br />

Management Module Web interface<br />

► Enabling the use of Secure Shell (SSH) to securely access the Management<br />

Module command line interface (CLI)<br />

For additional information about how to perform these tasks, refer to the<br />

<strong>IBM</strong> Eserver <strong>BladeCenter</strong> Management Module User’s Guide.<br />

5.2 LAN Switch I/O Module configuration<br />

This section explains the basic set up LAN Switch I/O Modules so you can work<br />

with <strong>BladeCenter</strong> <strong>JS20</strong>s. Advanced configuration of LAN Switch I/O Modules is<br />

beyond the scope of this book.<br />

<strong>The</strong> primary setup tasks for LAN Switch I/O Modules are to:<br />

► Assign IP addresses to the LAN Switch I/O Module management interfaces<br />

► Enable the LAN Switch I/O Module external Ethernet ports<br />

You can learn about the selection of IP addresses for the LAN Switch I/O Module<br />

management interfaces in Chapter 4, “Planning considerations” on page 53.<br />

Chapter 5. Hardware setup 73


When a new LAN Switch I/O Module is first installed, the Management Module<br />

assigns a default IP address to the management interface of the LAN Switch I/O<br />

Module. <strong>The</strong> default IP address is chosen based on the I/O module bay where<br />

the LAN Switch I/O Module is installed. Those I/O modules installed in I/O<br />

module bays 1, 2, 3, and 4 are assigned IP addresses 192.168.70.127,<br />

192.168.70.128, 192.168.70.129, and 192.168.70.130, respectively.<br />

Set the IP address of each LAN Switch I/O Module management interface and<br />

enable the external ports:<br />

1. Connect to the Management Module using the Web browser interface as<br />

explained in “Configuring the Management Module” on page 71.<br />

2. From the Management Module Web interface, select I/O Module Tasks →<br />

Management.<br />

3. In the form shown in Figure 5-3, scroll down to the entry for the LAN Switch<br />

I/O Module that you want to configure. Enter the IP address, subnet mask,<br />

and gateway that you want to assign to the LAN Switch I/O Module<br />

management interface. Click the Save button to activate the new IP address.<br />

Figure 5-3 LAN Switch I/O Module IP address<br />

74 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


4. You are prompted to confirm that you want to change the IP address.<br />

5. Select the Advanced Management link for the LAN Switch I/O Module.<br />

6. Scroll down to the Advanced Setup section of the displayed form. This is<br />

illustrated in Figure 5-4. For the External Ports field, select Enabled from the<br />

drop-down list and then click the Save button.<br />

Figure 5-4 LAN Switch I/O Module setup<br />

Chapter 5. Hardware setup 75


At this point, the LAN Switch I/O Module management interface has an IP<br />

address. Also, the external ports on the LAN switch module are enabled so that<br />

they can be used to communicate with blade servers.<br />

<strong>The</strong> SoL remote text console function of the Management Module depends on a<br />

VLAN provided by a LAN Switch I/O Module installed in I/O module bay 1. This<br />

VLAN is automatically provided by the 4-Port Gigabit Ethernet Switch Module<br />

and the Nortel Networks Layer 2-7 Gigabit Ethernet Switch Module using VLAN<br />

4095.<br />

If you are using a Cisco Systems Intelligent Gigabit Ethernet Switch Module in<br />

I/O module bay 1, you must manually configure this VLAN. A procedure is<br />

documented in the Cisco Systems Intelligent Gigabit Ethernet Switch Module for<br />

the <strong>IBM</strong> Eserver <strong>BladeCenter</strong> Installation Guide. This Cisco guide describes<br />

how to manually set up the VLAN that is needed to support the SoL remote text<br />

console function.<br />

5.3 Blade server configuration<br />

Minimal configuration is required for each <strong>BladeCenter</strong> <strong>JS20</strong> prior to installing an<br />

operating system. <strong>The</strong> two main configuration tasks are:<br />

► Assigning a name to the blade server<br />

► Setting the blade server boot sequence<br />

<strong>The</strong> easiest way to perform these tasks is through the Management Module Web<br />

interface.<br />

76 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


5.3.1 Assigning names to blade servers<br />

Set the name of each blade server in a <strong>BladeCenter</strong> chassis by using these<br />

steps:<br />

1. From the Management Module Web interface, select Blade Tasks →<br />

Configuration.<br />

2. In the Blade Information section of the form, as shown in Figure 5-5, type the<br />

name that you want to assign to each blade server.<br />

3. Scroll down the form and select the Save button to save the names that you<br />

have assigned to each blade server.<br />

Figure 5-5 Blade server naming<br />

Chapter 5. Hardware setup 77


5.3.2 Setting the boot sequence<br />

Set the boot sequence of each blade server by using this procedure:<br />

1. From the Management Module Web interface, select Blade Tasks →<br />

Configuration.<br />

2. Scroll down the form to the Boot Sequence section to see the current boot<br />

sequence for all the blade servers.<br />

3. If a blade server does not have the correct boot sequence, you can change<br />

the boot sequence by selecting the blade server name, which displays the<br />

form shown in Figure 5-6.<br />

4. Set the desired boot sequence using the drop-down lists and click the Save<br />

button to set the new boot sequence.<br />

<strong>The</strong> correct boot sequence depends on the method that you plan to use to install<br />

an operating system on the blade server.<br />

Figure 5-6 Boot sequence of blades<br />

78 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


5.4 Firmware<br />

You may need to upgrade various firmware components before you install an<br />

operating system on a <strong>BladeCenter</strong> <strong>JS20</strong>. <strong>The</strong> minimum firmware levels required<br />

and the actual firmware levels we used during the development of this book are<br />

listed in Table 5-1.<br />

To check the level of firmware components, use the following two steps:<br />

1. Find the latest firmware levels and update packages distributed on the <strong>IBM</strong><br />

support Web site.<br />

a. Go to the <strong>IBM</strong> support Web site:<br />

http://www.ibm.com/pc/support<br />

b. Under Download, select Driver matrices.<br />

c. On the Personal computing drivers Web page, select Servers.<br />

d. On the Software and device drivers - Servers Web page, select <strong>eServer</strong><br />

<strong>BladeCenter</strong>.<br />

e. On the “Software and device drivers - <strong>IBM</strong> <strong>eServer</strong> <strong>BladeCenter</strong> (Type<br />

8677)” Web page, you see the firmware levels for the <strong>BladeCenter</strong>.<br />

f. You can also select <strong>IBM</strong> <strong>eServer</strong> <strong>BladeCenter</strong> <strong>JS20</strong> - Software and<br />

device drivers to see <strong>JS20</strong>-specific firmware levels.<br />

2. Identify the current build levels and levels of firmware that are currently<br />

installed by displaying the firmware vital product data (VPD). You can view the<br />

firmware VPD for most firmware components, as shown in Table 5-1, through<br />

the Management Module Web interface. This is illustrated in Figure 5-7.<br />

Table 5-1 <strong>BladeCenter</strong> and <strong>BladeCenter</strong> <strong>JS20</strong> firmware components<br />

Firmware component Minimum level Level we used<br />

Management Module V1.09 (BRET58A) V1.14 (BRET67I)<br />

4-Port Gigabit Ethernet<br />

Switch Module<br />

V1.04 (ibmrun.081) V1.07 (ibmrun.095)<br />

<strong>BladeCenter</strong> <strong>JS20</strong> BIOS FW04310120 FW04310120<br />

<strong>BladeCenter</strong> <strong>JS20</strong> BSMP V1.00 (BQ8T16A) V1.0 (BQ8T20A)<br />

Note: FW04310120 for the <strong>JS20</strong> BIOS is the minimum required supported<br />

level to support AIX 5L releases like V5.2 (ML4).<br />

Chapter 5. Hardware setup 79


Figure 5-7 Viewing firmware levels on <strong>BladeCenter</strong><br />

<strong>The</strong> following sections explain how to update the various firmware components<br />

associated with the <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

5.4.1 Management Module firmware<br />

<strong>The</strong> easiest way to upgrade the firmware of the Management Module is through<br />

the Management Module Web interface.<br />

1. Download the firmware update package from the <strong>IBM</strong> support Web site. This<br />

is usually a ZIP file that contains several files with a PKT extension.<br />

80 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


2. Unzip the firmware update package into a directory on the workstation where<br />

you are running the Web browser that you are using to connect the<br />

Management Module Web interface.<br />

3. Update each PKT file to the Management Module as explained in the<br />

following steps.<br />

a. In the navigation panel of the Management Module Web interface<br />

(Figure 5-8), select MM Control → Firmware Update.<br />

b. In the Update MM Firmware pane (right side of Figure 5-8), select the<br />

Browse button and locate the PKT file in the directory where you<br />

unpacked it.<br />

c. Click the Update button to initiate the download of the PKT file to the<br />

Management Module. This step may take some time to complete.<br />

Figure 5-8 Updating the Management Module firmware<br />

d. A window opens that shows you the progress of the download.<br />

e. Click the Continue button when prompted to confirm the firmware update.<br />

This step may also take some time to complete.<br />

f. Another window opens that shows you the progress.<br />

Chapter 5. Hardware setup 81


4. After you download all the PKT files to the Management Module, restart the<br />

Management Module to make the firmware updates active. In the<br />

Management Module Web interface, select MM Control → Restart MM.<br />

5.4.2 LAN Switch I/O Module firmware<br />

<strong>The</strong> method you should use to upgrade the firmware of a LAN switch module<br />

depends on the type of switch module. We demonstrate how you can upgrade<br />

firmware for the 4-port Gigabit Ethernet Switch Module, which is the switch<br />

module we used in our testing.<br />

For information about upgrading firmware in the Nortel Networks Layer 2-7<br />

Gigabit Ethernet Switch Module, refer to the Nortel Networks Layer 2-7 GbE<br />

Switch Module for <strong>IBM</strong> Eserver <strong>BladeCenter</strong> Installation Guide.<br />

For information about upgrading firmware in the Cisco Systems Intelligent<br />

Gigabit Ethernet Switch Module, refer to the Cisco Systems Intelligent Gigabit<br />

Ethernet Switch Module for the <strong>IBM</strong> Eserver <strong>BladeCenter</strong> Installation Guide.<br />

1. Download the firmware update package from the <strong>IBM</strong> support Web site. This<br />

is usually a ZIP file that contains the firmware image file.<br />

2. Unzip the firmware update package into a directory on the workstation where<br />

you are running the Web browser that you use to connect the Management<br />

Module Web interface.<br />

3. Before you can upgrade the firmware of the 4-Port Gigabit Ethernet Switch<br />

Module, configure an IP address for the switch module management<br />

interface. This enables the capability to directly connect to the switch module<br />

Web interface. To configure the IP address of the switch module management<br />

interface, see 5.2, “LAN Switch I/O Module configuration” on page 73.<br />

4. Connect to the switch module Web interface via the Management Module<br />

Web interface using the following procedure:<br />

a. In the Management Module Web interface, select I/O Module Tasks →<br />

Management.<br />

b. Select the Advanced Management link for the I/O module bay where the<br />

switch module is installed.<br />

c. Scroll down to the bottom of the page to the Start Telnet/Web Session<br />

category. Click the Start Web Session button.<br />

d. A new window opens for the switch module Web interface. You are<br />

prompted to provide a user ID and a password. <strong>The</strong> default user ID is<br />

USERID, and the default password is PASSW0RD, where the number 0 is<br />

between the letters W and R in the password. Consider changing the<br />

defaults in a production environment.<br />

82 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


You can also connect directly to the switch module user interface from any<br />

Web browser if you know the IP address of the switch module.<br />

5. Upgrade the firmware using the following procedure:<br />

a. From the switch module Web interface, select Maintenance → Using<br />

Browser → Upgrade Firmware/Configuration File.<br />

b. You now see the form shown in Figure 5-9. Select the Browse button and<br />

locate the switch module firmware file in the directory where you put it<br />

earlier.<br />

c. Select the Start button to initiate the switch module firmware update<br />

process.<br />

d. A window opens requesting confirmation. After you confirm the update,<br />

the main window displays the status of the update until it completes and<br />

reboots the switch module.<br />

Figure 5-9 Firmware upgrade of 4-Port Gb Ethernet Switch Module<br />

6. After you restart the switch module, verify that the IP address of the switch<br />

module is still set correctly through the Management Module Web interface. In<br />

some cases, you may need to reset the IP address.<br />

Chapter 5. Hardware setup 83


5.4.3 <strong>BladeCenter</strong> <strong>JS20</strong> firmware (BIOS)<br />

<strong>The</strong>re is currently no way to upgrade the firmware (BIOS) of a <strong>BladeCenter</strong> <strong>JS20</strong><br />

without first installing an operating system on the blade server. This is usually not<br />

an issue, because newly shipped <strong>BladeCenter</strong> <strong>JS20</strong>s already have current<br />

firmware installed.<br />

For completeness, we explain how to upgrade the firmware of a <strong>BladeCenter</strong><br />

<strong>JS20</strong> after you install an operating system.<br />

Upgrading firmware under AIX<br />

AIX includes the update_flash utility as part of the AIX diagnostics package. To<br />

use the update_flash utility, you must first download the latest firmware from the<br />

<strong>IBM</strong> support Web site. <strong>The</strong>n transfer it to a disk that is accessible to the operating<br />

system running on the <strong>BladeCenter</strong> <strong>JS20</strong>. Next, you start the update_flash utility<br />

from the directory where the new firmware is located. <strong>The</strong> update_flash utility<br />

requires that you have root privileges.<br />

<strong>The</strong> following example illustrates the usage of update_flash:<br />

/usr/lpp/diagnostics/bin/update_flash -f JS1FW419A.IMG<br />

In this example, JS1FW419A.IMG is the firmware image that was previously<br />

downloaded from the <strong>IBM</strong> support Web site. <strong>The</strong> update_flash utility asks you to<br />

confirm that you want to update the firmware and then reboot the operating<br />

system to perform the actual update.<br />

If you subsequently have a problem with the new firmware and want to revert to<br />

the previous firmware level, use the following command:<br />

/usr/lpp/diagnostics/bin/update_flash -r<br />

Alternatively, if you are satisfied with the new firmware, commit the firmware<br />

update before you install any future firmware by using this command:<br />

/usr/lpp/diagnostics/bin/update_flash -c<br />

Upgrading firmware under Linux<br />

<strong>The</strong> mechanism used to update the firmware (BIOS) of the <strong>BladeCenter</strong> <strong>JS20</strong><br />

under the Linux operating system is based on a Linux kernel module called<br />

rtas_flash. This kernel module is available in most recent 2.4 kernels and in all<br />

2.6 kernels. All supported Linux distributions for the <strong>BladeCenter</strong> <strong>JS20</strong> include<br />

this kernel module.<br />

<strong>The</strong> rtas_flash kernel module creates interfaces for manipulating the flash<br />

memory where the firmware is stored in the /proc/ppc64/rtas directory. In most<br />

Linux distributions, this kernel module is normally not loaded at system boot.<br />

84 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


<strong>The</strong>refore, the interfaces to manipulate the flash memory are not normally<br />

present in /proc/ppc64/rtas.<br />

<strong>IBM</strong> provides a script for Linux called update_flash that uses the interfaces<br />

provided by the rtas_flash kernel module. It also supports the same options and<br />

syntax as the update_flash utility under AIX.<br />

You can obtain the latest version of the update_flash script by installing the<br />

diagnostic utilities for Linux on POWER that <strong>IBM</strong> distributes from the Web at:<br />

http://techsupport.services.ibm.com/server/lopdiags<br />

Note: Some Linux distributions provide a version of the update_flash script<br />

that is not fully functional. Always obtain the latest version of the update_flash<br />

script from <strong>IBM</strong> before you attempt to update firmware.<br />

To use the update_flash script, follow these steps:<br />

1. Download the latest firmware from the <strong>IBM</strong> support Web site.<br />

2. Transfer it to a disk that is accessible to the operating system running on the<br />

<strong>BladeCenter</strong> <strong>JS20</strong>.<br />

3. Enter the update_flash command from the directory where the new firmware<br />

is located. <strong>The</strong> update_flash command requires that you have root privileges.<br />

<strong>The</strong> following example illustrates the usage of update_flash:<br />

/usr/sbin/update_flash -f JS1FW419A.IMG<br />

In this example, JS1FW419A.IMG is the firmware image that was previously<br />

downloaded from the <strong>IBM</strong> support Web site. <strong>The</strong> update_flash utility copies the<br />

new firmware image into the kernel and then reboots the operating system to<br />

perform the actual firmware update.<br />

If you subsequently have a problem with the new firmware and want to revert to<br />

the previous firmware level, then use the following command:<br />

/usr/sbin/update_flash -r<br />

If you are happy with the new firmware, commit the firmware upgrade before you<br />

install any future firmware by using the following command:<br />

/usr/sbin/update_flash -c<br />

5.4.4 Integrated Systems Management Processor firmware<br />

<strong>The</strong> easiest way to upgrade the firmware of the <strong>BladeCenter</strong> <strong>JS20</strong> Integrated<br />

Systems Management Processor (ISMP) is through the Management Module<br />

Web interface.<br />

Chapter 5. Hardware setup 85


1. Download the firmware update package from the <strong>IBM</strong> support Web site. This<br />

package contains a file with a ZIP extension, which is the firmware for the<br />

ISMP.<br />

2. Place the file into a directory on the workstation where you will run the Web<br />

browser that you will use to connect the Management Module Web interface.<br />

3. Download the PKT file to the ISMP. Follow this upgrade procedure:<br />

a. As shown in Figure 5-10, select Blade Tasks → Firmware Update in the<br />

Management Module Web interface.<br />

b. <strong>The</strong> Update Blade Firmware pane is displayed (right side of Figure 5-10).<br />

In the Target drop-down list, select the blade server on which you want to<br />

upgrade the ISMPfirmware.<br />

c. Click the Browse button and locate the ISMP firmware ZIP file in the<br />

directory where you placed it earlier.<br />

d. Click the Update button to initiate download of the ZIP file to the<br />

Management Module. This step may take some time to complete.<br />

Figure 5-10 Blade firmware update<br />

86 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


e. A window opens that shows the progress of the download.<br />

f. Click the Continue button when prompted to confirm the firmware update.<br />

This step may also take some time to complete.<br />

g. A window opens that shows the progress.<br />

<strong>The</strong> new ISMP firmware is now active in the target <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

5.5 Providing a console for the <strong>JS20</strong> blades<br />

<strong>The</strong> <strong>BladeCenter</strong> <strong>JS20</strong> differs from other blade servers that are available for the<br />

<strong>BladeCenter</strong> in that it does not provide an interface to a keyboard, video monitor,<br />

and mouse (KVM) console. <strong>The</strong>refore, you must set up the SoL remote text<br />

console function to provide a console for a <strong>BladeCenter</strong> <strong>JS20</strong>. Use of SoL is<br />

optional for other types of blade server that support KVM consoles.<br />

<strong>The</strong> SoL remote text console function involves several different components in<br />

the <strong>BladeCenter</strong> infrastructure as illustrated in Figure 5-11.<br />

<strong>The</strong> SoL remote console function works as follows:<br />

1. Using a Telnet or SSH client, connect to the <strong>BladeCenter</strong> Management<br />

Module CLI. This is usually via an external management network that is<br />

connected to the Management Module’s 10/100BaseT Ethernet interface.<br />

2. From the Management Module’s CLI, initiate a SoL remote console session to<br />

the desired blade server.<br />

3. <strong>The</strong> Management Module uses a private VLAN provided by the LAN switch<br />

module in I/O module bay 1, to transport the SoL data stream to the Ethernet<br />

interface of the target blade server.<br />

4. <strong>The</strong> Ethernet controller of the target blade server passes the SoL data stream<br />

received from the private VLAN to the blade system management processor<br />

(BSMP), which manages the text console for the blade server.<br />

Chapter 5. Hardware setup 87


LAN Switch<br />

Module in<br />

I/O Module<br />

Bay 1<br />

BSMP<br />

Figure 5-11 Serial over LAN components<br />

<strong>The</strong> sections that follow explain how to configure and use the SoL remote text<br />

console function.<br />

88 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong><br />

Telnet<br />

or<br />

SSH Client<br />

External Management Network<br />

Ethernet<br />

Interface<br />

(eth0)<br />

Management<br />

Module<br />

External Interface<br />

(eth0)<br />

Internal Interface<br />

(eth1)<br />

Ethernet<br />

Interface<br />

(eth1)<br />

Ethernet Controller<br />

<strong>JS20</strong> Blade Server<br />

I/O Module<br />

Bay 2


5.5.1 Configuring Serial over LAN<br />

Before you attempt to configure the SoL remote text console function, verify that<br />

you have all the prerequisites in place:<br />

► A supported LAN switch module installed in I/O module bay 1<br />

This switch module is used to provide a VLAN that connects the Management<br />

Module to the first Ethernet interface on each blade server.<br />

► <strong>The</strong> minimum firmware levels described in 5.4, “Firmware” on page 79<br />

This is important if you install a <strong>BladeCenter</strong> <strong>JS20</strong> in an existing <strong>BladeCenter</strong><br />

chassis that may have back-level firmware in the Management Module or LAN<br />

switch modules.<br />

► A reliable Telnet or SSH client<br />

► A network connecting the Telnet or SSH client to the <strong>BladeCenter</strong><br />

Management Module external 10/100BaseT Ethernet interface<br />

► An identified range of IP addresses that will be used by the Management<br />

Module to communicate with the BSMP on each blade server via the private<br />

VLAN<br />

This was discussed in Chapter 4, “Planning considerations” on page 53.<br />

<strong>The</strong> easiest way to configure the SoL remote text console function is through the<br />

Management Module Web interface as explained in the following steps.<br />

To configure the SoL remote text console function, use the following procedure:<br />

1. From the Management Module Web interface, select Blade Tasks → Serial<br />

Over LAN.<br />

2. In the right pane, scroll down until you see the Serial Over LAN Configuration<br />

section (see Figure 5-12). Complete the following tasks:<br />

a. From the Serial over LAN list, select the Enabled option.<br />

b. Leave the value for SoL VLAN ID at the default (4095) if you have either a<br />

4-port Gigabit Ethernet Switch Module or a Nortel Networks Layer 2-7<br />

Gigabit Ethernet Switch Module installed in I/O module bay 1.<br />

If you have a Cisco Systems Intelligent Gigabit Ethernet Switch Module<br />

installed in I/O module bay 1, set the VLAN ID to the same value you used<br />

when you manually defined the VLAN that supports SoL.<br />

c. Enter the start of the IP address range that will be used by the<br />

Management Module to communicate with the BSMP on each blade<br />

server.<br />

d. Leave the values for Accumulate timeout, Send threshold, Retry count,<br />

and Retry interval at their defaults (5, 250, 3, and 250).<br />

Chapter 5. Hardware setup 89


e. Leave the values for User Defined Keystroke Sequences at their defaults.<br />

f. Click the Save button.<br />

Figure 5-12 Serial over LAN configuration<br />

3. Restart the Management Module.<br />

a. In the Management Module Web interface, select MM Control → Restart<br />

MM.<br />

b. Click the Restart button on the displayed form.<br />

c. You are prompted to confirm the restart before it occurs.<br />

4. After the Management Module has restarted and you have reconnected to the<br />

Management Module Web interface from your Web browser, enable the SoL<br />

remote text console for each blade server.<br />

a. Select Blade Tasks → Serial Over LAN from the Management Module<br />

Web interface.<br />

b. In the right pane, scroll down until you see the Serial Over LAN Status<br />

section (see Figure 5-13).<br />

c. Select the blade servers you want to enable for SoL. You can choose all of<br />

them by clicking the check box at the top of the table.<br />

90 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


d. Click the Enable Serial Over LAN link underneath the table. You may<br />

need to scroll down to see this link.<br />

Figure 5-13 Serial over LAN status<br />

e. After a few seconds, the window refreshes. In the SoL column of the table,<br />

verify that each blade server has a status of Enable.<br />

<strong>The</strong> configuration of the SoL remote text console function is now complete.<br />

5.5.2 Using Serial over LAN<br />

Access to the SoL remote text console function is via the Management Module<br />

CLI. <strong>The</strong> CLI is documented in the <strong>IBM</strong> ^ <strong>BladeCenter</strong> Management<br />

Module Command-Line Interface Reference Guide.<br />

You can connect to the Management Module CLI using either a Telnet or SSH<br />

client. Figure 5-14 shows an active Telnet session that displays the help<br />

command. <strong>The</strong> Management Module can support up to 20 concurrent CLI<br />

Chapter 5. Hardware setup 91


connections. This is sufficient to support the concurrent use of a SoL remote text<br />

console to each blade server in a full <strong>BladeCenter</strong> chassis. At the same time, it<br />

supports six additional CLI connections for other administrative activities.<br />

Figure 5-14 Management Module’s command line interface<br />

Restriction: You can only have one active SoL remote text console<br />

connection to each blade server.<br />

<strong>The</strong> Management Module CLI is context sensitive. When you issue CLI<br />

commands, you can accept the current context or override the context using the<br />

-T option provided on most commands. In our examples, we use the -T option to<br />

make it clear on which entity the command is operating. You can change the<br />

context for the Management Module CLI by using the env -T command.<br />

92 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


To use the SoL remote text console function, first connect to the Management<br />

Module from a Telnet or SSH client. You are prompted to enter a user ID and<br />

password. <strong>The</strong> default user ID is USERID and the default password is<br />

PASSW0RD, where 0 in the default password is a zero. Consider changing the<br />

defaults in a production environment.<br />

<strong>The</strong>re are several different commands that you can use to access the SoL<br />

remote text console function. <strong>The</strong> most useful commands are listed in Table 5-2.<br />

Table 5-2 Management Module commands for SoL<br />

Command What it does<br />

console Open a SoL remote text console for the blade server. This<br />

command fails if another SoL remote text console is already open<br />

for the blade server.<br />

console -o Terminate any existing SoL remote text console for the blade server<br />

and open a SoL remote text console for the blade server.<br />

boot -c Reset the blade server, and then open a SoL remote text console<br />

for the blade server.<br />

reset -c This is functionally equivalent to the boot -c command when used<br />

in a blade server context.<br />

power -on -c Power on the blade server and then open a SoL remote text<br />

console for the blade server.<br />

power -cycle -c Power on the blade server and then open a SoL remote text<br />

console for the blade server. If the blade server is already powered<br />

on, power it off first, and then power on.<br />

Here are some examples of how to use these commands. To open a SoL remote<br />

text console to the blade server in bay 3, you use the following command:<br />

console -T system:blade[3]<br />

To reset the blade server in bay 2 and then open a SoL remote text console to the<br />

blade server, you use the following command:<br />

boot -c -T system:blade[2]<br />

To terminate an active SoL remote text console, press the Esc key followed by an<br />

open parenthesis (Shift+9 on U.S. keyboards). When the SoL remote text<br />

console ends, you return to the Management Module CLI prompt.<br />

Inactivity timeout<br />

<strong>The</strong> Management Module CLI has an aggressive default inactivity timeout. If you<br />

do not use the CLI or a SoL remote text console accessed through the CLI for 2<br />

Chapter 5. Hardware setup 93


minutes, you are disconnected. You can increase the timeout using the<br />

telnetcfg command. For example, the following command increases the<br />

inactivity timeout to 10 minutes (600 seconds):<br />

telnetcfg -t 600 -T system:mm[1]<br />

Setting CRLF<br />

Over the years, there have been many variations on the interpretation of the<br />

Enter key in a Telnet session. Some Telnet daemons expect carriage return +<br />

linefeed (CRLF) and some expect carriage return alone (CR). From our testing,<br />

CR seems to be the most compatible. During your SoL session, if when you<br />

press the Enter key, the key behaves as though it was pressed twice and not<br />

once, then correct the CRLF option. Under Windows 2000 Telnet, do this as<br />

shown in Figure 5-15 before you open a connection to the Management Module.<br />

Microsoft (R) Windows 2000 (TM) Version 5.00 (Build 2195)<br />

Welcome to Microsoft Telnet Client<br />

Telnet Client Build 5.00.99206.1<br />

Escape Character is 'CTRL+]'<br />

Microsoft Telnet> unset crlf<br />

Microsoft Telnet> open bcmm<br />

Figure 5-15 Setting CRLF in Windows 2000 Telnet<br />

Refer to the developer of your client for other Telnet implementations.<br />

5.5.3 Open Firmware interfaces<br />

In some situations, you may need to access the Open Firmware interfaces to the<br />

<strong>BladeCenter</strong> <strong>JS20</strong> firmware. This section explains how to access Open Firmware<br />

interfaces and how to use some of the functions they provide.<br />

Activating the Open Firmware interfaces<br />

You can access the Open Firmware interface whenever you power on a<br />

<strong>BladeCenter</strong> <strong>JS20</strong>. Before you can access the Open Firmware interface, set up<br />

the SoL remote text console function as described in 5.5, “Providing a console for<br />

the <strong>JS20</strong> blades” on page 87.<br />

Use the following procedure to access the Open Firmware interface:<br />

1. Using a Telnet or SSH client, connect to the Management Module external<br />

Ethernet interface IP address.<br />

94 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


2. When prompted, enter a valid user ID and password. <strong>The</strong> default<br />

Management Module user ID is USERID, and the default password is<br />

PASSW0RD, where the 0 in the default password is a zero. Consider<br />

changing the defaults in any production environment.<br />

3. Power cycle the blade and start an SoL console by using the power -cycle -c<br />

command. For example, to power cycle and start a SoL remote text console<br />

with the blade server in the first bay (bay 1), use the command:<br />

power -cycle -c -T system:blade[1]<br />

4. After approximately 30 seconds, you see a sequence of checkpoint codes<br />

displayed on the console. <strong>The</strong>se codes are generated by the Power On Self<br />

Test (POST). <strong>The</strong> meaning of the checkpoint codes is documented in the<br />

<strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong> Type 8842 Installation and User’s Guide.<br />

5. Wait until the checkpoint code D5BB is displayed and then type 8. <strong>The</strong> POST<br />

pauses for approximately five seconds at this checkpoint code to give you<br />

time to respond to the checkpoint code. A number is displayed after the D5BB<br />

checkpoint code, indicating how many seconds remain before the POST will<br />

proceed further. For example, Figure 5-16 shows that four seconds remain.<br />

E153 U8842.P1Z.23A1073-P1-T7<br />

EAA1 U8842.P1Z.23A1073-P1<br />

E172 U8842.P1Z.23A1073-P1<br />

EAA1 U8842.P1Z.23A1073-P1<br />

E152 U8842.P1Z.23A1073-P1<br />

E153 U8842.P1Z.23A1073-P1<br />

E152 U8842.P1Z.23A1073-P1<br />

E153 U8842.P1Z.23A1073-P1<br />

EAA1 U8842.P1Z.23A1073-P1<br />

E172 U8842.P1Z.23A1073-P1<br />

EAA1 U8842.P1Z.23A1073-P1-C5<br />

D001<br />

D003<br />

D004<br />

E139<br />

E14A<br />

D008<br />

E1F0<br />

E1F1<br />

D099<br />

D5BB 4<br />

Figure 5-16 POST checkpoint codes<br />

Chapter 5. Hardware setup 95


6. As soon as you type 8, the Open Firmware command prompt is displayed.<br />

If you do not type any key when the D5BB checkpoint code is displayed, the<br />

blade server proceeds to boot using the default boot sequence. To learn how to<br />

configure the boot sequence, see 5.3, “Blade server configuration” on page 76.<br />

5.5.4 Specifying IP parameters to Open Firmware<br />

You can use a directed bootp to install a <strong>JS20</strong> blade from a NIM server. bootp<br />

does not require the NIM server to be on the same subnet as the <strong>JS20</strong> blade.<br />

Nor does this option require that you have the MAC address of the network<br />

adapter on the <strong>JS20</strong> blade.<br />

To perform a directed bootp, you need an SoL connection to the blade so that<br />

you can specify the IP parameters to Open Firmware. Currently you must have<br />

two network adapters to perform a NIM installation if you are using SoL. You<br />

cannot install AIX 5L over the same adapter that is using SoL.<br />

Step 1: Preparing your NIM server<br />

Follow these steps to prepare your NIM server.<br />

1. Create a SPOT, lpp_source, and any other resources that you need at the<br />

level of AIX 5L that you want to install on your NIM server. Your NIM server is<br />

usually the NIM master, but you can also set up a NIM client as a NIM server.<br />

For instructions on how to create NIM resources, see 7.1, “Minimal NIM<br />

installation” on page 124.<br />

2. Ensure that you have the information in the following worksheet (Table 5-3) for<br />

your <strong>JS20</strong> blade before you proceed with the installation.<br />

Table 5-3 Network configuration information worksheet<br />

Network attribute Value<br />

Network interface<br />

Host name<br />

IP address _______.________.________.________<br />

Network mask _______.________.________.________<br />

Name server _______.________.________.________<br />

Domain name _______.________.________.________<br />

Gateway _______.________.________.________<br />

96 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


3. Define the <strong>JS20</strong> blade as a NIM client on your NIM master by running the<br />

following command on the NIM master:<br />

smitty nim_mkmac<br />

This command creates a client definition for your <strong>JS20</strong> blade. You can also<br />

define the <strong>JS20</strong> blade using the define NIM operation on the command line.<br />

4. If you want to set the <strong>JS20</strong> blade's name server and domain name after the<br />

installation, use a resolv_conf resource.<br />

5. Set up your NIM master to install the <strong>JS20</strong> blade, by running this command:<br />

smitty nim_bosinst<br />

Select the <strong>JS20</strong> blade that you defined earlier as your target. <strong>The</strong>n select the<br />

type of install that you want to perform and select the installation resources<br />

that you want to use to install the <strong>JS20</strong> blade. You can also prepare the <strong>JS20</strong><br />

blade to install using the bos_inst NIM operation on the command line.<br />

Note: If the <strong>JS20</strong> blade is powered off or was never installed, set “Initiate<br />

reboot and installation now?” to no and press Enter in the SMIT interface.<br />

If the <strong>JS20</strong> blade is powered on and running AIX 5L, set “Initiate reboot<br />

and installation now?” to yes in the SMIT interface. If you choose this<br />

option, a directed bootp is initiated by default and you can skip step 2.<br />

Before you run this command, ensure that the <strong>JS20</strong> blade is a registered<br />

NIM client.<br />

1. Run smitty niminit on the <strong>JS20</strong> blade.<br />

2. Specify the host name of your NIM master and the interface you want to<br />

use for the installation.<br />

You can also initialize the <strong>JS20</strong> blade using the niminit command on the<br />

command line.<br />

Step 2: Specifying a directed bootp from the <strong>JS20</strong> blade<br />

Specify a directed bootp from the <strong>JS20</strong> blade by using these steps:<br />

1. Open a Web interface to the Management Module by navigating to the IP<br />

address or host name of the Management Module using a Web browser.<br />

2. Enable SoL to the <strong>JS20</strong> blade from the Management Module Web interface.<br />

Select Blade Tasks →Serial Over LAN.<br />

3. Select the <strong>JS20</strong> blade that you are installing and click Enable Serial Over<br />

LAN.<br />

4. Power on the <strong>JS20</strong> blade from the Management Module Web interface Blade<br />

Tasks →Power/Restart.<br />

Chapter 5. Hardware setup 97


5. Select the <strong>JS20</strong> blade that you are installing and click Power On Blade.<br />

6. Open a Serial over LAN connection to the <strong>JS20</strong> blade by using Telnet into the<br />

Management Module and running the console command. For example, if the<br />

<strong>JS20</strong> blade is in slot 3, run the following command:<br />

console -T blade[3]<br />

<strong>The</strong> Serial over LAN connection shows a series of LED numbers.<br />

7. Type 8 when you see D5BB to go to the Open Firmware prompt.<br />

8. Boot from the network by typing:<br />

boot net:bootp,server_ip,,client_ip,gateway_ip<br />

Because SoL uses ent0, you must use ent1. Run a command similar to this<br />

one (note the double comma):<br />

boot<br />

/pci@8000000f8000000/pci@0/ethernet@1,1:bootp,192.168.2.10,,192.168.1.11,192.168.1.1<br />

98 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong><br />

Note: You must specify the full device path name with this command. To<br />

determine the full path to your device, list the device tree by running the ls<br />

command at the Open Firmware prompt. This command displays output<br />

similar to this:<br />

0 > ls<br />

000000c87f18: /ibm,serial<br />

000000c88840: /chosen<br />

000000c88a98: /packages<br />

...<br />

000000d31488: /vdevice<br />

000000d327a8: /vty@0<br />

000000d32f88: /<strong>IBM</strong>,sp@4000<br />

000000d33f10: /rtc@4001<br />

000000d34a18: /pci@8000000f8000000<br />

000000d384d0: /pci@0<br />

000000d4bbd0: /ethernet@1<br />

000000d5af50: /ethernet@1,1<br />

000000d3be00: /pci@3<br />

000000d6a350: /usb@0<br />

000000d845f8: /hub@1<br />

000000d854b8: /usb@0,1<br />

000000d9f760: /hub@1<br />

000000d3f798: /pci@1f<br />

000000d45ed8: /ide@4,1<br />

000000d47b10: /disk@0<br />

<strong>The</strong> highlighted items are the path to the second Ethernet adapter. You<br />

pass this information to the boot command to initiate a network boot from<br />

the second Ethernet adapter.


After you run the boot command, the network installation begins. Output similar<br />

to Example 5-1 is displayed on the SoL connection.<br />

Example 5-1 Output of the boot command<br />

BOOTP: chosen-network-type =<br />

ethernet,auto,none,auto<br />

BOOTP: server IP = 192.168.2.10<br />

BOOTP: requested filename =<br />

BOOTP: client IP = 192.168.1.11<br />

BOOTP: client HW addr = 0 d 60 1e c cb<br />

BOOTP: gateway IP = 192.168.1.1<br />

BOOTP: device<br />

/pci@8000000f8000000/pci@0/ethernet@1,1<br />

BOOTP: loc-code U8842.P1Z.23A0984-P1-T7<br />

BOOTP R = 1<br />

FILE: /tftpboot/js20blade1.austin.ibm.com<br />

Load Addr=0x0000000000004000, Max<br />

Size=0x0000000000bfc000<br />

FINAL Packet Count = 21131<br />

FINAL File Size = 10818623 bytes.<br />

load-base=0x4000<br />

real-base=0xc00000<br />

Elapsed time since release of system<br />

processors: 2 mins 28 secs<br />

Chapter 5. Hardware setup 99


100 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Chapter 6. Installing Linux<br />

6<br />

This chapter explains how to install Red Hat Enterprise Linux (RHEL) and SUSE<br />

LINUX Enterprise Server (SLES) on the <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. 101


6.1 Installing Linux<br />

Both Red Hat Enterprise Linux and SUSE LINUX Enterprise Server are<br />

supported operating systems for the <strong>BladeCenter</strong> <strong>JS20</strong>. This chapter explains<br />

how to set up and conduct a network-based installation of both distributions.<br />

Because the general process of doing this is similar for both operating systems,<br />

the focus of this chapter is on the installation of Red Hat Enterprise Linux 3.0 and<br />

SUSE LINUX Enterprise Server 9.0.<br />

Tip: Prior to installing any operating system, we recommend that you upgrade<br />

all firmware to the latest level. See 5.4, “Firmware” on page 79, for instructions<br />

about how to complete this task.<br />

6.1.1 Configuring the sources<br />

Since we are doing a network installation, the first step is to configure a file space<br />

that contains the source files for the operating system installation. For all cited<br />

examples, we created the sources on a pSeries 690 that was connected to the<br />

same local area network (LAN) as the <strong>BladeCenter</strong>. Since there are special tools<br />

available within each distribution for this purpose, we configured each network<br />

install server with a similar distribution. For example, the server that installed<br />

RHEL 3 also ran RHEL 3.<br />

Tip: We recommend a 100 Mb or faster LAN that is local to both the machine<br />

that will host the files and the <strong>BladeCenter</strong>. We also recommend that the file<br />

spaces be shared via Network File System (NFS).<br />

Red Hat sources<br />

<strong>The</strong> Red Hat installer, Anaconda, makes it simple to set up a remote file share.<br />

Simply copy the ISOs for the source CDs into your NFS mount. To create an ISO<br />

of the first CD and place it in /mnt/exports/, enter:<br />

# dd if=/dev/cdrom of=/mnt/exports/RHEL-3.0-U2-disc1.iso bs=32M<br />

Repeat the previous step for all the CDs, changing the file name as appropriate.<br />

You may also increase the bs parameter as appropriate. This parameter controls<br />

block size. <strong>The</strong> larger the block size is, the more RAM is taken for the dd process,<br />

but the faster the process takes.<br />

Important: Ensure that the CD is not mounted before you begin the dd. Also<br />

ensure that the destination of the ISO has enough space to store all the data.<br />

Copying a full complement of RHEL 3 ISOs can take over 4 GB.<br />

102 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Configure the NFS daemon to export the mount point where the ISOs are<br />

located. You can do this by editing /etc/exports. Example 6-1 shows the export<br />

file we used.<br />

Example 6-1 /etc/exports file contents<br />

/mnt/exports/ 192.168.1.0/24(ro)<br />

After the /etc/exports file is created or updated, we enabled NFS as shown in<br />

Example 6-2.<br />

Example 6-2 Starting NFS<br />

# service nfs start<br />

Starting NFS services: [ OK ]<br />

Starting NFS quotas: [ OK ]<br />

Starting NFS daemon: [ OK ]<br />

Starting NFS mountd: [ OK ]<br />

#<br />

Attention: If NFS does not start properly, refer to the Red Hat Enterprise<br />

Linux Reference Guide at the following Web site for troubleshooting<br />

information:<br />

http://www.redhat.com/docs<br />

Chapter 6. Installing Linux 103


SuSE sources<br />

SuSE installations are usually administered via a program called Yet Another<br />

Setup Tool (YaST).<br />

1. Within YaST, there is an applet for configuring installation sources on the<br />

server. Access this from the Misc section in the left pane, as shown in<br />

Figure 6-1.<br />

Figure 6-1 YaST installation server configuration<br />

104 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


2. You now see the Source Configuration panel (Figure 6-2). Select SuSE SLES<br />

Version 9 (ppc).<br />

Figure 6-2 Selecting the distribution<br />

Chapter 6. Installing Linux 105


3. In the next panel as shown in Figure 6-3, you select which media to use as a<br />

source for the SuSE distribution. We found it easier to use ISOs rather than<br />

CDs. Proceeding past this window builds the source directory. You may be<br />

prompted for additional CDs (or locations of ISOs) while the tree is being built.<br />

Tip: If you have any service packs or additional CDs that you want to add,<br />

select the Prompt for Additional CDs check box.<br />

Figure 6-3 Source Configuration pane<br />

106 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


4. After the source files are copied into the installation share, click Finish as<br />

shown in Figure 6-4. This completes any final tasks such as starting the NFS<br />

daemon.<br />

Figure 6-4 Completing the setup<br />

6.1.2 <strong>The</strong> zImage.initrd file<br />

PowerPC-based hosts support booting directly from the network provided that<br />

they are served a zImage.initrd via BOOTP. <strong>The</strong> zImage.initrd is a package that<br />

contains both the kernel and initial RAM disk in a single file.<br />

Most distributions already have a zImage.initrd, which you can use for network<br />

booting. For RHEL, you can find this file on the first CD as images/netboot.img.<br />

For SuSE, this file is called install and is in the root directory of the first CD.<br />

Although these files are referred to by different names, they are the zImage.initrd<br />

file that we require.<br />

Building your own zImage.initrd<br />

Although most users use the zImage.initrd provided by their distribution, you can<br />

also compile your own.<br />

Tip: This section address the nuances of how to create a zImage.initrd. If you<br />

need more information about compiling a kernel, see the Kernel HOWTO at:<br />

http://www.tldp.org<br />

Chapter 6. Installing Linux 107


To build your own zImage.initrd, follow these steps:<br />

1. Obtain the necessary kernel sources and development utilities for your<br />

distribution as appropriate. If the necessary packages are not installed, you<br />

may experience some errors, which cause your compilation to fail. Consult<br />

your distribution’s documentation to determine the names of these packages.<br />

2. With the packages installed, start by cleaning the source tree. This is a good<br />

procedure to ensure that there are no remnants of an old compile left hanging<br />

around. Here is an example of how to do this.<br />

# make mrproper<br />

3. With the source tree clean, configure your kernel as required. <strong>The</strong> preferred<br />

method to do this is to use either make menuconfig (text based) or make<br />

xconfig (graphical; requires an X server).<br />

Both Red Hat and SuSE have their default kernel options selected when<br />

menuconfig / xconfig is launched after having done a make mrproper. From<br />

either interface, select the desired options.<br />

4. After you configure the kernel, perform the compilation. Since RHEL 3 is<br />

based on the 2.4 kernel and SLES 9 is based on 2.6, the next steps are<br />

slightly different. Refer to Example 6-3 and Example 6-4 for each distribution.<br />

Example 6-3 Compiling a kernel in Red Hat<br />

# make dep<br />

# make clean<br />

# make zImage<br />

# make modules<br />

# make modules_install<br />

# make install<br />

Example 6-4 Compiling a kernel in SuSE<br />

# make<br />

# make modules_install<br />

# make install<br />

108 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


5. After you compile the kernel, create the initial RAM disk. Both Red Hat and<br />

SuSE ship with a command called mkinitrd for this purpose. However, the<br />

syntax for each is slightly different, as shown in Example 6-5 and<br />

Example 6-6.<br />

Example 6-5 mkinitrd for Red Hat<br />

# cd arch/ppc64/boot<br />

# mkinitrd -f ramdisk.image 2.4.21-15.ELcustom<br />

# gzip ramdisk.image<br />

# cd /usr/src/linux-2.4<br />

# make zImage.initrd<br />

Example 6-6 mkinitrd for SuSE<br />

# make install<br />

# mkinitrd<br />

# cd arch/ppc64/boot/<br />

# cp /boot/initrd-2.6.5-7.39-pseries64 ramdisk.image<br />

# gzip ramdisk.image<br />

# cd /usr/src/linux<br />

# make zImage.initrd<br />

Attention: <strong>The</strong> file names referred to are all relative to the kernel version<br />

strings within the Makefile for the kernel. <strong>The</strong> files on your system may be<br />

different.<br />

After you complete these steps, you should have a bootable zImage.initrd file in<br />

/boot/ppc64/.<br />

6.2 Configuring BOOTP and TFTP<br />

<strong>The</strong>re are three steps to network booting a POWER/PowerPC-based server.<br />

1. Perform a Power On Self Test (POST) and then receive BOOTP information<br />

from BOOTP server, including the IP, Trivial File Transfer Protocol (TFTP)<br />

server IP, and zImage.initrd file name.<br />

2. TFTP server sends the zImage.initrd file.<br />

3. <strong>The</strong> server boots using the zImage.initrd file.<br />

Both RHEL and SLES ship with Dynamic Host Configuration Protocol (DHCP)<br />

and TFTP packages. Ensure that you have installed the appropriate RPMs.<br />

Chapter 6. Installing Linux 109


6.2.1 BOOTP<br />

DHCP is an extension to the original BOOTP specification. As a result, you can<br />

use the DHCP daemon (dhcpd) to provide the BOOTP information for booting via<br />

the network.<br />

<strong>The</strong> DHCP daemon is configured through the /etc/dhcpd.conf file. Example 6-7<br />

shows the configuration that we used in our lab.<br />

With dhcpd configured, start the DHCP daemon as shown here:<br />

# /etc/init.d/dhcpd start<br />

Example 6-7 /etc/dhcpd.conf<br />

always-reply-rfc1048 true;<br />

allow bootp;<br />

deny unknown-clients;<br />

not authoritative;<br />

default-lease-time 600;<br />

max-lease-time 7200;<br />

subnet 192.168.1.0 netmask 255.255.255.0 {<br />

group {<br />

next-server 192.168.1.1;<br />

}<br />

Attention: Keep in mind that Example 6-7 contains IP and MAC address<br />

information specific to our lab environment. See your product documentation<br />

or the dhcpd man page to customize this for your site.<br />

}<br />

110 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong><br />

host blade1 {<br />

fixed-address 192.168.1.117;<br />

hardware ethernet 00:0d:60:1e:0f:89;<br />

filename "zImage.initrd.suse";<br />

}<br />

host blade2 {<br />

fixed-address 192.168.1.118;<br />

hardware ethernet 00:0d:60:1e:0e:fb;<br />

filename "zImage.initrd.redhat";<br />

}


6.2.2 Trivial File Transfer Protocol<br />

<strong>The</strong> TFTP daemon for both distributions considers the /tftpboot directory its<br />

home. Any file name statements made in the /etc/dhcpd.conf configuration file<br />

should be relative to this path.<br />

Both RHEL and SLES spawn TFTP from the xinetd daemon. By default, tftpboot<br />

is disabled on both distributions. To enable both xinetd and TFTP, enter:<br />

[root@blade1 root]# chkconfig tftp on<br />

[root@blade1 root]# /etc/init.d/xinetd restart<br />

Your installation server is now ready to network boot the <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

6.3 Preparing an unattended install<br />

RHEL and SLES both have their own methods of conducting an unattended<br />

installation. RHEL uses Kickstart, and SLES uses a method known as AutoYaST.<br />

Chapter 6. Installing Linux 111


6.3.1 Red Hat Kickstart<br />

Red Hat provides a utility called redhat-config-kickstart to assist with the<br />

creation of Kickstart files. When you launch this utility, you see the Kickstart<br />

Configurator window (Figure 6-5).<br />

Figure 6-5 redhat-config-kickstart main window<br />

<strong>The</strong> separate sections of this application allow you to specify different attributes<br />

of the installation.<br />

► Basic Configuration<br />

You can configure language support, keyboard, mouse, and root password<br />

here. You may also specify whether you prefer a text installation instead of a<br />

graphical installation. Lastly you may choose if the system should reboot after<br />

the automated installation is complete.<br />

► Installation Method<br />

You can choose an upgrade or clean installation here. An upgrade usually<br />

allows the preservation of existing data by upgrading the installed applications<br />

112 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


from a previous version of Red Hat. You can also specify the source of the<br />

installation (NFS, HTTP, CDROM, etc.) here.<br />

► Boot Loader Options<br />

Select the type of bootloader (GRUB or LILO) along with any relevant options<br />

from here. You may also specify kernel parameters from this window.<br />

► Partition Information<br />

You specify the partition size and type information here. You also specify the<br />

action to take with existing partitions and new disks.<br />

► Network Configuration<br />

Define population network information, such as IP, DNS, and routers, from<br />

here.<br />

► Authentication<br />

You may configure NIS, Keberos, and other authentication server options<br />

here. You may also specify local password encryption and security.<br />

► Firewall Configuration<br />

Configure iptables rules from here.<br />

► Xconfiguration<br />

Select the video card, resolution, monitor, and window manager here.<br />

► Package Selection<br />

Select which programs are installed on the system here.<br />

► Pre-Installation Script<br />

Identify the commands to run before the installation.<br />

► Post-Installation Script<br />

Identify the commands to run after the installation.<br />

When all your options are specified, save the file. This automatically checks for<br />

any missing options. If everything is correct, you may now use the ks.cfg file in an<br />

unattended installation.<br />

Tip: We found that the redhat-config-kickstart program, by default, places a<br />

skipx parameter in the ks.cfg file. We had to comment out this prior to using<br />

ks.cfg. Otherwise, the installation would fail.<br />

Chapter 6. Installing Linux 113


6.3.2 SuSE AutoYaST<br />

SLES provides a YaST applet to assist with the creation and editing of AutoYAST<br />

files. You can launch it from the Misc section of YaST, as shown in Figure 6-6.<br />

Figure 6-6 AutoYaST module within YaST<br />

114 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


After the AutoYaST editor is created, you see a window similar to the one in<br />

Figure 6-7.<br />

Figure 6-7 Main AutoYaST window<br />

You can use the following sections to configure your AutoYaST file.<br />

► Software<br />

Select the packages to install, and configure the automatic update client.<br />

► Hardware<br />

Configure partitioning, audio, printing, and X.<br />

Chapter 6. Installing Linux 115


► System<br />

Set the general system information. You also configure the language, time<br />

zone, and other locale-related settings here. And specify boot loader, logging,<br />

and runlevel information here.<br />

► Network Devices<br />

Set network adapter information. Kernel module information as well as IP<br />

details can be set here.<br />

► Network Services<br />

Network clients and daemons can be configured from here. <strong>The</strong>re are over 15<br />

different daemons to choose from. Refer to the Network Services section for<br />

more information.<br />

► Security and Users<br />

From here, you may create users that you want to be present after install. You<br />

may also change the security policy to make things such as password<br />

changes more stringent. And you can configure VPN and firewall settings<br />

from here.<br />

► Misc<br />

This section allows you to add complete configuration files, or to add special<br />

scripts to run before and after installation.<br />

Tip: AutoYaST also lets you import Kickstart files from Red Hat or Alice (a<br />

precursor to AutoYaST) files. You do this by selecting File → Import.<br />

After you are satisfied with the selections, save the file. This validates that the<br />

required minimum options are populated prior to saving.<br />

6.4 Performing an unattended installation<br />

<strong>The</strong> unattended installation process needs to pass the zImage.initrd the location<br />

of the unattended answer file during boot. Unfortunately, BOOTP does not have a<br />

facility to provide anything more than the location to the zImage.initrd and the<br />

tftpboot server to the client. To pass the required parameters, they need to be<br />

input into Open Firmware.<br />

116 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


6.4.1 Open Firmware<br />

For instructions about accessing the Open Firmware prompt, refer to “Activating<br />

the Open Firmware interfaces” on page 94.<br />

When you reach the Open Firmware prompt, pass bootparm through the boot-file<br />

variable. Example 6-8 and Example 6-9 show the settings for SuSE and Red Hat<br />

respectively.<br />

Example 6-8 Open Firmware settings for SuSE<br />

0 > setenv boot-file autoyast=nfs://192.168.1.113:/mnt/SUSE-Source/autoyast.xml<br />

install=nfs://192.168.1.113:/mnt/SLES-Source/SLES9/ netdevice=eth1<br />

Example 6-9 Open Firmware settings for Red Hat<br />

0 > setenv boot-file ks=nfs:192.168.1.112:/mnt/install-images/ks.cfg<br />

ksdevice=eth1<br />

After the installation is complete, remember to clear the boot-file value. If this<br />

value is not cleared, it is passed to the kernel as bootparm next time you boot.<br />

You can clear the value by using the nvsetenv command. We recommend that<br />

you run this command after the installation completes via either AutoYaST or<br />

Kickstart, as appropriate.<br />

This command is provided by the ppc64-utils RPM in RHEL and the util-linux<br />

RPM in SLES. <strong>The</strong> syntax to clear the boot-file variable is:<br />

# /sbin/nvsetenv boot-file ""<br />

6.4.2 mkzimage_cmdline: SuSE only<br />

mkzimage_cmdline is a utility contained in the SLES 9 lilo RPM. <strong>The</strong> SuSE kernel<br />

has special markers that allow mkzimage_cmdline to pass bootparm information<br />

directly through the zImage.initrd. Example 6-10 shows the syntax.<br />

Example 6-10 mkzimage_cmdline<br />

# /lib/lilo/chrp/mkzimage_cmdline -a 1 -c -s<br />

"autoyast=nfs://192.168.1.113:/mnt/SUSE-Source/autoyast.xmlinstall=nfs://192.16<br />

8.1.113:/mnt/SLES-Source/SLES9/ netdevice=eth1" netboot.suse.img<br />

# /lib/lilo/chrp/mkzimage_cmdline netboot.suse.img<br />

cmd_line size:512<br />

cmd_line: autoyast=nfs://192.168.1.113:/mnt/SUSE-Source/autoyast.xml<br />

install=nfs://192.168.1.113:/mnt/SLES-Source/SLES9/ netdevice=eth1<br />

active: 1<br />

Chapter 6. Installing Linux 117


<strong>The</strong> syntax for mkzimage_cmdline is straightforward. When entered with only a file<br />

name parameter. <strong>The</strong> current bootparm is displayed. You can use the following<br />

options:<br />

-h Display help.<br />

-v Display version.<br />

-a 0|1 Disable/enable command line. Overrides bootparm passed from<br />

Open Firmware.<br />

-s STRING Stores STRING as bootparm in zImage.initrd.<br />

-c Clears previously stored bootparm from zImage.initrd before<br />

applying a new one.<br />

6.5 Performing a network installation<br />

When installing several machines, we recommend that you perform a network<br />

installation instead of a CD-based installation. <strong>The</strong> speed and ability to serve<br />

multiple machines makes it ideal for installing to a full complement of<br />

<strong>BladeCenter</strong> <strong>JS20</strong>s.<br />

Attention: <strong>The</strong> cited examples are provided for illustration purposes only. For<br />

detailed instructions about how to install each distribution, refer to the product<br />

documentation.<br />

6.5.1 SUSE LINUX Enterprise Server 9<br />

For the SUSE LINUX Enterprise Server 9, follow this procedure:<br />

1. After you load the zImage.initrd file, you see a language display similar to the<br />

one in Figure 6-8. In this example, we selected English.<br />

118 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Select the language.<br />

1) Bosnia<br />

2) Cestina<br />

3) Deutsch<br />

4) English<br />

5) Español<br />

6) Français<br />

7) Hellenic<br />

8) Italiano<br />

9) Japanese<br />

10) Magyar<br />

11) Nederlands<br />

12) Polski<br />

13) Português<br />

14) Português Brasileiro<br />

15) Russian<br />

16) Slovencina<br />

><br />

Figure 6-8 Language selection<br />

2. After you select the language, the main installer display appears with a<br />

number of options. In this example, we chose to load additional kernel<br />

modules, as shown in Figure 6-9.<br />

>>> Linuxrc v1.6 (Kernel 2.6.5-7.69-pseries64) (c) 1996-2004 SUSE LINUX AG<br />


3. We select the install option and start the installation, as shown in Figure 6-10.<br />

Start Installation or System<br />

1) Start Installation or Update<br />

2) Boot Installed System<br />

3) Start Rescue System<br />

><br />

Figure 6-10 Installation menu<br />

4. We instruct the installer to use the network as a source for the installation, as<br />

shown in Figure 6-11.<br />

Choose the source medium.<br />

1) CD-ROM<br />

2) Network<br />

3) Hard Disk<br />

><br />

Figure 6-11 Source menu<br />

5. We chose NFS as the protocol to use from the menu in Figure 6-12.<br />

Choose the network protocol.<br />

1) FTP<br />

2) HTTP<br />

3) NFS<br />

4) TFTP<br />

><br />

Figure 6-12 Protocol menu<br />

120 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


6. We chose the following site-specific information. You may have different<br />

information for your environment, but the general process is the same as<br />

shown in Figure 6-13.<br />

Automatic configuration via DHCP?<br />

1) Yes<br />

2) No<br />

> 1<br />

Sending DHCP request...<br />

Enter the IP address of the NFS server [192.168.1.113]><br />

Enter the directory on the server<br />

[/mnt/SLES/SLES9-RC1/]><br />

Figure 6-13 Installation location<br />

With these steps complete the normal YaST process begins.<br />

Tip: With this, the network specific portion of the installation is complete. Refer<br />

to the product documentation for further information.<br />

6.5.2 Red Hat Enterprise Linux 3<br />

After the zImage.initrd is served by the TFTP boot server, the installation process<br />

is the same as a CD-based installation. Refer to the product documentation for<br />

more information regarding this.<br />

Chapter 6. Installing Linux 121


122 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Chapter 7. Installing AIX on the <strong>JS20</strong><br />

7<br />

This chapter discusses NIM. It explains how to configure NIM to install AIX and<br />

how to install AIX on an <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong> using NIM.<br />

<strong>BladeCenter</strong> <strong>JS20</strong> supports the installation of AIX 5.2f and later. We recommend<br />

that you install AIX using a network installation via NIM. NIM is a unified console<br />

that allows the management of several AIX logical partitions (LPARs) or<br />

machines. You may install AIX on new machines or manage existing ones from a<br />

unified interface.<br />

Tip: Prior to installing any operating system, we recommend that you upgrade<br />

all firmware to the latest level. See 5.4, “Firmware” on page 79, for instructions<br />

about how to complete this task.<br />

To use NIM, there should be an available AIX instance with NIM configured on<br />

the same network segment as the <strong>BladeCenter</strong>. In the interest of speed and<br />

security, a private dedicated subnet is recommended.<br />

Our NIM master was configured on a pSeries 690 connected to the same local<br />

area network (LAN) as the <strong>BladeCenter</strong>. This p690 was running the same level of<br />

AIX that we installed on the <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

Note: An installation of AIX may be several gigabytes in size. We recommend<br />

that you use a 100 Mb or faster network for NIM-based installations.<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. 123


7.1 Minimal NIM installation<br />

Follow these steps to install NIM:<br />

1. Launch the Web-based Systems Manager (WebSM) and point to the server<br />

that is running NIM. Select Configure the Network Installation<br />

Management environment from the NIM object shown in Figure 7-1.<br />

Figure 7-1 NIM configuration window<br />

2. In the window that opens (Figure 7-2), select Configure this host as NIM<br />

Master.<br />

124 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Figure 7-2 Configuring as a NIM Master<br />

3. Select the network interface that is bound to the same network as the<br />

<strong>BladeCenter</strong>, and name it appropriately, as shown in Figure 7-3.<br />

Figure 7-3 Selecting the NIM interface and network name<br />

Chapter 7. Installing AIX on the <strong>JS20</strong> 125


4. In the next window (Figure 7-4), select Initialize Minimal NIM Environment.<br />

Figure 7-4 Choosing to initialize minimal NIM environment<br />

5. Click View Settings to confirm the gathered settings and complete the<br />

minimal NIM environment as shown in Figure 7-5.<br />

Figure 7-5 Finalizing the NIM environment settings<br />

126 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


7.2 Adding resources<br />

Follow these steps to add resources:<br />

1. With the minimal NIM environment complete, configure lpp_sources and a<br />

SPOT on the NIM master. Return to the NIM configuration window within<br />

WebSM and select Add a new resource to the Network Installation<br />

Management environment.<br />

2. In the Add New Resource window (Figure 7-6), select the lpp_source &<br />

SPOT option.<br />

Figure 7-6 lpp_source and SPOT configuration<br />

Chapter 7. Installing AIX on the <strong>JS20</strong> 127


3. In the Configure NIM Environment window (Figure 7-7), select the medium<br />

that you will use for your AIX source. We use the DVD RAM drive.<br />

Figure 7-7 AIX source<br />

4. As shown in Figure 7-8, select the default values and begin populating the<br />

NIM resources.<br />

Figure 7-8 Default resource settings<br />

128 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


5. Click View Settings, as shown in Figure 7-9, to confirm the settings.<br />

Assuming a CD-based installation, insert the first CD and proceed to<br />

complete the addition of the new resources.<br />

Figure 7-9 Completing the resource installation<br />

Tip: This step make take some time to complete, since many files must be<br />

incorporated into the resource.<br />

Chapter 7. Installing AIX on the <strong>JS20</strong> 129


7.3 Adding machine information<br />

With the resources added, add information regarding the machine on which you<br />

want to perform a base operating system (BOS) installation.<br />

1. From the main NIM menu (Figure 7-10), select Add a new Machine to the<br />

Network Installation Management environment.<br />

Figure 7-10 Adding new machine information to NIM<br />

130 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


2. In the Add New Machine window (Figure 7-11), enter the host name of the<br />

server that you want to add to the NIM environment from this window.<br />

Figure 7-11 Host name of new machine<br />

Tip: <strong>The</strong> NIM server must be able to resolve the host name of the new<br />

server. Ensure that this host name can be resolved by DNS or a hosts file.<br />

Chapter 7. Installing AIX on the <strong>JS20</strong> 131


3. In the next window (Figure 7-12), select Multiprocessor and confirm the<br />

other metrics of the machine that you are adding to the NIM environment.<br />

Figure 7-12 Entering machine-specific information<br />

132 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


4. As shown in Figure 7-13, enter the MAC address for the machine that you are<br />

adding to the Network adapter hardware address field and proceed to<br />

complete the addition of the new machine to the NIM environment.<br />

Figure 7-13 Entering MAC address information<br />

With the NIM resources created and the machine object created, simply proceed<br />

with the installation.<br />

NIM allows you to install new operating systems on pSeries servers by booting<br />

the machine via BOOTP. This diskless or network installation makes it simple to<br />

install new operating systems to servers in a matter of minutes.<br />

Chapter 7. Installing AIX on the <strong>JS20</strong> 133


7.4 Preparing the NIM master<br />

With all the NIM resources in place, it is simple to begin the installation of a new<br />

operating system to the target machine, in this case a <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

1. Launch WebSM and view the configured machine objects from the NIM<br />

section.<br />

2. Right-click the object for the desired machine and select Install Operating<br />

System, as shown in Figure 7-14.<br />

Figure 7-14 Installing an operating system from NIM<br />

3. Select the location of the SPOT and lpp_source resources created earlier.<br />

134 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


4. Proceed with the installation by setting the BOS preferences as shown in<br />

Figure 7-15.<br />

Figure 7-15 Setting BOS preferences<br />

Tip: To fully automate the AIX installation, create a bosinst_data resource and<br />

associate it with the operating system installation in NIM settings. Although<br />

this is beyond the scope of this book, you can find instructions in the AIX<br />

product manuals.<br />

7.5 Preparing the client<br />

With the NIM server configuration complete, simply boot the <strong>BladeCenter</strong> <strong>JS20</strong><br />

from the network as explained in 5.5.3, “Open Firmware interfaces” on page 94.<br />

When the NIM master supplies the boot image via TFTP, the AIX installation<br />

begins just as though you booted from the CD.<br />

Chapter 7. Installing AIX on the <strong>JS20</strong> 135


1. On the first panel, select the desired language. We selected English, as<br />

shown in Figure 7-16.<br />

******* Please define the System Console. *******<br />

Type a 1 and press Enter to use this terminal as the<br />

system console.<br />

Pour definir ce terminal comme console systeme, appuyez<br />

sur 1 puis sur Entree.<br />

Taste 1 und anschliessend die Eingabetaste druecken, um<br />

diese Datenstation als Systemkonsole zu verwenden.<br />

Premere il tasto 1 ed Invio per usare questo terminal<br />

come console.<br />

Escriba 1 y pulse Intro para utilizar esta terminal como<br />

consola del sistema.<br />

Escriviu 1 1 i premeu Intro per utilitzar aquest<br />

terminal com a consola del sistema.<br />

Digite um 1 e pressione Enter para utilizar este terminal<br />

como console do sistema.<br />

Figure 7-16 Installation language panel<br />

2. In the next panel (Figure 7-17), you may proceed with the default installation<br />

options or customize them. We chose to customize them.<br />

Welcome to Base Operating System<br />

Installation and Maintenance<br />

Type the number of your choice and press Enter. Choice is indicated by >>>.<br />

>>> 1 Start Install Now with Default Settings<br />

Figure 7-17 Operating system selection<br />

136 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong><br />

2 Change/Show Installation Settings and Install<br />

3 Start Maintenance Mode for System Recovery<br />

88 Help ?<br />

99 Previous Menu<br />

>>> Choice [1]:


3. In the Installation and Settings window (Figure 7-18), we proceeded with the<br />

default installation type and language environment values.<br />

Installation and Settings<br />

Either type 0 and press Enter to install with current settings, or type the<br />

number of the setting you want to change and press Enter.<br />

1 System Settings:<br />

Method of Installation.............Preservation<br />

Disk Where You Want to Install.....hdisk0<br />

2 Primary Language Environment Settings (AFTER Install):<br />

Cultural Convention................C (POSIX)<br />

Language...........................C (POSIX)<br />

Keyboard...........................C (POSIX)<br />

3 More Options (Desktop, Security, Kernel, Software, ...)<br />

>>> 0 Install with the settings listed above.<br />

+-----------------------------------------------------<br />

88 Help ? | WARNING: Base Operating System Installation will<br />

99 Previous Menu |destroy or impair recovery of SOME data on the<br />

|destination disk hdisk0.<br />

>>> Choice [0]:<br />

Figure 7-18 Installation and language options<br />

Chapter 7. Installing AIX on the <strong>JS20</strong> 137


4. As illustrated in Figure 7-19, we elected to include additional options such as<br />

the 64-bit kernel and JFS2. We then selected 0 and began the installation<br />

process.<br />

Install Options<br />

1. Enable Trusted Computing Base.................................... no<br />

2. Enable CAPP and EAL4+ Technology................................. no<br />

(English only, 64-bit kernel enablement, JFS2 file systems)<br />

3. Enable 64-bit Kernel............................................. yes<br />

4. Create JFS2 File Systems......................................... yes<br />

5. Graphics Software................................................ yes<br />

6. Documentation Services Software.................................. no<br />

7. Enable System Backups to install any system...................... no<br />

(Installs all devices and kernels)<br />

>>> 8. Install More Software<br />

Figure 7-19 Kernel options<br />

138 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong><br />

0 Install with the settings listed above.<br />

88 Help ?<br />

99 Previous Menu<br />

>>> Choice [8]:<br />

Tip: As with any operating system, AIX enables you to customize many<br />

different options during the installation process. We have provided a sample of<br />

what you can select.


Chapter 8. System management<br />

scenarios<br />

8<br />

Your installation may be comprised of a single <strong>BladeCenter</strong> chassis with only a<br />

few <strong>BladeCenter</strong> <strong>JS20</strong>s. If so, you may find that the capabilities provided by the<br />

<strong>BladeCenter</strong> Management Module Web interface and command line interface<br />

(CLI) are sufficient to meet your system management requirements. For larger<br />

installations comprised of multiple <strong>BladeCenter</strong> chassis and many <strong>BladeCenter</strong><br />

<strong>JS20</strong>s, consider using <strong>IBM</strong> Director, <strong>IBM</strong> Cluster Systems Management (CSM),<br />

or both.<br />

We demonstrate how to use many of the functions provided by the <strong>BladeCenter</strong><br />

Management Module Web interface and CLI in Chapter 5, “Hardware setup” on<br />

page 69.<br />

This chapter provides information about setting up and configuring the <strong>IBM</strong><br />

Director and the <strong>IBM</strong> Cluster Systems Management facilities. It also describes<br />

several systems management scenarios for the <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. 139


8.1 <strong>IBM</strong> Director<br />

<strong>IBM</strong> Director is a comprehensive systems-management solution. A powerful<br />

suite of tools and utilities, <strong>IBM</strong> Director automates many of the processes<br />

required to manage systems proactively, including preventive maintenance,<br />

diagnostic monitoring, troubleshooting, and more. It offers a graphical user<br />

interface (GUI) that provides system administrators easy access to both local<br />

and remote systems. <strong>IBM</strong> Director can be used in environments with multiple<br />

operating systems (heterogeneous environments).<br />

At the time of writing, the current version of <strong>IBM</strong> Director was Version 4.2.<br />

<strong>IBM</strong> Director is the industry-leading client/server workgroup manager. <strong>IBM</strong><br />

Director’s tools provide customers with flexible capabilities to realize maximum<br />

system availability and lower IT costs. With <strong>IBM</strong> Director, IT administrators can<br />

view and track the hardware configuration of remote systems in detail and<br />

monitor the usage and performance of critical components, such as processors,<br />

disks, and memory.<br />

Extensions to <strong>IBM</strong> Director are available for clients who want advanced<br />

capabilities. <strong>The</strong>se extensions include:<br />

► Server Plus Pack<br />

► Software Distribution Premium Edition<br />

► Remote Deployment Manager<br />

► Application Workload Manager<br />

Some of the features that you find in <strong>IBM</strong> Director include:<br />

► Self-managing, smart tools that are automated, and proactive capabilities that<br />

reduce IT costs and maximize uptime<br />

► Support for non-<strong>IBM</strong> hardware<br />

Innovative use of industry standards from Common Information Model (CIM)<br />

to Simple Network Mail Protocol (SNMP) enables heterogeneous hardware<br />

management, protecting your existing IT investment.<br />

► Seamless Integration<br />

<strong>IBM</strong> Director protects your investments in other management packages,<br />

complementing them with more extensive hardware manageability.<br />

► Single-click management GUI<br />

A convenient user interface delivers the ability to drag and drop tasks to<br />

specific systems or groups of systems.<br />

140 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


► Integrated, centralized SQL database<br />

An internal database makes system-related data available, even when the<br />

specific system is not directly available.<br />

► Multiple operating system support<br />

<strong>IBM</strong> Director smoothly handles a variety of popular operating systems.<br />

► Comprehensive <strong>BladeCenter</strong> support<br />

<strong>IBM</strong> Director provides an easy, single point of deployment and management<br />

of new blade server architectures.<br />

► Consistent framework that can be extended with plug-ins for advanced<br />

management<br />

► CLI support for many <strong>IBM</strong> Director management tasks<br />

<strong>IBM</strong> Director 4.20 has many new features, support for new operating systems,<br />

and enhancements to existing functions. Together, these changes provide<br />

significantly richer and broader capabilities for superior hardware management.<br />

Key enhancements include:<br />

► Expanded Linux support including aggregated, color-coded systems health<br />

status for systems running Linux<br />

► Convenient subscription service included with the <strong>IBM</strong> Director license to<br />

proactively deliver the latest product updates<br />

► Enhanced optional Server Plus Pack, a collection of advanced tools to help<br />

optimize server performance and availability<br />

Some of the new features include XML report format for the Capacity<br />

Manager and System Availability tools, the ability to schedule System<br />

Availability reports, and Active PCI Manager support for Linux.<br />

► <strong>The</strong> latest hardware support including support for additional hardware<br />

industry standards for manageability, ASF 2.0 and IPMI, to further facilitate<br />

the management of heterogeneous hardware environments<br />

► A multitude of new features and refinements to make <strong>IBM</strong> Director even<br />

easier to install and use<br />

Chapter 8. System management scenarios 141


8.1.1 Setting up an <strong>IBM</strong> Director Server<br />

Before Director clients can be installed, you must first configure the Director<br />

server. <strong>The</strong> following instructions explain how we set up our server.<br />

1. Mount the Director CD.<br />

sysmgr:/ # mount -t iso9660 -o map=off /dev/cdrom /media/cdrom<br />

Important: It is critical that the CD is mounted in this manner. If you do not<br />

specify -o map=off, the capitalization of RPM names is not preserved and<br />

the installation script fails.<br />

2. With the CD mounted, copy the contents of the appropriate directory to the<br />

hard drive of the server.<br />

sysmgr:/root/Director-Source/ # cp<br />

/media/cdrom/director/server/linux/i386/* .<br />

3. With the sources copied to the hard drive, launch the installer.<br />

sysmgr:/root/Director-Source/ # ./dirinstall<br />

4. After the Director code is installed, configure a database for it to use. Because<br />

PostgreSQL ships as a standard part of SLES, we chose to use it.<br />

Example 8-1 shows how we configured it.<br />

Example 8-1 Configuring PostgreSQL<br />

sysmgr:/root/Director-Source/ # cd /usr/share/pgsql<br />

sysmgr:/usr/share/pgsql/ # ln -s jdbc7.1-1.2.jar postgresql.jar<br />

sysmgr:/usr/share/pgsql/ # cd /etc/TWGserver/<br />

sysmgr:/etc/TWGServer/ # echo export CLASSPATH=/usr/share/pgsql/psotgresql.jar<br />

> setup_env<br />

sysmgr:/etc/TWGServer/ # chmod u+x setup_env<br />

Important: We assume that PostgreSQL is configured and running<br />

properly on the system. Refer to the documentation regarding how to do<br />

this.<br />

5. By default, PostgreSQL on SLES does not bind to a TCP/IP socket. To enable<br />

this, modify /etc/sysconfig/postgresql and add the following line:<br />

POSTGRES_OPTIONS="-i"<br />

6. Begin the database configuration, as shown in Example 8-2.<br />

Example 8-2 Launching the database configuration<br />

sysmgr:/etc/TWGServer/ # passwd postgres<br />

Changing password for postgres<br />

142 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


New password:<br />

Re-enter new password:<br />

Password changed<br />

sysmgr:/etc/TWGServer/ # cd /opt/<strong>IBM</strong>/director/bin/<br />

sysmgr:/opt/<strong>IBM</strong>/director/bin # ./cfgdb<br />

7. <strong>The</strong> installer launches. In the first window (Figure 8-1), select PostgreSQL.<br />

Figure 8-1 Selecting the type of database<br />

Chapter 8. System management scenarios 143


8. Enter the database connection information, as shown in Figure 8-2. Click<br />

Next to complete the configuration of the database.<br />

Figure 8-2 Database connection information<br />

144 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


9. We decided to use encrypted communication for our <strong>IBM</strong> Director<br />

environment. To do this, execute cfgsecurity as shown in Figure 8-3.<br />

sysmgr:/opt/<strong>IBM</strong>/director/bin # cfgsecurity<br />

Do you want <strong>IBM</strong> Director to encrypt data communications on this system?<br />

0 - No<br />

1 - Yes<br />

Enter number corresponding to desired security setting (0 is default):<br />

1<br />

Which encryption algorithm do you want to use?<br />

1 - DES encryption (56-bit key)<br />

2 - Triple DES encryption (128-bit key)<br />

2<br />

0 - No encryption<br />

Encryption set or unset successfully.<br />

All <strong>IBM</strong> Director servers and encrypted systems are set to "secured"<br />

automatically.<br />

Setting secured or unsecured system successful.<br />

Figure 8-3 Setting Director server encryption<br />

10.After you configure the database and security, start the Director daemon.<br />

sysmgr:/opt/<strong>IBM</strong>/director/bin # ./twgstart<br />

With the daemon started, the Director server configuration is now complete.<br />

8.1.2 Installing an <strong>IBM</strong> Director Agent<br />

Follow these steps to install an <strong>IBM</strong> Director Agent.<br />

1. Mount the Director CD and copy the files as explained in steps 1 and 2 on<br />

page 142. This time, substitute /media/cdrom/director/server/linux/i386/ with<br />

/media/cdrom/director/agent/linux/ppc/ in step 2.<br />

Chapter 8. System management scenarios 145


2. With the sources copied to the hard drive, launch the installer, as shown in<br />

Figure 8-4.<br />

blade:/root/Director-Source/ # ./dirinstall<br />

Attempting to install ITDAgent-4.12-1.ppc.rpm<br />

Preparing...<br />

#################################################<br />

ITDAgent<br />

#################################################<br />

To start the <strong>IBM</strong> Director Agent manually, run /opt/ibm/director/bin/twgstart<br />

Installation of selected components is successful.<br />

To start the <strong>IBM</strong> Director Agent manually, run /opt/ibm/director/bin/twgstart<br />

Figure 8-4 Installing the Director Agent<br />

3. After the installation is complete, configure security as desired. We chose to<br />

use encryption, as shown in Figure 8-5.<br />

blade:/etc/TWGServer/ # cd /opt/<strong>IBM</strong>/director/bin/<br />

blade:/opt/<strong>IBM</strong>/director/bin/ # cfgsecurity<br />

Do you want <strong>IBM</strong> Director to encrypt data communications on this system?<br />

0 - No<br />

1 - Yes<br />

Enter number corresponding to desired security setting (0 is default):<br />

1<br />

Encryption set or unset successfully.<br />

All <strong>IBM</strong> Director servers and encrypted systems are set to "secured"<br />

automatically.<br />

Setting secured or unsecured system successful.<br />

Figure 8-5 Configuring Director Agent security<br />

4. With the security set, start the Director Agent.<br />

blade:/opt/<strong>IBM</strong>/director/bin # ./twgstart<br />

146 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


8.1.3 Using Director to manage <strong>JS20</strong>s<br />

With the agent installed, you can add the <strong>BladeCenter</strong> <strong>JS20</strong> to the Director<br />

console as another managed server. For details about how to do this, see the<br />

<strong>IBM</strong> Director 4.12 Installation and Configuration Guide, SC09-P290-50.<br />

<strong>The</strong> following subset of Director management features are available for the<br />

<strong>BladeCenter</strong> <strong>JS20</strong>:<br />

► Software Distribution Standard Edition<br />

► Software Distribution Premium Edition<br />

► Update Assistant<br />

► Inventory Hardware<br />

► Inventory Software<br />

► Event Action PLans<br />

► Hardware Health Status<br />

► Resource Monitors<br />

► Eventlog<br />

► Agent Discovery<br />

► Process Management<br />

► SNMP Alerts<br />

► SNMP Browser<br />

► Rack Manager<br />

► Remote Session<br />

► File Transfer<br />

8.2 <strong>IBM</strong> Cluster Systems Management<br />

<strong>BladeCenter</strong> <strong>JS20</strong> installations may involve clusters of many blade servers. Many<br />

of these installations use <strong>IBM</strong> Cluster Systems Management to both install and<br />

manage <strong>BladeCenter</strong> <strong>JS20</strong>s.<br />

CSM for Linux has been under development since early 2000 and was first<br />

released as a product in June 2001. <strong>The</strong> original requirement for this product was<br />

to provide common cluster management for Linux and AIX, leveraging <strong>IBM</strong>’s<br />

existing cluster software portfolio to work with Open Source packages and best<br />

practices from the Linux cluster market.<br />

CSM provides basic cluster management functions, such as hardware control<br />

and operating system and software installation. It also provides significant<br />

value-added functions such as configuration file management, a robust<br />

monitoring infrastructure, automated event management, diagnostic probes, and<br />

common management of AIX and Linux clusters.<br />

Chapter 8. System management scenarios 147


As CSM was being developed, the Extreme Cluster Administration Toolkit<br />

(xCAT), a script-based package, was developed by <strong>IBM</strong>’s advanced technical<br />

sales support team. It was provided to clients who purchase Linux clusters based<br />

on <strong>IBM</strong> servers. It was also used by <strong>IBM</strong> Global Services to address their need<br />

for tools to deploy and manage Linux clusters. In the interim, xCAT became a<br />

proving ground for new concepts and practices centered on the management of<br />

Linux clusters. It can be a model for collaborative development between <strong>IBM</strong> and<br />

our clients.<br />

At the time of the development of this book, CSM was available in its fourth<br />

release (V1.4.0.4) with expanded scalability and hardware support. Many xCAT<br />

functions have been integrated into CSM, along with utilities to automate the<br />

process of migrating from xCAT to CSM. Through the use of CSM, clients who<br />

have implemented xCAT can now have many of these functions with full product<br />

support from <strong>IBM</strong>.<br />

<strong>The</strong> key benefits of CSM can help to reduce costs, provide higher availability of<br />

the cluster for productive use, and improve system utilization. Customers who<br />

have existing AIX 5L operating system-based cluster systems can leverage those<br />

skills to manage their Linux clusters.<br />

System administrators can automate repetitive installation and configuration<br />

tasks and automate problem determination and recovery. <strong>The</strong>y can monitor and<br />

report health information and resource utilization, and automatically recover from<br />

node, storage, or network failures. This can lead to overall simplification of cluster<br />

administration.<br />

This section describes a fast-path CSM implementation to install and manage<br />

<strong>BladeCenter</strong> <strong>JS20</strong>s. For a more comprehensive view, refer to the CSM for Linux<br />

Planning and Installation Guide, SA22-7853.<br />

Our implementation of CSM assumes the type of network configuration<br />

described in 4.1.1, “Minimal network requirements” on page 54. Review this<br />

along with other CSM planning considerations described in 4.3.2, “<strong>IBM</strong> Cluster<br />

Systems Management” on page 64.<br />

8.2.1 Setting up a CSM management node<br />

<strong>The</strong> first step in installing CSM is to set up a management node. We focus on<br />

how to set up a management node based on an xSeries server. We believe this<br />

is a common scenario for users of <strong>BladeCenter</strong> <strong>JS20</strong>s, particularly if they also<br />

have other blade servers that use Intel central processing units (CPUs).<br />

Using an xSeries server to install and manage <strong>BladeCenter</strong> <strong>JS20</strong>s cluster nodes<br />

introduces some complications that are not required when the management<br />

148 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


node is the same architecture as the compute nodes. We focus our discussion<br />

around these issues.<br />

We use CSM to both install and manage <strong>BladeCenter</strong> <strong>JS20</strong>s. <strong>The</strong>refore, we must<br />

use a compatible operating system on both the management node and the<br />

<strong>BladeCenter</strong> <strong>JS20</strong> cluster nodes. <strong>The</strong> steps we require to set up the CSM<br />

management node are:<br />

1. Install a supported operating system.<br />

2. Configure network interfaces and set up host name resolution.<br />

3. Obtain the latest CSM software and prerequisites.<br />

4. Install the CSM base RPM.<br />

5. Run the CSM installation program.<br />

6. Copy the operating system, service packs, CSM code, and prerequisites for<br />

<strong>BladeCenter</strong> <strong>JS20</strong> compute nodes.<br />

7. Accept the CSM license.<br />

8. Set up hardware control point passwords.<br />

9. Verify the installation.<br />

We review each of these steps in the sections that follow. For our tests, we<br />

performed a standard installation of SUSE LINUX Enterprise Server (SLES) 9 for<br />

x86 on our management node directly from the distribution CDs.<br />

<strong>The</strong> management node network configuration<br />

Our implementation of CSM assumes the type of network configuration<br />

described in 4.1.1, “Minimal network requirements” on page 54.<br />

<strong>The</strong> CSM management node is connected to both the hardware management<br />

subnet and the operating system management subnet using separate network<br />

interfaces. You can use additional network interfaces to connect the management<br />

node to other networks if required.<br />

In our scenario, we used the IP address ranges 9.3.5/24 for the hardware<br />

management subnet and 192.168.1/24 for the operating system management<br />

subnet. <strong>The</strong>re should be no other active DHCP servers on the operating system<br />

management subnet, because the CSM management node performs the DHCP<br />

server function on that subnet.<br />

Define host names and allocate IP addresses for the following interfaces before<br />

you attempt to use CSM:<br />

► <strong>The</strong> management node network interface that connects to the operating<br />

system management subnet<br />

Chapter 8. System management scenarios 149


This is the interface that will be used to install the operating system on the<br />

<strong>BladeCenter</strong> <strong>JS20</strong> cluster nodes.<br />

► <strong>The</strong> network interface on each <strong>BladeCenter</strong> <strong>JS20</strong> that connects to the<br />

operating system management subnet<br />

► <strong>The</strong> external network interface of each <strong>BladeCenter</strong> Management Module<br />

You can either set up a hosts file on the management node or use a Domain<br />

Name System (DNS) server to define the host name to IP address mapping. In<br />

our scenario, we used a hosts file as shown in Example 8-3.<br />

Example 8-3 /etc/hosts on management node<br />

127.0.0.1 localhost<br />

192.168.1.20 mgr<br />

192.168.1.117 blade1<br />

192.168.1.118 blade2<br />

192.168.1.119 blade3<br />

9.3.5.190 bcmm<br />

Obtaining the latest CSM software and prerequisites<br />

You can obtain the latest CSM software from:<br />

http://techsupport.services.ibm.com/server/cluster/fixes/csmfixhome.html<br />

Check this Web site for the latest CSM software even if you have CSM available<br />

on CD, because new CSM software releases occur every few months.<br />

<strong>The</strong> CSM software available at the previous Web site is under a 60-day Try and<br />

Buy license. If you are performing a permanent CSM installation, you need to<br />

have purchased an appropriate quantity of CSM licenses. You also need the full<br />

license file that is contained on the CSM product CD.<br />

We downloaded two different versions of CSM: one for the management node<br />

(xSeries based) and one for the <strong>BladeCenter</strong> <strong>JS20</strong> cluster nodes. <strong>The</strong> two<br />

versions of CSM that we downloaded were:<br />

► CSM for Linux on xSeries V1.4.0.4<br />

► CSM for Linux on pSeries V1.4.0.4<br />

Follow these steps:<br />

1. Type the following command:<br />

mkdir /tmp/csm_i386<br />

2. Now type this command:<br />

mkdir /tmp/csm_js20<br />

150 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


3. Download the two CSM tarballs into their respective directories created in<br />

steps 1 and 2.<br />

4. Change to /tmp/csm_i386 and run the following command:<br />

tar -xvzf csm-linux-1.4.0.0.i386.tar.gz<br />

5. Change to /tmp/csm_js20 and run the following command:<br />

tar -xvzf csm-linux-1.4.0.ppc64.tar.gz<br />

6. We created a directory for such prerequisite packages as Autoupdate called<br />

/tmp/csmpreq. Autoupdate is a package that is used by CSM. We<br />

downloaded Version 5.3.11-1 from the following URL and placed it into the<br />

/tmp/csmpreq directory.<br />

http://ftp.mat.univie.ac.at/pub/teschl/autoupdate/index.html<br />

Installing core CSM RPM<br />

You must be logged into the management node as root to perform this step.<br />

We installed the core CSM RPM on the management node using this command:<br />

# rpm -ivh /tmp/csm_i386/csm.core*.i386.rpm<br />

When you run the rpm command and install the CSM core, CSM sets the $PATH<br />

and $MANPATH environment variables. <strong>The</strong>refore, you must log out and then log<br />

in again to the management node after performing the installation to accept the<br />

changes to the environment variables that are required when performing<br />

subsequent installation steps.<br />

Running CSM installation program<br />

After logging back into the management node as root, we ran the CSM<br />

installation program using the following command:<br />

# installms -p /tmp/csm_i386:/tmp/csmpreq<br />

Note: <strong>The</strong> installms command is found in the /opt/csm/bin directory after you<br />

install the CMS core.<br />

<strong>The</strong> -p option of the installms command allows you to specify the location where<br />

the CSM installation program should search for the files it needs. We specified<br />

the two directories where we placed the CSM software and prerequisites for the<br />

management node hardware architecture that we were using.<br />

<strong>The</strong> installms command prompts you to insert operating system CDs if it cannot<br />

find the operating system packages that it needs in one of the directories listed<br />

after the -p option. You are also prompted to replace the Trivial File Transfer<br />

Protocol (TFTP) daemon provided with the Linux distribution with a replacement<br />

Chapter 8. System management scenarios 151


version packaged with CSM. And you are prompted to accept the licenses<br />

associated with some software packages distributed with CSM.<br />

Make sure you resolve any errors reported by the installms command before<br />

you continue with the next step.<br />

Copying the code required for <strong>BladeCenter</strong> <strong>JS20</strong>s<br />

<strong>The</strong> <strong>BladeCenter</strong> <strong>JS20</strong> has a different hardware architecture than the xSeries<br />

server that we used for our management node. <strong>The</strong>refore, you must copy the<br />

code required for the <strong>BladeCenter</strong> <strong>JS20</strong>s to the management node. This<br />

includes:<br />

► SUSE LINUX Enterprise Server 9 for <strong>IBM</strong> POWER<br />

► CSM for Linux on pSeries<br />

► CSM prerequisites<br />

We used the CSM copycds and copycsmpkgs commands to copy the operating<br />

system, service pack, CSM software, and prerequisites for the <strong>BladeCenter</strong> <strong>JS20</strong><br />

cluster nodes to the management node. Both of these commands require you to<br />

specify any attributes that are different to the operating system environment on<br />

the management node. In our scenario, we specified the InstallPkgArchitecture<br />

attribute, because the blade server cluster nodes have a different architecture<br />

than the management node.<br />

We copied the operating system CDs using the following command:<br />

# copycds InstallPkgArchitecture=ppc64<br />

We copied the CSM software and prerequisites for the <strong>BladeCenter</strong> <strong>JS20</strong> cluster<br />

nodes using this command:<br />

# copycsmpkgs -p /tmp/csm_js20:/tmp/csmpreq InstallPkgArchitecture=ppc64<br />

<strong>The</strong> copycsmpkgs command prompts you to insert operating system CDs if it<br />

cannot find the operating system packages that it needs in one of the directories<br />

listed after the -p option.<br />

Accepting the CSM licence<br />

Accept the CSM licence. If you want to use a 60-day trial licence, use the<br />

following command:<br />

# csmconfig -L<br />

If you want to accept the permanent license, mount the CSM product CD and run<br />

the following command:<br />

# csmconfig -L /media/cdrom/csmlum.full<br />

152 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Setting up hardware control point passwords<br />

CSM requires user IDs and passwords to communicate with <strong>BladeCenter</strong><br />

Management Modules so that it can:<br />

► Control power to <strong>BladeCenter</strong> <strong>JS20</strong>s<br />

► Provide remote console access to <strong>BladeCenter</strong> <strong>JS20</strong>s<br />

► Retrieve the MAC addresses for the Gigabit Ethernet interfaces on<br />

<strong>BladeCenter</strong> <strong>JS20</strong>s<br />

CSM lets you define one user ID and password for power control and a different<br />

user ID and password for remote console access for each <strong>BladeCenter</strong><br />

Management Module. Both of these profiles must match login profiles that are<br />

defined in the corresponding <strong>BladeCenter</strong> Management Module. You can<br />

configure these profiles through the <strong>BladeCenter</strong> Management Module Web<br />

interface.<br />

In a production environment, you should define two different login profiles, one<br />

for power control and the other for remote console access. You should also limit<br />

the access for each of these profiles to the minimum needed. In our test<br />

environment, we used the same profile for both power control and remote<br />

console access.<br />

We recommend that you use the same user ID and password for power control<br />

for all <strong>BladeCenter</strong> Management Modules in a cluster and define these as the<br />

default user ID and password for power control in CSM.<br />

You define a default <strong>BladeCenter</strong> Management Module user ID and password for<br />

power control using the following command:<br />

# systemid -p blade USERID<br />

<strong>The</strong> systemid command prompts you to enter the password associated with the<br />

default user ID you specified.<br />

Similarly, we recommend that you use the same user ID and password for remote<br />

console access for all <strong>BladeCenter</strong> Management Modules in a cluster and define<br />

these as the default user ID and password for remote console access in CSM.<br />

You define a default <strong>BladeCenter</strong> Management Module user ID and password for<br />

remote console access using the following command:<br />

# systemid -c -p blade USERID<br />

<strong>The</strong> systemid command prompts you to enter the password associated with the<br />

default user ID you specified. This step is necessary for the Serial over LAN<br />

(SoL) access.<br />

Chapter 8. System management scenarios 153


You can verify what user IDs you configured as the default for <strong>BladeCenter</strong> power<br />

control and remote console access by issuing the systemid command with no<br />

parameters. This displays output similar to what is shown in Example 8-4.<br />

Example 8-4 Output from systemid command<br />

blade_#DEFAULT# USERID<br />

blade_#DEFAULT#_console USERID<br />

Verifying the installation<br />

CSM is provided with an installation verification probe that can pick up many<br />

installation errors. You can run this probe by using the following command:<br />

# probemgr -p ibm.csm.ms -l 0<br />

Review the output of this command and investigate any error messages.<br />

8.2.2 Installing and managing <strong>BladeCenter</strong> <strong>JS20</strong>s using CSM<br />

After you establish a CSM management node, you can install and manage<br />

<strong>BladeCenter</strong> <strong>JS20</strong>s.<br />

<strong>The</strong> required steps are:<br />

1. Define your <strong>BladeCenter</strong> <strong>JS20</strong> cluster nodes to CSM.<br />

2. Set up the installation environment.<br />

3. Start installation and wait for completion.<br />

We examine each of these steps in more detail in the sections that follow.<br />

Defining the <strong>BladeCenter</strong> <strong>JS20</strong> cluster nodes<br />

We find that the easiest way to define <strong>BladeCenter</strong> <strong>JS20</strong> cluster nodes to CSM is<br />

to use a node definition file. <strong>The</strong>re are many other ways to define nodes in CSM.<br />

Review the CSM for Linux Planning and Installation Guide, SA22-7853, to<br />

determine the most appropriate method for your environment.<br />

<strong>The</strong> node definition file that we used for our small cluster of <strong>BladeCenter</strong> <strong>JS20</strong>s is<br />

shown in Example 8-5.<br />

Example 8-5 CSM node definition file<br />

default:<br />

ManagementServer=mgr<br />

PowerMethod=blade<br />

HWControlPoint=bcmm<br />

ConsoleMethod=blade<br />

Consol<strong>eServer</strong>Name=bcmm<br />

154 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


lade1:<br />

blade2:<br />

blade3:<br />

InstallDiskType=ide<br />

InstallPkgArchitecture=ppc64<br />

InstallAdapterName=eth1<br />

HWControlNodeId=blade1<br />

ConsolePortNum=1<br />

HWControlNodeId=blade2<br />

ConsolePortNum=2<br />

HWControlNodeId=blade3<br />

ConsolePortNum=3<br />

<strong>The</strong> default stanza of the node definition file contains the node attributes that are<br />

common to all the <strong>BladeCenter</strong> <strong>JS20</strong> cluster nodes in our cluster. You can have<br />

multiple default stanzas in a node definition file, with the new defaults applying to<br />

those stanzas listed in the node definition file after the new default stanza.<br />

<strong>The</strong> remaining stanzas define each <strong>BladeCenter</strong> <strong>JS20</strong> cluster node, and contain<br />

only those node attributes that are unique to that cluster node. <strong>The</strong> name of the<br />

stanza for each <strong>BladeCenter</strong> <strong>JS20</strong> cluster node should match the host name that<br />

you assigned to that cluster node in your hosts file or DNS.<br />

Use the following guide when you specify attribute values:<br />

► ManagementServer: Host name of management node network interface<br />

connected to the operating system management subnet<br />

► HWControlPoint: Host name of the <strong>BladeCenter</strong> Management Module<br />

► Consol<strong>eServer</strong>Name: Host name of the <strong>BladeCenter</strong> Management Module<br />

► InstallAdapterName: Set to eth1 if you are using the network configuration<br />

described in 4.1.1, “Minimal network requirements” on page 54<br />

This network configuration performs operating system network installation via<br />

eth1 to prevent issues with the SoL remote text console function, which uses<br />

eth0.<br />

► HWControlNodeId: Same name you assigned the <strong>BladeCenter</strong> <strong>JS20</strong> cluster<br />

node when you set up the hardware<br />

We recommend that you make it the same as the host name. We demonstrate<br />

how to configure this in 5.3, “Blade server configuration” on page 76.<br />

► ConsolePortNum: Bay in the <strong>BladeCenter</strong> chassis where the <strong>BladeCenter</strong><br />

<strong>JS20</strong> cluster node is installed<br />

Chapter 8. System management scenarios 155


► PowerMethod: Should always be blade for <strong>BladeCenter</strong> <strong>JS20</strong> cluster nodes<br />

► ConsoleMethod: Should always be blade for <strong>BladeCenter</strong> <strong>JS20</strong> cluster<br />

nodes<br />

After you create a node definition file for your cluster, you can define the nodes to<br />

CSM using the following command:<br />

# definenode -f /tmp/nodedefs<br />

After the definenode command runs, the management server is set up with all<br />

the node information for CSM, and you are ready to verify the node definitions.<br />

Since the actual node installation has not happened yet, you can make changes<br />

to any of the node definitions at this time. To determine whether the nodes have<br />

been defined, enter the lsnode command from the management server.<br />

After you run the lsnode command, the system responds with a line for each<br />

node that was successfully defined. If a node is not defined, it does not appear in<br />

the output for the lsnode command.<br />

To display all the information about each node, use the lsnode command from<br />

the management server with the –l (lowercase L) option, such as:<br />

# lsnode -l<br />

<strong>The</strong> system responds with a list (output) that contains the extended information<br />

for each node that was successfully defined. If a node is not defined, it does not<br />

appear in the output for the lsnode command.<br />

Some of the attributes for a node may have null values at this point. If a node is<br />

not defined, it is not installed. If you must correct the information, you can change<br />

the attributes of the nodes either by running the chnode command, or by<br />

rerunning the definenode command with the -m (modify) flag. <strong>The</strong> definenode -m<br />

command accepts a new nodedef file or a new command line and only modifies<br />

nodes that have changed attributes. To change attribute values of a node, enter<br />

the chnode command from the management server:<br />

chnode hostname attr=value<br />

Setting up the installation environment<br />

After you successfully define the node, configure how the node is to be installed<br />

in the cluster. For Red Hat AS or Red Hat EL 3 installations, use Kickstart, and<br />

run the csmsetupks command. For SUSE LINUX Enterprise Server installations,<br />

use AutoYaST, and run the csmsetupyast command. <strong>The</strong> csmsetupks and<br />

csmsetupyast commands configure the install process, collect MAC addresses,<br />

and configure the network boot of the nodes.<br />

156 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


In our lab environment using SuSE, the csmsetupyast did several things,<br />

including:<br />

► Retrieved the MAC addresses for the Gigabit Ethernet interfaces of each<br />

<strong>BladeCenter</strong> <strong>JS20</strong> cluster node<br />

► Built a Autoyast configuration file for each <strong>BladeCenter</strong> <strong>JS20</strong> cluster node<br />

from a template<br />

► Built the DHCP server configuration file /etc/dhcpd.conf<br />

Before you attempt to use csmsetupyast for the first time, we recommend that<br />

you verify that power control and remote console functionality is working<br />

correctly. You can verify that power control is working using the following<br />

command:<br />

# rpower -a query<br />

This command displays the current power status of all nodes in the cluster. If this<br />

is not successful, determine why before you proceed any further.<br />

You can verify that the remote consoles are working by using the following<br />

command:<br />

# rconsole -a<br />

This command opens a separate console window for every cluster node that is<br />

defined to CSM. If a <strong>BladeCenter</strong> <strong>JS20</strong> cluster node is not powered on, you see<br />

messages indicating that the SoL is not ready. Close each remote console<br />

window using the key sequence Ctrl+E c . before you continue further.<br />

When you are satisfied that power control and remote console functionality is<br />

working correctly, verify that the DHCP server on the management node is<br />

configured to listen on the correct network interface.<br />

Under SUSE LINUX Enterprise Server, the interfaces that the DHCP server uses<br />

to listen are configured in the file /etc/sysconfig/dhcpd. Set the<br />

DHCP_INTERFACE variable in this file to include the Ethernet interface on the<br />

management node that is connected to the operating system management<br />

subnet described in “Operating system management subnet” on page 57.<br />

Now set up the installation environment using the CSM command:<br />

# csmsetupyast -x -P<br />

We use the -x option because we already copied the necessary files to support<br />

the network installation of the operating system to the management node. If you<br />

do not specify this option, you are prompted to insert operating system CDs. We<br />

use the -P option to specify all nodes that are in the premanaged state and are<br />

waiting to be installed.<br />

Chapter 8. System management scenarios 157


<strong>The</strong> csmsetupyast command uses an XML template file found in /opt/csm/install<br />

as a default template for the Autoyast configuration files that it creates for each<br />

<strong>BladeCenter</strong> <strong>JS20</strong> cluster node. Depending upon the release of SuSE that you<br />

are using at the time, the file has the format:<br />

yastcfg.InstallDistributionNameInstallDistributionVersion-Arch.xml<br />

For example, in our lab environment, the file was called<br />

yastcfg.SLES9-ppc64.xml.<br />

If you want to use a different template, first create the template and then specify it<br />

on the csmsetupyast command using the -k option. We recommend that you<br />

copy the default template when creating new templates to ensure that you<br />

include all the elements that CSM requires in the template.<br />

Review the output of the csmsetupyast command carefully and resolve any errors<br />

before you proceed further.<br />

Starting the cluster node installation<br />

Start the cluster node installation using the installnode command. We<br />

recommend that you try to install a few nodes at first and make sure it works<br />

before you try to install many the nodes concurrently. To start installation of a<br />

single node, use the command:<br />

installnode -n blade1<br />

<strong>The</strong> installnode command takes a few minutes to complete when used with<br />

<strong>BladeCenter</strong> <strong>JS20</strong> cluster nodes. This is because the installnode command<br />

must wait until the <strong>BladeCenter</strong> <strong>JS20</strong> boot process has reached the point where<br />

it can pass network installation parameters to the firmware.<br />

When the installation is underway, the installnode command exits and displays<br />

the status for each node it was asked to install. You can monitor the status of the<br />

install as it progresses using the monitorinstall command. You can also open a<br />

remote console to watch the installation progress using the rconsole command.<br />

For more information about installing CSM, see CSM for Linux: Planning and<br />

Installation Guide, SA22-7853.<br />

158 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Abbreviations and acronyms<br />

ac alternating current<br />

AIX Advanced Interactive<br />

eXecutive<br />

BIU Bus Interface Unit<br />

BOOTP Bootstrap Protocol<br />

BOS base operating system<br />

BPU branch processing unit<br />

BSMP Blade System Management<br />

Processor<br />

CISC complex instruction set<br />

computer<br />

CLI command line interface<br />

CRLF carriage return/line feed<br />

CSM Cluster Systems<br />

Management<br />

CTR count register<br />

DC direct current<br />

DDR double data rate<br />

DHCP Dynamic Host Configuration<br />

Protocol<br />

DNS Domain Name Server<br />

ECC Error Correction Code<br />

ERAT effective to real address<br />

translation<br />

ESM Ethernet Switch Module<br />

FPR Floating Point Register<br />

FPU Floating Point Unit<br />

FSB Front Side Bus<br />

FTP File Transfer Protocol<br />

FXU Fixed Point Unit<br />

Gb gigabit<br />

GB gigabyte<br />

gcc GNU C Compiler<br />

GPR General Purpose Register<br />

GUI Graphical User Interface<br />

HTTP Hypertext Transfer Protocol<br />

HTTPS Hypertext Transfer<br />

Protocol-Secure<br />

I/O input/output<br />

<strong>IBM</strong> International Business<br />

Machines Corporation<br />

IGESM CISCO Intelligent Gigabit<br />

Ethernet Switch Module<br />

IP Internet Protocol<br />

IPL Initial Program Load (Boot)<br />

ITSO International Technical<br />

Support Organization<br />

KVM keyboard, video and mouse<br />

LAN local area network<br />

LPAR logical partition<br />

LR Link Register<br />

MAC Media Access<br />

Mb megabit<br />

MB megabyte<br />

MFLOPS Million Floating Point<br />

Operations Per Second<br />

MP multiprocessor<br />

NFS Network File System<br />

NIC Network Interface Controller<br />

NIM Network Install Manager<br />

PKT packet<br />

POSIX Portable Operating System for<br />

UNIX<br />

POST Power On Self Test<br />

POWER Performance Optimization<br />

With Enhanced RISC<br />

RAM random access memory<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. 159


RHEL Red Hat Enterprise Linux<br />

RISC Reduced Instruction Set<br />

Computer<br />

ROM Read Only Memory<br />

RTC Real Time Clock<br />

SAN storage area network<br />

SCSI small computer systems<br />

interface<br />

SMP Simultaneous Multi-Processor<br />

SOL Serial over LAN<br />

SPOT Shared Product Object Tree<br />

SSH Secure Shell<br />

TCP/IP Transmission Control<br />

Protocol/Internet Protocol<br />

TFTP Trivial File Transfer Protocol<br />

TLB Translation Lookaside Buffer<br />

USB Universal Serial Bus<br />

UXBC UpdateXpress firmware<br />

update scripts for<br />

<strong>BladeCenter</strong><br />

VLAN virtual local area network<br />

VMX Vector Multimedia eXtensions<br />

VPN Virtual Private Network<br />

VXU Vector Processing Unit<br />

WAN Wide Area Network<br />

XER Fixed Point Exception<br />

Register<br />

XML eXtended Markup Language<br />

YAST Yet Another Setup Tool<br />

(SuSE)<br />

160 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Related publications<br />

<strong>IBM</strong> <strong>Redbooks</strong><br />

<strong>The</strong> publications listed in this section are considered particularly suitable for a<br />

more detailed discussion of the topics covered in this redbook.<br />

For information about ordering these publications, see “How to get <strong>IBM</strong><br />

<strong>Redbooks</strong>” on page 164. Note that some of the documents referenced here may<br />

be available in softcopy only.<br />

► Implementing System Management Solutions using <strong>IBM</strong> Director, SG24-6188<br />

► <strong>The</strong> Cutting Edge: <strong>IBM</strong> Eserver <strong>BladeCenter</strong>, REDP-3581<br />

► <strong>IBM</strong> Eserver <strong>BladeCenter</strong> Systems Management, REDP-3582<br />

► Deploying Lotus Domino on <strong>IBM</strong> Eserver <strong>BladeCenter</strong>, REDP-3584<br />

► Deploying Apache on <strong>IBM</strong> Eserver <strong>BladeCenter</strong>, REDP-3588<br />

► Deploying Samba on <strong>IBM</strong> Eserver <strong>BladeCenter</strong>, REDP-3595<br />

► <strong>IBM</strong> Eserver <strong>BladeCenter</strong> Networking Options, REDP-3660<br />

► <strong>IBM</strong> Eserver <strong>BladeCenter</strong> Layer 2-7 Network Switching, REDP-3755<br />

► <strong>IBM</strong> Eserver <strong>BladeCenter</strong> Systems Management with <strong>IBM</strong> Director V4.1<br />

and Remote Deployment Manager 4.1, REDP-3776<br />

► <strong>IBM</strong> Eserver <strong>BladeCenter</strong> Configuration Tips, TIPS-0454<br />

Other publications<br />

<strong>The</strong>se publications are also relevant as further information sources:<br />

► Peterson and Davie, Computer Networks: A Systems Approach, 3rd Edition,<br />

Morgan Kaufmann, 2003, ISBN 1-55-860832-3<br />

► Stevens, TCP/IP Illustrated, Volume 1: <strong>The</strong> Protocols, Addison-Wesley, 1993,<br />

ISBN 0-20-163346-9<br />

► Tanenbaum, Computer Networks, 4th Edition, Prentice Hall, 2002, ISBN<br />

0-13-066102-3<br />

► Tanenbaum, Modern Operating Systems, 2nd Edition, Prentice Hall, 2001,<br />

ISBN 0-13-031358-0<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. 161


Online resources<br />

► CSM for Linux: Planning and Installation Guide, SA22-7853<br />

http://publib.boulder.ibm.com/epubs/pdf/am7il103.pdf<br />

► <strong>IBM</strong> Cluster Systems Management for Linux Planning and Installation Guide,<br />

SA22-7873<br />

► <strong>IBM</strong> Cluster Systems Management for AIX5L Planning and Installation Guide,<br />

SA22-7919<br />

► <strong>IBM</strong> Director 4.12 Installation and Configuration Guide, SC09- P290-50<br />

► <strong>IBM</strong> Director 4.0 for <strong>BladeCenter</strong> Products: Systems Management Guide,<br />

SC01-R051-40<br />

► CSM for Linux Planning and Installation Guide, SA22-7853<br />

<strong>The</strong>se Web sites and URLs are also relevant as further information sources:<br />

► Installing AIX 5L on the <strong>IBM</strong> Sserver <strong>BladeCenter</strong> <strong>JS20</strong> whitepaper<br />

http://www.ibm.com/servers/aix/whitepapers/aix_js20.pdf<br />

► <strong>IBM</strong> Eserver Support Web site<br />

http://www.ibm.com/pc/support<br />

► Red Hat Linux<br />

http://www.redhat.com<br />

► SUSE LINUX<br />

http://www.suse.com<br />

► Instructions for installing SuSE from CD<br />

http://www.ibm.com/pc/support/site.wss/document.do?sitestyle=ibm&lndocid=<br />

MIGR-54819<br />

► PowerPC 970 Training<br />

http://www.technonics.com/powerpc.htm<br />

► Linux at <strong>IBM</strong><br />

http://www-1.ibm.com/linux<br />

► Linux Kernel Archives<br />

http://www.kernel.org<br />

► AIX Version 5L Operating System<br />

http://www-1.ibm.com/servers/aix<br />

162 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


► <strong>IBM</strong> Microelectronics Division - PowerPC Technical Documentation<br />

http://www.chips.ibm.com<br />

► <strong>IBM</strong> ^ <strong>BladeCenter</strong> Power Module Upgrade Guidelines Technical<br />

Update<br />

http://www.ibm.com/pc/support/site.wss/document.do?lndocid=MIGR-53353<br />

► Myrinet Web site<br />

http://www.myri.com<br />

► <strong>IBM</strong> ^ Cluster 1350 product<br />

http://www.ibm.com/eserver/cluster<br />

► PowerPC Architecture books from the <strong>IBM</strong> DeveloperWorks Web site<br />

http://www.ibm.com/developerworks/eserver/articles/archguide.html<br />

► <strong>IBM</strong> white paper POWER3: Next Generation 64-bit PowerPC Processor<br />

Design<br />

http://www.ibm.com/servers/eserver/pseries/hardware/whitepapers/power3wp.ht<br />

ml<br />

► <strong>IBM</strong> white paper entitled POWER4 System Microarchitecture<br />

http://www.ibm.com/servers/eserver/pseries/hardware/whitepapers/power4.html<br />

► PowerPC Microprocessor Family: AltiVec Technology Programming<br />

Environments Manual<br />

http://www-306.ibm.com/chips/techlib/techlib.nsf/techdocs/FBFA164F824370F98<br />

7256D6A006F424D<br />

► RETAIN tip H181655<br />

http://www-307.ibm.com/pc/support/site.wss/document.do?lndocid=MIGR-55282<br />

► CSM publication site<br />

http://publib.boulder.ibm.com/clresctr/windows/public/clusterbooks.html<br />

► Service and productivity tools for Linux on POWER<br />

http://techsupport.services.ibm.com/server/lopdiags<br />

► <strong>The</strong> Linux Documentation Project - Kernel HOWTO<br />

http://www.tldp.org<br />

► <strong>The</strong> latest CSM software<br />

http://techsupport.services.ibm.com/server/cluster/fixes/csmfixhome.html<br />

► CSM for Linux: Planning and Installation Guide, SA22-7853<br />

http://publib.boulder.ibm.com/epubs/pdf/am7il103.pdf<br />

Related publications 163


How to get <strong>IBM</strong> <strong>Redbooks</strong><br />

Help from <strong>IBM</strong><br />

You can search for, view, or download <strong>Redbooks</strong>, Redpapers, Hints and Tips,<br />

draft publications and Additional materials, as well as order hardcopy <strong>Redbooks</strong><br />

or CD-ROMs, at this Web site:<br />

ibm.com/redbooks<br />

<strong>IBM</strong> Support and downloads<br />

ibm.com/support<br />

<strong>IBM</strong> Global Services<br />

ibm.com/services<br />

164 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


Index<br />

Symbols<br />

/boot/ppc64 109<br />

/etc/dhcpd.conf 110<br />

/etc/exports 103<br />

/tftpboot 111<br />

Numerics<br />

64-bit computing 32, 34, 36–37, 42, 46<br />

A<br />

activating the open firmware interface 94<br />

Aggressive branch prediction 44–45<br />

AIX operating system 48, 54, 57, 60, 62, 64, 79,<br />

84–85<br />

installing 123<br />

installing with NIM 124–138<br />

Altivec 40<br />

AMD 3DNow! 40<br />

Apple Computer 36<br />

ARP function 56<br />

array 39<br />

array processing 39<br />

AS/400 37<br />

AutoYaST 114<br />

B<br />

blade servers 21<br />

assigning names 77<br />

configuration 76–78<br />

setting the boot sequence 78<br />

Blade Systems Management Processor (BSMP)<br />

15, 22, 26<br />

<strong>BladeCenter</strong><br />

advantages<br />

high availability 4<br />

lower cost 4<br />

SAN optimization 4<br />

server consolidation 4<br />

switch technology 4<br />

chassis 8–10<br />

management module 13<br />

HS20 blade server 21<br />

I/O modules 16–21<br />

Management Module 13–16<br />

power module options 10<br />

power options 10–163<br />

power requirements 5<br />

standard features 9<br />

supported bays 8<br />

versus <strong>IBM</strong> rack server 5<br />

<strong>BladeCenter</strong> T 10<br />

boot sequence 78<br />

Bootstrap protocol (BOOTP) 60<br />

Brocade Enterprise SAN Switch Module 17, 19<br />

Brocade Entry SAN Switch Module 17, 20<br />

Bus Interface Unit (BIU) 45<br />

C<br />

C2 security 48<br />

cache structure 44<br />

Cisco Systems Intelligent Gigabit Ethernet Switch<br />

Module 17–18, 76, 82<br />

Command line interface (CLI) 56, 91<br />

env command 92<br />

Complex instruction set computer (CISC) 30<br />

console for <strong>JS20</strong> blades 87<br />

copycds command 152<br />

copycsmpkgs command 152<br />

count register (CTR) 35<br />

CSM<br />

see <strong>IBM</strong> Cluster Systems Management<br />

csmconfig command 152<br />

D<br />

definenode command 156<br />

DHCP (dynamic host configuration protocol) 60–61<br />

dynamic host configuration protocol (DHCP) 54,<br />

60–61, 70, 110, 149<br />

E<br />

effective to real address translation (ERAT) 44<br />

en0 62<br />

en1 62<br />

env command 92<br />

© Copyright <strong>IBM</strong> Corp. 2005. All rights reserved. 165


eth0 62<br />

eth1 62<br />

F<br />

Fibre Channel Expansion Card 28<br />

firmware 79–87<br />

<strong>JS20</strong> blade server 84<br />

LAN switch I/O module 82<br />

Management Module 80<br />

upgrading under Linux 84<br />

vital product data (VPD) 79<br />

fixed-point exception register (XER) 34<br />

floating-point status and control register (FPSCR)<br />

34<br />

G<br />

general purpose registers (GPRs) 34<br />

Gigabit Ethernet Expansion Card 28<br />

GNU C Compiler (gcc) 49<br />

H<br />

high availability 4<br />

HTTP 16<br />

HTTPS 16<br />

I<br />

I/O modules 16–21, 73–75<br />

configuration 73<br />

<strong>IBM</strong> 4-Port Gigabit Ethernet Switch Module 17, 76,<br />

79, 82<br />

<strong>IBM</strong> AS/400 37<br />

<strong>IBM</strong> Cluster Systems Management (CSM) 16,<br />

50–51, 56, 60, 64, 66, 147–158<br />

obtaining latest software and prerequisites 150<br />

setting up 148<br />

<strong>IBM</strong> Director 16, 56, 63–64, 140<br />

agent 64<br />

installing agent 145<br />

<strong>IBM</strong> General Parallel File System (GPFS) 59<br />

<strong>IBM</strong> Microelectronics Division 46<br />

chip documentation 46<br />

<strong>IBM</strong> Microprocessors<br />

POWER2 31<br />

POWER3 37<br />

POWER4 37<br />

POWER5 38<br />

PowerPC 970 22, 29, 38, 42, 46<br />

166 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong><br />

PowerPC 970FX 22, 29, 38, 42, 46<br />

<strong>IBM</strong> microprocessors<br />

PowerPC 601 36<br />

PowerPC 603 36<br />

PowerPC 604 36<br />

<strong>IBM</strong> rack server 5<br />

<strong>IBM</strong> RS/6000 36<br />

installms command 151<br />

installnode command 158<br />

integrated switch technology 4<br />

Integrated Systems Management Processor (ISMP)<br />

85<br />

Intel Multimedia Extensions (MMX) 40<br />

Intel Streaming SIMD Extensions (SSE) 40<br />

J<br />

<strong>JS20</strong> Blade Server<br />

base features 24<br />

ethernet controller 25<br />

I/O expansion option connectors 25<br />

I/O subsystem 24<br />

IDE controller 25<br />

memory subsystem 24<br />

miscellaneous I/O 26<br />

optional features 27<br />

expansion cards 28<br />

hard disk drives 27<br />

memory modules 27<br />

processor subsystem 24<br />

USB controllers 26<br />

K<br />

keyboard, video monitor and mouse (KVM) 14<br />

KVM (keyboard, video monitor and mouse) 14<br />

L<br />

link register (LR) 35<br />

Linux<br />

compiling a Red Hat kernel 108<br />

compiling a SuSe kernel 108<br />

configuring the sources 102<br />

installing 102<br />

Red Hat sources 102<br />

SuSE sources 104<br />

local area network (LAN) 16<br />

lsnode command 156


M<br />

Management Module 13–16, 70–73<br />

configuration 70–73<br />

console commands 93<br />

default IP address 70<br />

default password 72<br />

default userid 71<br />

firmware 80<br />

mkinitrd command 109<br />

mkzimage_cmdline command 117<br />

Motorola 36<br />

Myrinet 16, 20, 22, 28, 58–59<br />

N<br />

Network File System (NFS) 60<br />

network installation 118<br />

network installation planning 60<br />

network interface selection 62<br />

network planning 54<br />

minimal requirements 54<br />

network services 60<br />

Nortel Networks Layer 2-7 Gigabit Ethernet Switch<br />

Module 17–18, 76, 82<br />

O<br />

obtaining latest CSM software and prerequisites<br />

150<br />

Open Firmware 117<br />

Operating Environment Architecture (OEA) 33<br />

operating system support 48<br />

Optical Pass-thru Module 20<br />

P<br />

PCI<br />

see Peripheral Component Interconnect<br />

performing a network installation 118<br />

performing an unattended installation 116<br />

Peripheral Component Interconnect (PCI) 5, 25,<br />

141<br />

POWER architecture<br />

definition 30<br />

POWER2 31<br />

POWER3 37<br />

POWER4 37<br />

POWER5 38<br />

Power On Self Test (POST) 109<br />

power requirements for <strong>BladeCenter</strong> 5<br />

PowerPC architecture<br />

background 32<br />

definition 30<br />

Operating Environment Architecture (OEA) 33<br />

PowerPC 601 36<br />

PowerPC 603 36<br />

PowerPC 604 36<br />

PowerPC 970 22, 29, 38, 42, 46<br />

PowerPC 970FX 22, 29, 38, 42, 46<br />

User Instruction Set Architecture (UISA) 33<br />

Virtual Environment Architecture (VEA) 33<br />

preparing an unattended installation 111<br />

probemgr command 154<br />

R<br />

rack server 5<br />

rconsole command 157<br />

Red Hat Enterprise Linux 3 48, 121<br />

Red Hat’s Kickstart 112<br />

<strong>Redbooks</strong> Web site 164<br />

Contact us xv<br />

reduced instruction set computer (RISC) 30<br />

Register renaming 44<br />

required network services 60<br />

rpm command 151<br />

rpower command 157<br />

RS64 processors 36<br />

rtas_flash kernel module 84<br />

S<br />

segment lookaside buffer (SLB) 44<br />

Serial over LAN (SoL) 14, 50, 57, 62, 76, 87<br />

components 88<br />

configuring 89<br />

using command line interface 91<br />

server consolidation 4<br />

setting up a CSM management node 148<br />

setting up network installation 61<br />

small computer system interface (SCSI) 5<br />

SoL (Serial over LAN) 14<br />

SSH 16, 89, 91<br />

storage area network (SAN) 16<br />

optimization 4<br />

superscalar features 44<br />

SuSE AutoYaST 114<br />

SUSE LINUX Enterprise Server 9 49, 118<br />

symmetrical multiprocessor (SMP) 36<br />

systemid command 153<br />

Index 167


T<br />

translation lookaside buffer (TLB) 44<br />

Trivial File Transfer Protocol (TFTP) 60, 111<br />

U<br />

unattended installation 111, 116<br />

Universal Manageability Initiative 49<br />

update_flash script 85<br />

User Instruction Set Architecture (UISA) 33<br />

V<br />

vector 39<br />

vector length 38<br />

vector register file (VRF) 41<br />

Vector/SIMD Multimedia eXtension (VMX) 38<br />

Virtual Environment Architecture (VEA) 33<br />

Virtual LAN (VLAN) 15, 17–18, 54, 57, 61, 76, 87,<br />

89<br />

vital product data (VPD) 79<br />

VMX extensions to PowerPC architecture 40<br />

VMX status and control register (VSCR) 41<br />

VPD (vital product data) 79<br />

VRP (vector register file) 41<br />

VSCR (VMX status and control register) 41<br />

X<br />

xinetd daemon 111<br />

Y<br />

Yet Another Setup Tool (YaST) 104<br />

Z<br />

zImage.initrd 116<br />

zImage.initrd file 107, 109<br />

building your own 107–108<br />

168 <strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


<strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> <strong>JS20</strong>


<strong>The</strong> <strong>IBM</strong> Eserver<br />

<strong>BladeCenter</strong> <strong>JS20</strong><br />

Learn about <strong>IBM</strong><br />

Eserver<br />

<strong>BladeCenter</strong> <strong>JS20</strong><br />

and blade server<br />

technology<br />

Understand the<br />

PowerPC 970<br />

microprocessor<br />

architecture<br />

Plan for, install, and<br />

configure<br />

<strong>BladeCenter</strong> <strong>JS20</strong><br />

SG24-6342-01 ISBN 073849061X<br />

Back cover<br />

Blade servers are a relatively new technology. <strong>The</strong>y have<br />

captured industry focus because of their modular design,<br />

which can reduce cost with a more efficient use of valuable<br />

floor space. <strong>The</strong>y offer simplified management, which can<br />

help to speed such tasks as installing, reprovisioning,<br />

updating, and troubleshooting hundreds of blade servers. You<br />

can do all of this remotely using one graphical console with<br />

<strong>IBM</strong> Director systems management tools.<br />

In addition, blade servers provide improved performance by<br />

doubling current rack density. By integrating resources and<br />

sharing key components, costs decrease and availability<br />

increases.<br />

<strong>The</strong> <strong>IBM</strong> Eserver <strong>BladeCenter</strong> boasts innovative modular<br />

technology, leadership density, and availability. It was<br />

designed to help solve a multitude of real-world problems.<br />

This <strong>IBM</strong> Redbook takes an in-depth look at the <strong>IBM</strong><br />

Eserver <strong>BladeCenter</strong> <strong>JS20</strong>. This is a two-way blade server<br />

for applications requiring 64-bit computing. It is ideal for<br />

computer-intensive applications and transactional Internet<br />

servers. This <strong>IBM</strong> Redbook helps you to install, tailor, and<br />

configure the <strong>IBM</strong> Eserver® <strong>BladeCenter</strong> <strong>JS20</strong>.<br />

INTERNATIONAL<br />

TECHNICAL<br />

SUPPORT<br />

ORGANIZATION<br />

®<br />

BUILDING TECHNICAL<br />

INFORMATION BASED ON<br />

PRACTICAL EXPERIENCE<br />

<strong>IBM</strong> <strong>Redbooks</strong> are developed by<br />

the <strong>IBM</strong> International Technical<br />

Support Organization. Experts<br />

from <strong>IBM</strong>, Customers and<br />

Partners from around the world<br />

create timely technical<br />

information based on realistic<br />

scenarios. Specific<br />

recommendations are provided<br />

to help you implement IT<br />

solutions more effectively in<br />

your environment.<br />

For more information:<br />

ibm.com/redbooks

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!