SAS 9.1.3 Intelligence Platform: System Administration Guide

110 Concurrent Queries/Clients R Chapter 14
metadata is loaded into memory, and no I/O for Column metadata queries should
happen again until the server is paused and resumed or stopped and restarted.
Concurrent Queries/Clients
The number of processors on the server determines the number of concurrent
queries that can run on the server.
If you expect a small volume of concurrent clients, you can probably use a
single-processor machine. If you are expecting 1,000 concurrent clients, you will want
an eight-processor machine.
When configuring the metadata server, set the MAXACTIVETHREADS option in the
omaconfig.xml server configuration file to the number of processors. Also be sure to
set the THREADSMIN (TMIN) and THREADSMAX (TMAX) OBJECTSERVERPARMS
options in the metadata server start-up command such that TMIN=TMAX and
TMAX=(#_processors * 2) + 1. For more information about how the metadata server
handles concurrent queries and threads, see “Configuring the Number of Threads Used
by the Metadata Server” on page 103.
Estimating Metadata Storage Requirements
The metadata storage requirement is largely fixed and can be roughly estimated by
using the following formula:
50MB + size_of_all_repositories
In this formula, 50MB is the approximate amount of memory that the server uses to
register the first repository, and size_of_all_repositories is roughly equal to the
size of all of the metadata type containers in all repositories on the metadata server.
A metadata type container is a SAS data set that stores the attributes of metadata
object instances as observations. A metadata type container is created when the first
object instance of that type is created in a repository. Therefore, the amount of required
RAM increases as containers are added, until each repository on the server has
containers for all of the metadata types in the SAS Metadata Model.
The server also creates several containers in each repository that do not map to a
metadata type. For example:
3 an MDASSOC container stores information about all metadata object associations
3 a TXTPAGE stores overflow information from metadata type containers
3 an MRRGSTRY container stores information about the object IDs that are used in
the repository
3 a CNTAINER container maintains a list of other containers that have been created
The memory requirement for stored metadata is also affected by:
3 the number of observations created in each container
3 the value of the PHYSICAL_DELETE server configuration option. When the
PHYSICAL_DELETE option is on (the default), then the memory used by an
observation will be recovered when a metadata object instance is deleted. When
PHYSICAL_DELETE is turned off, then an observation for every metadata object
ever created will be held in memory until the deleted records are purged from the
repository.
A downloadable program called ContainerStatsExport.sas measures the size of
the containers in a repository. The program also reports statistics about the containers
that can be used to manage memory, such as the observation length, and the number of
logical observations versus the number of physical observations. If you create a