Scientific and Technical Aerospace Reports Volume 38 July 28, 2000

More documents

Recommendations

Info

tency in shared-memory multiprocessor systems. Design of a parallel, discrete-event simulation of this monitor is in progress. A design of a tracing tool that facilitates the reconciliation of constraint violations is near completion. Author Complex Systems; Computer Systems Performance; Design Analysis; Expert Systems; Software Engineering 20000064626 NASA Ames Research Center, Moffett Field, CA USA Use Computer-Aided Tools to Parallelize Large CFD Applications Jin, H., NASA Ames Research Center, USA; Frumkin, M., NASA Ames Research Center, USA; Yan, J., NASA Ames Research Center, USA; February 2000; In English; See also 20000064579; No Copyright; Abstract Only; Available from CASI only as part of the entire parent document Porting applications to high performance parallel computers is always a challenging task. It is time consuming and costly. With rapid progressing in hardware architectures and increasing complexity of real applications in recent years, the problem becomes even more sever. Today, scalability and high performance are mostly involving handwritten parallel programs using message-passing libraries (e.g. MPI). However, this process is very difficult and often error-prone. The recent reemergence of shared memory parallel (SMP) architectures, such as the cache coherent Non-Uniform Memory Access (ccNUMA) architecture used in the SGI Origin 2000, show good prospects for scaling beyond hundreds of processors. Programming on an SMP is simplified by working in a globally accessible address space. The user can supply compiler directives, such as OpenMP, to parallelize the code. As an industry standard for portable implementation of parallel programs for SMPs, OpenMP is a set of compiler directives and callable runtime library routines that extend FORTRAN, C and C++ to express shared memory parallelism. It promises an incremental path for parallel conversion of existing software, as well as scalability and performance for a complete rewrite or an entirely new development. Perhaps the main disadvantage of programming with directives is that inserted directives may not necessarily enhance performance. In the worst cases, it can create erroneous results. While vendors have provided tools to perform error-checking and profiling, automation in directive insertion is very limited and often failed on large programs, primarily due to the lack of a thorough enough data dependence analysis. to overcome the deficiency, we have developed a toolkit, CAPO, to automatically insert OpenMP directives in FORTRAN programs and apply certain degrees of optimization. CAPO is aimed at taking advantage of detailed inter-procedural dependence analysis provided by CAPTools, developed by the University of Greenwich, to reduce potential errors made by users. Earlier tests on NAS Benchmarks and ARC3D have demonstrated good success of this tool. In this study, we have applied CAPO to parallelize three large applications in the area of computational fluid dynamics (CFD): OVERFLOW, TLNS3D and INS3D. These codes are widely used for solving Navier-Stokes equations with complicated boundary conditions and turbulence model in multiple zones. Each one comprises of from 50K to 1,00k lines of FORTRAN77. As an example, CAPO took 77 hours to complete the data dependence analysis of OVERFLOW on a workstation (SGI, 175MHz, R10K processor). A fair amount of effort was spent on correcting false dependencies due to lack of necessary knowledge during the analysis. Even so, CAPO provides an easy way for user to interact with the parallelization process. The OpenMP version was generated within a day after the analysis was completed. Due to sequential algorithms involved, code sections in TLNS3D and INS3D need to be restructured by hand to produce more efficient parallel codes. An included figure shows preliminary test results of the generated OVERFLOW with several test cases in single zone. The MPI data points for the small test case were taken from a handcoded MPI version. As we can see, CAPO’s version has achieved 18 fold speed up on 32 nodes of the SGI O2K. For the small test case, it outperformed the MPI version. These results are very encouraging, but further work is needed. For example, although CAPO attempts to place directives on the outer- most parallel loops in an interprocedural framework, it does not insert directives based on the best manual strategy. In particular, it lacks the support of parallelization at the multi-zone level. Future work will emphasize on the development of methodology to work in a multi-zone level and with a hybrid approach. Development of tools to perform more complicated code transformation is also needed. Author Applications Programs (Computers); Software Engineering; Software Development Tools; Parallel Processing (Computers); Interprocessor Communication; Computational Fluid Dynamics; Computer Systems Performance 20000064718 Jet Propulsion Lab., California Inst. of Tech., Pasadena, CA USA A Comparative Study of Interferometric Regridding Algorithms Hensley, Scott, Jet Propulsion Lab., California Inst. of Tech., USA; Safaeinili, Ali, Jet Propulsion Lab., California Inst. of Tech., USA; [1999]; 17p; In English; PIERS 99, 23 Mar. 1999, Pasadena, CA, USA; No Copyright; Avail: CASI; A03, Hardcopy; A01, Microfiche THe paper discusses regridding options: (1) The problem of interpolating data that is not sampled on a uniform grid, that is noisy, and contains gaps is a difficult problem. (2) Several interpolation algorithms have been implemented: (a) Nearest neighbor - Fast and easy but shows some artifacts in shaded relief images. (b) Simplical interpolator - uses plane going through three points 168
containing point where interpolation is required. Reasonably fast and accurate. (c) Convolutional - uses a windowed Gaussian approximating the optimal prolate spheroidal weighting function for a specified bandwidth. (d) First or second order surface fitting - Uses the height data centered in a box about a given point and does a weighted least squares surface fit. Derived from text Algorithms; Computational Grids; Weighting Functions; Interpolation 20000064723 Logistics Management Inst., McLean, VA USA The Economic Benefits of Advanced Product Data Final Report Kordell, Michelle M.; Gentsch, Eric L.; Dec. 1999; 66p; In English Contract(s)/Grant(s): DASW01-99-F-3309 Report No.(s): AD-A377031; LMI-DL910T1; No Copyright; Avail: CASI; A04, Hardcopy; A01, Microfiche The benefits of computer-aided design (CAD) are widely accepted for new designs. What, however, are the economic benefits of converting drawings to CAD models for military mechanical spare parts procurement and production? Based on our analysis, we estimate that repetitive conversions from raster to vector drawings following the first spares procurement cost the Defense Logistics Agency an extra $48 million over the last 10 years for consumable spare and replacement parts procurements. Had DoD taken delivery of CAD data from the original equipment manufacturer, an additional $35 million potentially could have been avoided. We conducted this analysis in order to assess the opportunity for future cost savings through the acquisition and distribution of CAD data. This analysis also serves as a benchmark to measure the offsetting cost of this opportunity, such as data management and possibly the development and implementation costs associated with neutral CAD formats. DTIC Computer Aided Design; Costs; Data Management 20000064730 Carnegie-Mellon Univ., School of Computer Science, Pittsburgh, PA USA Fast and Flexible Application-level Networking on Exokernel Systems Ganger, Gregory R.; Engler, Dawson R.; Kaashoek, M. F.; Briceno, Hector M.; Hunt, Russell; Mar. 2000; 31p; In English; Prepared in cooperation with Stanford Univ., Stanford, CA, Massachusetts Inst of Tech, Cambridge, MA and ExoTec Inc. Contract(s)/Grant(s): F30602-97-2-0288 Report No.(s): AD-A377106; CMU-CS-00-117; No Copyright; Avail: CASI; A03, Hardcopy; A01, Microfiche Application-level networking is a promising software organization for improving performance and functionality for important network services. The xok/ExOS exokernel system includes application-level support for standard network services, while at the same time allowing application writers to specialize networking services. This paper describes how xok/ExOS’s kernel mechanisms and library operating system organization achieves this flexibility, and shares our experiences and lessons learned (both positive and negative). It also describes how we have used this flexibility to build and specialize three network data services: the Cheetah HTTP server, the webswamp web benchmarking tool, and an application-level TCP forwarder. Overall measurements show large performance improvements relative to similar services built on conventional interfaces in each case reaching the maximum possible end-to-end performance for the experimental platform. For example, Cheetah provides factor of 2-4 increases in throughput compared to highly-tuned socket-based implementations and 3-8 compared to conventional systems. Webswamp can offer loads that are 2-8 compared heavier. The TCP forwarder provides 50-300% higher throughput while also providing end-toend TCP semantics that cannot be achieved with POSIX sockets. With more detailed measurements and profiling, these overall performance improvements are also broken down and attributed to the specific specializations described, providing server writers with insights into where to focus their optimization efforts. DTIC Client Server Systems; Software Engineering; Operating Systems (Computers) 20000064914 Office of Management and Budget, Washington, DC USA Progress on Year 2000 Conversion, 8th Quarterly Report (U.S. Office of Management and Budget) Mar. 18, 1999; 56p; In English Report No.(s): PB2000-106044; No Copyright; Avail: CASI; A04, Hardcopy; A01, Microfiche This report is the eighth in a series of quarterly reports to Congress on the Administrations progress in fixing the year 2000 (Y2K) computer problem in Federal systems. This report builds on previous reports by updating previous measures of progress and cost including information on Federally supported, State-run programs. This report also provides more information on the efforts that Federal agencies are undertaking on business continuity and contingency planning. This report summarizes data 169
Page 1 and 2:
Scientific and Technical Aerospace
Page 3 and 4:
Introduction Scientific and Technic
Page 5 and 6:
Subject Divisions Table of Contents
Page 7 and 8:
ground equipment and systems. For a
Page 9 and 10:
Subject Categories of the Division
Page 11 and 12:
48 Oceanography 139 Includes the ph
Page 13 and 14:
72 Atomic and Molecular Physics 191
Page 15 and 16:
Document Availability Information T
Page 17 and 18:
Avail: NASA Public Document Rooms.
Page 19 and 20:
NASA CASI Price Tables — Effectiv
Page 21 and 22:
ALABAMA AUBURN UNIV. AT MONTGOMERY
Page 23 and 24:
��
Page 25 and 26:
20000061433 Pisa Univ., Dipartiment
Page 27 and 28:
English; 34th; 34th Thermophysics C
Page 29 and 30:
ciency of the generated derivative
Page 31 and 32:
20000063509 NASA Glenn Research Cen
Page 33 and 34:
05 AIRCRAFT DESIGN, TESTING AND PER
Page 35 and 36:
A long tradition of aerodynamic des
Page 37 and 38:
sonic and transonic cases will be p
Page 39 and 40:
20000061449 British Aerospace Publi
Page 41 and 42:
20000064593 NASA Langley Research C
Page 43 and 44:
performance required for a proposed
Page 45 and 46:
ment cost and create better product
Page 47 and 48:
tion (NPSS), is focused on the inte
Page 49 and 50:
surface measurement techniques used
Page 51 and 52:
with emphasis on the search for lin
Page 53 and 54:
20000063520 NASA Kennedy Space Cent
Page 55 and 56:
primary issues for it’s adaptatio
Page 57 and 58:
Footage shows the night vertical ta
Page 59 and 60:
Page 61 and 62:
necessary for ignition modeling, or
Page 63 and 64:
engine lifetime, and high power ope
Page 65 and 66:
24 COMPOSITE MATERIALS ��
Page 67 and 68:
ments. For the screening of liner m
Page 69 and 70:
formation of a neutral, fluorescent
Page 71 and 72:
20000065655 Ames Lab., IA USA Funda
Page 73 and 74:
This is a second Triennial Conferen
Page 75 and 76:
of completely charged PSSNa solutio
Page 77 and 78:
20000064100 NASA Langley Research C
Page 79 and 80:
Spacecraft in low Earth orbit (LEO)
Page 81 and 82:
Pressure gradients, i.e. pressure h
Page 83 and 84:
agreed that the NO(sub x) levels co
Page 85 and 86:
tectural approach to extending the
Page 87 and 88:
20000064078 Physics and Electronics
Page 89 and 90:
20000064925 Institute for Human Fac
Page 91 and 92:
20000066607 Institute for Human Fac
Page 93 and 94:
20000062455 National Inst. of Stand
Page 95 and 96:
Added Analysis (Risk Reduction); 6)
Page 97 and 98:
shipboard electrical power generati
Page 99 and 100:
20000067664 Jet Propulsion Lab., Ca
Page 101 and 102:
other, The theoretical framework go
Page 103 and 104:
figures compare the lift and pitchi
Page 105 and 106:
ment phase demonstrate desired feat
Page 107 and 108:
esolution mode with resolving power
Page 109 and 110:
to track 1% or 0.5% changes was hig
Page 111 and 112:
20000067643 NASA Marshall Space Fli
Page 113 and 114:
37 MECHANICAL ENGINEERING ��
Page 115 and 116:
solver based on overset grid techno
Page 117 and 118:
20000067659 NASA Marshall Space Fli
Page 119 and 120:
20000064621 Rensselaer Polytechnic
Page 121 and 122:
that for a specific example of a be
Page 123 and 124:
km x 1000 km). As with the nearly c
Page 125 and 126:
20000064527 Analytical Imaging and
Page 127 and 128:
20000064533 Austin State Univ., Dep
Page 129 and 130:
20000064539 California Univ., Inst.
Page 131 and 132:
Page 133 and 134:
in Yellowstone National Park. We ar
Page 135 and 136:
terns of community hysteresis based
Page 137 and 138:
Page 139 and 140: tion model (DEM) fitting to the sca
Page 141 and 142: The Carolina slate belt is a 10- to
Page 143 and 144: 20000063373 Los Alamos National Lab
Page 145 and 146: 20000064912 New Energy and Industri
Page 147 and 148: and discusses the issues and resear
Page 149 and 150: The objective of this study was to
Page 151 and 152: distributions with engine test resu
Page 153 and 154: 7.6, while the 1926 events appear t
Page 155 and 156: We investigate the compositional di
Page 157 and 158: Flight Center, USA; Moore, T. E., N
Page 159 and 160: This report presents the Applied Me
Page 161 and 162: in Goa, India, by 3 - 4 days. Wind
Page 163 and 164: 51 LIFE SCIENCES (GENERAL) Includes
Page 165 and 166: 20000062717 Joint Inst. for Nuclear
Page 167 and 168: 20000064015 Institute for Human Fac
Page 169 and 170: during 50% (SD 4%) of the breathing
Page 171 and 172: 20000064711 Massachusetts Inst. of
Page 173 and 174: The analysis of this smaller data s
Page 175 and 176: vided a functional helmet mount. Th
Page 177 and 178: 61 COMPUTER PROGRAMMING AND SOFTWAR
Page 179 and 180: The goal of multi-image classificat
Page 181 and 182: 20000063526 Defence Science and Tec
Page 183 and 184: 85 dB amplifier design evolved by o
Page 185 and 186: 20000064594 New Mexico Univ., Elect
Page 187 and 188: 20000064615 NASA Ames Research Cent
Page 189: declarations with a pattern recogni
Page 193 and 194: has his own responsibility, while t
Page 195 and 196: the people, documents and projects
Page 197 and 198: and experimental technology, many o
Page 199 and 200: tems. In particular this applicatio
Page 201 and 202: Over the past decade, high performa
Page 203 and 204: expensive than a single analysis. I
Page 205 and 206: 20000064726 Carnegie-Mellon Univ.,
Page 207 and 208: 20000064099 Teledyne Brown Engineer
Page 209 and 210: 20000066597 Geophysical Observatory
Page 211 and 212: 20000065633 Institute TNO of Applie
Page 213 and 214: important conclusion is that for su
Page 215 and 216: est frame of the string. The modifi
Page 217 and 218: isotopes. The method and program we
Page 219 and 220: A detailed study is undertaken, usi
Page 221 and 222: 20000063497 Kaiser Electro-Optics,
Page 223 and 224: delivers 60% of the x-ray flux from
Page 225 and 226: 76 SOLID-STATE PHYSICS ��
Page 227 and 228: 20000064660 Abdus Salam Internation
Page 229 and 230: 20000067648 NASA Marshall Space Fli
Page 231 and 232: 20000064703 Institute of Atomic Ene
Page 233 and 234: The BRST approach is applied to the
Page 235 and 236: 20000067671 Hampton Univ., VA USA A
Page 237 and 238: 82 DOCUMENTATION AND INFORMATION SC
Page 239 and 240: ecoming almost too cumbersome to be
Page 241 and 242:
physical world. These are the two p
Page 243 and 244:
with WFPC2 data. Tests of HSTphot
Page 245 and 246:
A brief description is given of the
Page 247 and 248:
20000064070 NASA Ames Research Cent
Page 249 and 250:
gas, show variations that have rema
Page 251 and 252:
egy, called Power In that would all
Page 253 and 254:
Page 255 and 256:
20000064089 Smithsonian Astrophysic
Page 257 and 258:
A ABERRATION, 143, 198 ABRASION, 22
Page 259 and 260:
COMPRESSIBLE FLOW, 83 COMPRESSION L
Page 261 and 262:
FATIGUE (MATERIALS), 50, 93, 96 FAT
Page 263 and 264:
LASING, 90 LATE STARS, 224 LATITUDE
Page 265 and 266:
PATCH ANTENNAS, 74 PATHOLOGY, 98 PA
Page 267 and 268:
SPACE PLATFORMS, 38 SPACE PROBES, 2
Page 269 and 270:
WIND PROFILES, 137 WIND TUNNEL TEST
Page 271 and 272:
Brown, Andy, 38 Brown, April S., 20
Page 273 and 274:
Grosvenor, C. A., 70 Grosvenor, J.
Page 275 and 276:
Lu, Gui-Ying, 50 Lugg, Desmond J.,
Page 277 and 278:
Robertson, Tony, 39 Robinson, Jeffr
Page 279:
Warner, J. D., 63 Warren, James, 16
show all

Scientific and Technical Aerospace Reports Volume 38 July 28, 2000

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?