Evaluating Performance of OpenMP and MPI on the SGI Origin 2000 With Benchmarks of Realistic Problem Sizes

reportActive / Technical Report | Accession Number: ADA383931 | Open PDF

Abstract:

Six application benchmarks, including four numerical aerodynamic simulation NAS codes, provided by H. Jin and J. Wu, were previously parallelized using OpenMP and message-passing interface MPI and run on a 128-processor Silicon Graphics Inc. SGI Origin 2000. Detailed profile data were collected to understand the factors causing imperfect scalability. The results show that load imbalance and cost of remote accesses are the main factors in limited speedup of the OpenMP versions, whereas communication costs are the single major factor in the performance of the MPI versions.

Author(s):
Zoltani, Csaba K. ; Satya-narayana, Punyam ; Hisley, Dixie
Author Organization(s):
ARMY RESEARCH LAB ABERDEEN PROVING GROUND MD
Descriptive Note:
Final rept. Oct 1998-Sep 1999
Supplementary Note:
DOI: 10.21236/ADA383931
Pagination:
0028

Security Markings

DOCUMENT & CONTEXTUAL SUMMARY

Distribution:
Approved For Public Release

RECORD

Collection: TR
Identifying Numbers
Report Number(s):
ARL-TR-2324
Monitor Series:
ARL/APG
 Subject Terms
Joint Capability Areas:
JCA_6_Net Centric; JCA_6.2_Enterprise Services; JCA_6.1_Information Transport; JCA_5_Command and Control; JCA_1.2.5_Lessons Learned
Modernization Areas:
Quantum Science and Computing; Microelectronics
Communities of Interest:
Ground and Sea Platforms
Descriptor(s):
*PARALLEL PROCESSING, *COMPUTER BENCHMARKING, DISTRIBUTED DATA PROCESSING, COMPUTER COMMUNICATIONS, PERFORMANCE(ENGINEERING), COMPUTER PROGRAM VERIFICATION, MESSAGE PROCESSING
Field(s)/Group(s):
Computer Programming and Software, Computer Systems
Keyword(s):
MPI(MESSAGE PASSING INTERFACE)
Report Date:
2000 Sep 01
Creation Date:
2000 Nov 29