Large System Performance of SPEC OMP Benchmark Suites

Hideki Saito; Greg Gaertner; Wesley Jones; Rudolf Eigenmann; Hidetoshi Iwashita; Ron Lieberman; Matthijs Van Waveren; Brian Whitney

doi:10.1023/A:1023086618401

Large System Performance of SPEC OMP Benchmark Suites

Hideki Saito, Greg Gaertner, Wesley Jones, Rudolf Eigenmann, Hidetoshi Iwashita, Ron Lieberman, Matthijs Van Waveren, Brian Whitney

Computational Science

Research output: Contribution to journal › Article › peer-review

8 Scopus Citations

Abstract

Performance characteristics of application programs on large-scale systems are often significantly different from those on smaller systems. SPEC OMP includes benchmark suites intended for measuring performance of modern shared memory parallel systems. SPEC OMPM2001 is developed for medium-scale (4- to 16-way) systems whereas SPEC OMPL2001 is developed for large-scale systems (32-way and larger). We present our experiences on benchmark development in achieving good scalability using the OpenMP API. We also analyze the published results of these benchmark suites on large systems (32-way and larger), based on application program behavior and systems' architectural features.

Original language	American English
Pages (from-to)	197-209
Number of pages	13
Journal	International Journal of Parallel Programming
Volume	31
Issue number	3
DOIs	https://doi.org/10.1023/A:1023086618401 https://doi.org/10.1023/A:1023086618401
State	Published - 2003

NREL Publication Number

NREL/JA-530-34463

Keywords

Benchmark
High-performance computing
OpenMP
Performance evaluation
SPEC OMP

Access to Document

Cite this

@article{df56a2f44ed54423961674f682fb0596,

title = "Large System Performance of SPEC OMP Benchmark Suites",

abstract = "Performance characteristics of application programs on large-scale systems are often significantly different from those on smaller systems. SPEC OMP includes benchmark suites intended for measuring performance of modern shared memory parallel systems. SPEC OMPM2001 is developed for medium-scale (4- to 16-way) systems whereas SPEC OMPL2001 is developed for large-scale systems (32-way and larger). We present our experiences on benchmark development in achieving good scalability using the OpenMP API. We also analyze the published results of these benchmark suites on large systems (32-way and larger), based on application program behavior and systems' architectural features.",

keywords = "Benchmark, High-performance computing, OpenMP, Performance evaluation, SPEC OMP",

author = "Hideki Saito and Greg Gaertner and Wesley Jones and Rudolf Eigenmann and Hidetoshi Iwashita and Ron Lieberman and {Van Waveren}, Matthijs and Brian Whitney",

year = "2003",

doi = "10.1023/A:1023086618401",

language = "American English",

volume = "31",

pages = "197--209",

journal = "International Journal of Parallel Programming",

issn = "0885-7458",

publisher = "Springer New York",

number = "3",

}

TY - JOUR

T1 - Large System Performance of SPEC OMP Benchmark Suites

AU - Saito, Hideki

AU - Gaertner, Greg

AU - Jones, Wesley

AU - Eigenmann, Rudolf

AU - Iwashita, Hidetoshi

AU - Lieberman, Ron

AU - Van Waveren, Matthijs

AU - Whitney, Brian

PY - 2003

Y1 - 2003

N2 - Performance characteristics of application programs on large-scale systems are often significantly different from those on smaller systems. SPEC OMP includes benchmark suites intended for measuring performance of modern shared memory parallel systems. SPEC OMPM2001 is developed for medium-scale (4- to 16-way) systems whereas SPEC OMPL2001 is developed for large-scale systems (32-way and larger). We present our experiences on benchmark development in achieving good scalability using the OpenMP API. We also analyze the published results of these benchmark suites on large systems (32-way and larger), based on application program behavior and systems' architectural features.

AB - Performance characteristics of application programs on large-scale systems are often significantly different from those on smaller systems. SPEC OMP includes benchmark suites intended for measuring performance of modern shared memory parallel systems. SPEC OMPM2001 is developed for medium-scale (4- to 16-way) systems whereas SPEC OMPL2001 is developed for large-scale systems (32-way and larger). We present our experiences on benchmark development in achieving good scalability using the OpenMP API. We also analyze the published results of these benchmark suites on large systems (32-way and larger), based on application program behavior and systems' architectural features.

KW - Benchmark

KW - High-performance computing

KW - OpenMP

KW - Performance evaluation

KW - SPEC OMP

UR - http://www.scopus.com/inward/record.url?scp=0347763428&partnerID=8YFLogxK

U2 - 10.1023/A:1023086618401

DO - 10.1023/A:1023086618401

M3 - Article

AN - SCOPUS:0347763428

SN - 0885-7458

VL - 31

SP - 197

EP - 209

JO - International Journal of Parallel Programming

JF - International Journal of Parallel Programming

IS - 3

ER -

Large System Performance of SPEC OMP Benchmark Suites

Abstract

NREL Publication Number

Keywords

Access to Document

Other files and links

Fingerprint

Cite this