High-performance conjugate-gradient benchmark: A new metric for ranking high-performance computing systems

Jack Dongarra; Michael A. Heroux; Piotr Luszczek

doi:10.1177/1094342015593158

High-performance conjugate-gradient benchmark: A new metric for ranking high-performance computing systems

Jack Dongarra, Michael A. Heroux, Piotr Luszczek

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

107 Scopus citations

Abstract

We describe a new high-performance conjugate-gradient (HPCG) benchmark. HPCG is composed of computations and data-access patterns commonly found in scientific applications. HPCG strives for a better correlation to existing codes from the computational science domain and to be representative of their performance. HPCG is meant to help drive the computer system design and implementation in directions that will better impact future performance improvement.

Original language	English
Pages (from-to)	3-10
Number of pages	8
Journal	International Journal of High Performance Computing Applications
Volume	30
Issue number	1
DOIs	https://doi.org/10.1177/1094342015593158
State	Published - Feb 1 2016
Externally published	Yes

Keywords

HPC benchmarking
Preconditioned conjugate gradient
additive Schwarz
multigrid smoothing
validation and verification

Access to Document

10.1177/1094342015593158

Cite this

@article{b2eeb08a6f1b495d91d9ee844ee46142,

title = "High-performance conjugate-gradient benchmark: A new metric for ranking high-performance computing systems",

abstract = "We describe a new high-performance conjugate-gradient (HPCG) benchmark. HPCG is composed of computations and data-access patterns commonly found in scientific applications. HPCG strives for a better correlation to existing codes from the computational science domain and to be representative of their performance. HPCG is meant to help drive the computer system design and implementation in directions that will better impact future performance improvement.",

keywords = "HPC benchmarking, Preconditioned conjugate gradient, additive Schwarz, multigrid smoothing, validation and verification",

author = "Jack Dongarra and Heroux, {Michael A.} and Piotr Luszczek",

note = "Publisher Copyright: {\textcopyright} The Author(s) 2015.",

year = "2016",

month = feb,

day = "1",

doi = "10.1177/1094342015593158",

language = "English",

volume = "30",

pages = "3--10",

journal = "International Journal of High Performance Computing Applications",

issn = "1094-3420",

publisher = "SAGE Publications",

number = "1",

}

TY - JOUR

T1 - High-performance conjugate-gradient benchmark

T2 - A new metric for ranking high-performance computing systems

AU - Dongarra, Jack

AU - Heroux, Michael A.

AU - Luszczek, Piotr

N1 - Publisher Copyright: © The Author(s) 2015.

PY - 2016/2/1

Y1 - 2016/2/1

N2 - We describe a new high-performance conjugate-gradient (HPCG) benchmark. HPCG is composed of computations and data-access patterns commonly found in scientific applications. HPCG strives for a better correlation to existing codes from the computational science domain and to be representative of their performance. HPCG is meant to help drive the computer system design and implementation in directions that will better impact future performance improvement.

AB - We describe a new high-performance conjugate-gradient (HPCG) benchmark. HPCG is composed of computations and data-access patterns commonly found in scientific applications. HPCG strives for a better correlation to existing codes from the computational science domain and to be representative of their performance. HPCG is meant to help drive the computer system design and implementation in directions that will better impact future performance improvement.

KW - HPC benchmarking

KW - Preconditioned conjugate gradient

KW - additive Schwarz

KW - multigrid smoothing

KW - validation and verification

UR - http://www.scopus.com/inward/record.url?scp=84958963839&partnerID=8YFLogxK

U2 - 10.1177/1094342015593158

DO - 10.1177/1094342015593158

M3 - Article

AN - SCOPUS:84958963839

SN - 1094-3420

VL - 30

SP - 3

EP - 10

JO - International Journal of High Performance Computing Applications

JF - International Journal of High Performance Computing Applications

IS - 1

ER -

High-performance conjugate-gradient benchmark: A new metric for ranking high-performance computing systems

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this