Algorithm and data optimization techniques for scaling to massively threaded systems

John A. Stratton; Christopher Rodrigues; I. Jui Ray Sung; Li Wen Chang; Nasser Anssari; Geng Daniel Liu; Wen-Mei W Hwu; Nady Obeid

doi:10.1109/MC.2012.194

Algorithm and data optimization techniques for scaling to massively threaded systems

John A. Stratton, Christopher Rodrigues, I. Jui Ray Sung, Li Wen Chang, Nasser Anssari, Geng Daniel Liu, Wen-Mei W Hwu, Nady Obeid

Research output: Contribution to specialist publication › Article

Abstract

A study of the implementation patterns among massively threaded applications for many-core GPUs reveals that each of the seven most commonly used algorithm and data optimization techniques can enhance the performance of applicable kernels by 2 to 10× in current processors while also improving future scalability.

Original language	English (US)
Pages	26-32
Number of pages	7
Volume	45
No	8
Specialist publication	Computer
DOIs	https://doi.org/10.1109/MC.2012.194
State	Published - Aug 29 2012

Keywords

Parboil benchmarks
accelerators
massively threaded systems
optimization patterns
scalability

ASJC Scopus subject areas

Computer Science(all)

Access to Document

10.1109/MC.2012.194

Fingerprint Dive into the research topics of 'Algorithm and data optimization techniques for scaling to massively threaded systems'. Together they form a unique fingerprint.

Cite this

Stratton, J. A., Rodrigues, C., Sung, I. J. R., Chang, L. W., Anssari, N., Liu, G. D., Hwu, W-M. W., & Obeid, N. (2012). Algorithm and data optimization techniques for scaling to massively threaded systems. Computer, 45(8), 26-32. https://doi.org/10.1109/MC.2012.194

@misc{e0a57f31151f4ef1aac03c5f707ceb47,

title = "Algorithm and data optimization techniques for scaling to massively threaded systems",

abstract = "A study of the implementation patterns among massively threaded applications for many-core GPUs reveals that each of the seven most commonly used algorithm and data optimization techniques can enhance the performance of applicable kernels by 2 to 10× in current processors while also improving future scalability.",

keywords = "Parboil benchmarks, accelerators, massively threaded systems, optimization patterns, scalability",

author = "Stratton, {John A.} and Christopher Rodrigues and Sung, {I. Jui Ray} and Chang, {Li Wen} and Nasser Anssari and Liu, {Geng Daniel} and Hwu, {Wen-Mei W} and Nady Obeid",

year = "2012",

month = aug,

day = "29",

doi = "10.1109/MC.2012.194",

language = "English (US)",

volume = "45",

pages = "26--32",

journal = "Computer",

issn = "0018-9162",

publisher = "IEEE Computer Society",

}

TY - GEN

T1 - Algorithm and data optimization techniques for scaling to massively threaded systems

AU - Stratton, John A.

AU - Rodrigues, Christopher

AU - Sung, I. Jui Ray

AU - Chang, Li Wen

AU - Anssari, Nasser

AU - Liu, Geng Daniel

AU - Hwu, Wen-Mei W

AU - Obeid, Nady

PY - 2012/8/29

Y1 - 2012/8/29

N2 - A study of the implementation patterns among massively threaded applications for many-core GPUs reveals that each of the seven most commonly used algorithm and data optimization techniques can enhance the performance of applicable kernels by 2 to 10× in current processors while also improving future scalability.

AB - A study of the implementation patterns among massively threaded applications for many-core GPUs reveals that each of the seven most commonly used algorithm and data optimization techniques can enhance the performance of applicable kernels by 2 to 10× in current processors while also improving future scalability.

KW - Parboil benchmarks

KW - accelerators

KW - massively threaded systems

KW - optimization patterns

KW - scalability

UR - http://www.scopus.com/inward/record.url?scp=84865331893&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84865331893&partnerID=8YFLogxK

U2 - 10.1109/MC.2012.194

DO - 10.1109/MC.2012.194

M3 - Article

AN - SCOPUS:84865331893

VL - 45

SP - 26

EP - 32

JO - Computer

JF - Computer

SN - 0018-9162

ER -