Accurate data redistribution cost estimation in software distributed shared memory systems

Donald G. Morris, David K Lowenthal

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Distributing data is one of the key problems in implementing efficient distributed-memory parallel programs. The problem becomes more difficult in programs where data redistribution between computational phases is considered. The global data distribution problem is to find the optimal distribution in multi-phase parallel programs. Solving this problem requires accurate knowledge of data redistribution cost. We are investigating this problem in the context of a software distributed shared memory (SDSM) system, in which obtaining accurate redistribution cost estimates is difficult. This is because SDSM communication is implicit: It depends on access patterns, page locations, and the SDSM consistency protocol. We have developed integrated compile- and run-time analysis for SDSM systems to determine accurate redistribution cost estimates with low overhead. Our resulting system, SUIF-Adapt, can efficiently and accurately estimate execution time, including redistribution, to within 5% of the actual time in all of our test cases and is often much closer. These precise costs enable SUIF-Adapt to find efficient global data distributions in multiple-phase programs.

Original languageEnglish (US)
Pages (from-to)62-71
Number of pages10
JournalSIGPLAN Notices (ACM Special Interest Group on Programming Languages)
Volume36
Issue number7
StatePublished - Jul 2001
Externally publishedYes

Fingerprint

Computer systems
Data storage equipment
Costs
Network protocols
Communication

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Software

Cite this

@article{e054ca9132224569982fecc5e6892294,
title = "Accurate data redistribution cost estimation in software distributed shared memory systems",
abstract = "Distributing data is one of the key problems in implementing efficient distributed-memory parallel programs. The problem becomes more difficult in programs where data redistribution between computational phases is considered. The global data distribution problem is to find the optimal distribution in multi-phase parallel programs. Solving this problem requires accurate knowledge of data redistribution cost. We are investigating this problem in the context of a software distributed shared memory (SDSM) system, in which obtaining accurate redistribution cost estimates is difficult. This is because SDSM communication is implicit: It depends on access patterns, page locations, and the SDSM consistency protocol. We have developed integrated compile- and run-time analysis for SDSM systems to determine accurate redistribution cost estimates with low overhead. Our resulting system, SUIF-Adapt, can efficiently and accurately estimate execution time, including redistribution, to within 5{\%} of the actual time in all of our test cases and is often much closer. These precise costs enable SUIF-Adapt to find efficient global data distributions in multiple-phase programs.",
author = "Morris, {Donald G.} and Lowenthal, {David K}",
year = "2001",
month = "7",
language = "English (US)",
volume = "36",
pages = "62--71",
journal = "ACM SIGPLAN Notices",
issn = "1523-2867",
publisher = "Association for Computing Machinery (ACM)",
number = "7",

}

TY - JOUR

T1 - Accurate data redistribution cost estimation in software distributed shared memory systems

AU - Morris, Donald G.

AU - Lowenthal, David K

PY - 2001/7

Y1 - 2001/7

N2 - Distributing data is one of the key problems in implementing efficient distributed-memory parallel programs. The problem becomes more difficult in programs where data redistribution between computational phases is considered. The global data distribution problem is to find the optimal distribution in multi-phase parallel programs. Solving this problem requires accurate knowledge of data redistribution cost. We are investigating this problem in the context of a software distributed shared memory (SDSM) system, in which obtaining accurate redistribution cost estimates is difficult. This is because SDSM communication is implicit: It depends on access patterns, page locations, and the SDSM consistency protocol. We have developed integrated compile- and run-time analysis for SDSM systems to determine accurate redistribution cost estimates with low overhead. Our resulting system, SUIF-Adapt, can efficiently and accurately estimate execution time, including redistribution, to within 5% of the actual time in all of our test cases and is often much closer. These precise costs enable SUIF-Adapt to find efficient global data distributions in multiple-phase programs.

AB - Distributing data is one of the key problems in implementing efficient distributed-memory parallel programs. The problem becomes more difficult in programs where data redistribution between computational phases is considered. The global data distribution problem is to find the optimal distribution in multi-phase parallel programs. Solving this problem requires accurate knowledge of data redistribution cost. We are investigating this problem in the context of a software distributed shared memory (SDSM) system, in which obtaining accurate redistribution cost estimates is difficult. This is because SDSM communication is implicit: It depends on access patterns, page locations, and the SDSM consistency protocol. We have developed integrated compile- and run-time analysis for SDSM systems to determine accurate redistribution cost estimates with low overhead. Our resulting system, SUIF-Adapt, can efficiently and accurately estimate execution time, including redistribution, to within 5% of the actual time in all of our test cases and is often much closer. These precise costs enable SUIF-Adapt to find efficient global data distributions in multiple-phase programs.

UR - http://www.scopus.com/inward/record.url?scp=17644392481&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=17644392481&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:17644392481

VL - 36

SP - 62

EP - 71

JO - ACM SIGPLAN Notices

JF - ACM SIGPLAN Notices

SN - 1523-2867

IS - 7

ER -