skip to main content
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Tuning collective communication for Partitioned Global Address Space programming models

Abstract

Partitioned Global Address Space (PGAS) languages offer programmers the convenience of a shared memory programming style combined with locality control necessary to run on large-scale distributed memory systems. Even within a PGAS language programmers often need to perform global communication operations such as broadcasts or reductions, which are best performed as collective operations in which a group of threads work together to perform the operation. In this study we consider the problem of implementing collective communication within PGAS languages and explore some of the design trade-offs in both the interface and implementation. In particular, PGAS collectives have semantic issues that are different than in send–receive style message passing programs, and different implementation approaches that take advantage of the one-sided communication style in these languages. We present an implementation framework for PGAS collectives as part of the GASNet communication layer, which supports shared memory, distributed memory and hybrids. The framework supports a broad set of algorithms for each collective, over which the implementation may be automatically tuned. In conclusion, we demonstrate the benefit of optimized GASNet collectives using application benchmarks written in UPC, and demonstrate that the GASNet collectives can deliver scalable performance on a variety of state-of-the-art parallel machines includingmore » a Cray XT4, an IBM BlueGene/P, and a Sun Constellation system with InfiniBand interconnect.« less

Authors:
 [1];  [2];  [2];  [2]
  1. Univ. of California, Berkeley, CA (United States)
  2. Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
Publication Date:
Research Org.:
Univ. of California, Berkeley, CA (United States); Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR) (SC-21); National Science Foundation (NSF)
OSTI Identifier:
1407106
Grant/Contract Number:  
AC02-05CH11231; FC03-01ER25509; FC02-07ER25799; AC02-06CH11357; AC05-00OR22725; OCI-0749190
Resource Type:
Accepted Manuscript
Journal Name:
Parallel Computing
Additional Journal Information:
Journal Volume: 37; Journal Issue: 9; Journal ID: ISSN 0167-8191
Publisher:
Elsevier
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; Partitioned Global Address Space languages; collective communication; one-sided communication

Citation Formats

Nishtala, Rajesh, Zheng, Yili, Hargrove, Paul H., and Yelick, Katherine A. Tuning collective communication for Partitioned Global Address Space programming models. United States: N. p., 2011. Web. doi:10.1016/j.parco.2011.05.006.
Nishtala, Rajesh, Zheng, Yili, Hargrove, Paul H., & Yelick, Katherine A. Tuning collective communication for Partitioned Global Address Space programming models. United States. doi:10.1016/j.parco.2011.05.006.
Nishtala, Rajesh, Zheng, Yili, Hargrove, Paul H., and Yelick, Katherine A. Sun . "Tuning collective communication for Partitioned Global Address Space programming models". United States. doi:10.1016/j.parco.2011.05.006. https://www.osti.gov/servlets/purl/1407106.
@article{osti_1407106,
title = {Tuning collective communication for Partitioned Global Address Space programming models},
author = {Nishtala, Rajesh and Zheng, Yili and Hargrove, Paul H. and Yelick, Katherine A.},
abstractNote = {Partitioned Global Address Space (PGAS) languages offer programmers the convenience of a shared memory programming style combined with locality control necessary to run on large-scale distributed memory systems. Even within a PGAS language programmers often need to perform global communication operations such as broadcasts or reductions, which are best performed as collective operations in which a group of threads work together to perform the operation. In this study we consider the problem of implementing collective communication within PGAS languages and explore some of the design trade-offs in both the interface and implementation. In particular, PGAS collectives have semantic issues that are different than in send–receive style message passing programs, and different implementation approaches that take advantage of the one-sided communication style in these languages. We present an implementation framework for PGAS collectives as part of the GASNet communication layer, which supports shared memory, distributed memory and hybrids. The framework supports a broad set of algorithms for each collective, over which the implementation may be automatically tuned. In conclusion, we demonstrate the benefit of optimized GASNet collectives using application benchmarks written in UPC, and demonstrate that the GASNet collectives can deliver scalable performance on a variety of state-of-the-art parallel machines including a Cray XT4, an IBM BlueGene/P, and a Sun Constellation system with InfiniBand interconnect.},
doi = {10.1016/j.parco.2011.05.006},
journal = {Parallel Computing},
number = 9,
volume = 37,
place = {United States},
year = {2011},
month = {6}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 17 works
Citation information provided by
Web of Science

Save / Share: