Paper Info
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 paper 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. Finally, 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.
Year
DOI
Venue
2011
10.1016/j.parco.2011.05.006
Parallel Computing
Keywords
Field
DocType
memory system,gasnet communication layer,partitioned global address space,one-sided communication,pgas language,global communication operation,collective communication,different implementation approach,pgas collective,one-sided communication style,programming model,pgas language programmer,partitioned global address space languages,implementation framework,shared memory,distributed memory,message passing
InfiniBand,Shared memory,Programming paradigm,Computer science,Parallel computing,Distributed memory,Programming style,Theoretical computer science,Partitioned global address space,Message passing,Scalability,Distributed computing
Journal
Volume
Issue
ISSN
37
9
Parallel Computing
Citations 
PageRank 
References 
8
0.50
8
Authors
4
Name
Order
Citations
PageRank
Rajesh Nishtala136524.36
Yili Zheng218315.22
Paul H. Hargrove3453.53
Katherine A. Yelick43494407.23