1532-0634/asset/olbannerleft.gif?v=1&s=a4e4e145787de94e1d91eaab3c8c29d8a9d96a26)
Research Article
Managing distributed shared arrays in a bulk-synchronous parallel programming environment
Article first published online: 7 JAN 2004
DOI: 10.1002/cpe.768
Copyright © 2004 John Wiley & Sons, Ltd.
Issue
1532-0634/asset/cover.gif?v=1&s=6094df24c795ce080ff6df6ff3b6bcec19adb708 "Concurrency and Computation: Practice and Experience")
Concurrency and Computation: Practice and Experience
Special Issue: Compilers for Parallel Computers
Volume 16, Issue 2-3, pages 133–153, February - March 2004
Additional Information
How to Cite
Kessler, C. W. (2004), Managing distributed shared arrays in a bulk-synchronous parallel programming environment. Concurrency Computat.: Pract. Exper., 16: 133–153. doi: 10.1002/cpe.768
Publication History
- Issue published online: 7 JAN 2004
- Article first published online: 7 JAN 2004
- Manuscript Accepted: 1 JUL 2002
- Manuscript Revised: 29 MAY 2002
- Manuscript Received: 20 SEP 2001
- Abstract
- References
- Cited By
Keywords:
- NestStep;
- BSP model;
- bulk-synchronous parallelism;
- parallel programming language;
- distributed shared array;
- runtime scheduling of communication
Abstract
NestStep is a parallel programming language for the BSP (bulk-hronous parallel) programming model. In this article we describe the concept of distributed shared arrays in NestStep and its implementation on top of MPI. In particular, we present a novel method for runtime scheduling of irregular, direct remote accesses to sections of distributed shared arrays. Our method, which is fully parallelized, uses conventional two-sided message passing and thus avoids the overhead of a standard implementation of direct remote memory access based on one-sided communication. The main prerequisite is that the given program is structured in a BSP-compliant way. Copyright © 2004 John Wiley & Sons, Ltd.