A Programming Framework for Incremental Data Distribution in Iterative Applications

被引：1

作者：

Chan, Philip ^{[1
]}

Abramson, David ^{[1
]}

机构：

[1] Monash Univ, Caulfield Sch Informat Technol, MESSAGE Lab, Caulfield, Vic, Australia

来源：

PROCEEDINGS OF THE 2008 INTERNATIONAL SYMPOSIUM ON PARALLEL AND DISTRIBUTED PROCESSING WITH APPLICATIONS | 2008年

关键词：

D O I：

10.1109/ISPA.2008.105

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

Successful HPC over desktop grids and non-dedicated NOWs is challenging, since good performance is difficult to achieve due to dynamic workloads. On iterative data-parallel applications, this is addressed by dynamic data distribution. However, current approaches migrate an application from one distribution to another in one single phase, which can impact performance. In this paper, we present D-3-ARC, a programming framework to support adaptive and incremental data distribution, so that data migration takes place over several successive iterations. D-3-ARC consists of a runtime system and an API for specifying the distribution of arrays as well as how data redistribution takes place. We demonstrate how D-3-ARC can be used to develop an incremental strategy for data distribution in a Poisson solver, utilising a runtime feedback mechanism to determine how much data to migrate during each iteration.

引用

页码：244 / 251

页数：8

共 21 条

[1]

[Anonymous], 2003, Parallel Processing Letters

[2]

[Anonymous], 1994, MPI MESS PASS INT ST

[3]

ARABE JNC, 1995, SUPERCOMPUTING 95

[4]

Balay S, 1997, MODERN SOFTWARE TOOLS FOR SCIENTIFIC COMPUTING, P163

[5]

BALAY S, 2004, ANL9511

[6]

Deltz SJ, 2004, NINTH INTERNATIONAL WORKSHOP ON HIGH-LEVEL PARALLEL PROGRAMMING MODELS AND SUPPORTIVE ENVIRONMENTS, PROCEEDINGS, P42

[7]

Devine K, 2002, COMPUT SCI ENG, V4, P90, DOI 10.1109/5992.988653

[8]

EDJLALI G, 1995, P 9 INT PAR PROC S S

[9] Program Control Language: a programming language for adaptive distributed applications [J].

Ensink, B ;

Stanley, J ;

Adve, V .

JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 2003, 63 (11) :1082-1104

[10] Efficient run-time support for irregular block-structured applications [J].

Fink, SJ ;

Baden, SB ;

Kohn, SR .

JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 1998, 50 (1-2) :61-82

← 1 2 3 →