Contention-free communication scheduling for array redistribution

被引:30
作者
Guo, MY
Nakata, I
Yamashita, Y
机构
[1] Univ Tsukuba, Inst Informat Sci & Elect, Tsukuba, Ibaraki 3050006, Japan
[2] Univ Lib & Informat Sci, Tsukuba, Ibaraki 3058550, Japan
关键词
parallelizing compilers; HPF; array redistribution; communication scheduling; distributed memory machines;
D O I
10.1016/S0167-8191(00)00027-2
中图分类号
TP301 [理论、方法];
学科分类号
081202 [计算机软件与理论];
摘要
Array redistribution is required often in programs on distributed memory parallel computers. It is essential to use efficient algorithms for redistribution, otherwise the performance of the programs may degrade considerably. The redistribution overheads consist of two parts: index computation and interprocessor communication. If there is no communication scheduling in a redistribution algorithm, the communication contention may occur, which increases the communication waiting time. In order to solve this problem, in this paper, we propose a technique to schedule the communication so that it becomes contention-free, Our approach initially generates a communication table to represent the communication relations among sending nodes and receiving nodes. According to the communication table, we then generate another table named communication scheduling table. Each column of communication scheduling table is a permutation of receiving node numbers in each communication step. Thus the communications in our redistribution algorithm are contention-free. Our approach can deal with multi-dimensional "shape changing redistribution". (C) 2000 Elsevier Science B.V. All rights reserved.
引用
收藏
页码:1325 / 1343
页数:19
相关论文
共 16 条
[1]
Scheduling block-cyclic array redistribution [J].
Desprez, F ;
Dongarra, J ;
Petitet, A ;
Randriamaro, C ;
Robert, Y .
IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 1998, 9 (02) :192-205
[2]
GUO M, 1998, P 3 INT WORKSH HIGH
[3]
GUO M, 1998, THESIS U TSUKUBA
[4]
Guo MY, 1998, IEICE T INF SYST, VE81D, P1195
[5]
Processor mapping techniques toward efficient data redistribution [J].
Kalns, ET ;
Ni, LM .
IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 1995, 6 (12) :1234-1247
[6]
Kaushik S.D., 1994, P 8 ACM INT C SUP JU
[7]
KAUSHIK SD, 1995, MULTIPHASE REDISTRIB
[8]
LIM YW, 1997, P 26 INT C PAR PROC
[9]
LIM YW, 1996, P IEEE S PAR DISTR P
[10]
CP-PACS: A massively parallel processor at the University of Tsukuba [J].
Nakazawa, K ;
Nakamura, H ;
Boku, T ;
Nakata, I ;
Yamashita, Y .
PARALLEL COMPUTING, 1999, 25 (13-14) :1635-1661