Packing/Unpacking Using MPI User-Defined Datatypes for Efficient Data Redistribution(Software Support and Optimization Techniques)(<Special Section>Hardware/Software Support for High Performance Scientific and Engineering Computing)

スポンサーリンク

概要

• 論文の詳細を見る

• In many parallel programs, run-time data redistribution is usually required to enhance data locality and reduce remote memory access on the distributed memory multicomputers. Research on data redistribution algorithms has recently matured. The time required to generate data sets and processor sets is much lesser than before. Therefore, packing/unpacking has become a relatively high cost in redistribution. In this paper, we present methods to perform BLOCK-CYCLIC(s) to BLOCK-CYCLIC(t) redistribution, using MPI user-defined datatypes. This method reduces the required memory buffers and avoids unnecessary movement of data. Theoretical models are presented to determine the best method for redistribution. The methods were implemented on an IBM SP2 parallel machine to evaluate the performance of the proposed methods. The experimental results indicate that this approach can clearly improve the redistribution in most cases.

• 社団法人電子情報通信学会の論文
• 2004-07-01

著者

• Yang Chu-sing

  Department Of Computer Science And Engineering National Sun Yat-sen University

• Bai Sheng-wen

  Department Of Computer Science And Engineering National Sun Yat-sen University

• Huang Tsung-chuan

  Department Of Electrical Engineering National Sun Yat-sen University

関連論文

• Multiresolution Motion Estimation with Zerotree Coding Aware Metric(Multimedia Systems)
• Wavelet Domain Half-Pixel Motion Compensation Using H-Transform(Image Processing, Image Pattern Recognition)
• Packing/Unpacking Using MPI User-Defined Datatypes for Efficient Data Redistribution(Software Support and Optimization Techniques)(Hardware/Software Support for High Performance Scientific and Engineering Computing)

スポンサーリンク