A Performance-based Parallel Loop Self-Scheduling Using Hybrid OpenMP and MPI Programming on Multicore SMP Clusters
碩士 === 東海大學 === 資訊工程與科學系碩士在職專班 === 97 === Parallel loop self-scheduling on parallel and distributed systems has been a critical problem and it is becoming more difficult to deal with in the emerging heterogeneous cluster computing environments. In the past, some self-scheduling schemes have been pro...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | en_US |
Published: |
2009
|
Online Access: | http://ndltd.ncl.edu.tw/handle/34506415836160772354 |
id |
ndltd-TW-097THU00392007 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-097THU003920072015-11-20T04:18:28Z http://ndltd.ncl.edu.tw/handle/34506415836160772354 A Performance-based Parallel Loop Self-Scheduling Using Hybrid OpenMP and MPI Programming on Multicore SMP Clusters 一個在多核對稱式處理器叢集上使用混合OpenMP與MPI程式以效能為準的平行迴圈自我排程法 Jen-Hsiang Chang 張壬相 碩士 東海大學 資訊工程與科學系碩士在職專班 97 Parallel loop self-scheduling on parallel and distributed systems has been a critical problem and it is becoming more difficult to deal with in the emerging heterogeneous cluster computing environments. In the past, some self-scheduling schemes have been proposed as applicable to heterogeneous cluster computing environments. In recent years, multicore computers have been widely included in cluster systems. However, previous researches on parallel loop self-scheduling did not consider the features of multicore computers. It is more suitable for shared-memory multiprocessors to adopt OpenMP for parallel programming. In this paper, we propose a performance-based approach, by using hybrid OpenMP and MPI parallel programming, that partitions loop iterations according to the performance weighting of multicore nodes in a cluster. Because the iterations assigned to one MPI process will be processed in parallel by OpenMP threads running by the processor cores in the same computational node, the number of loop iterations to be allocated to one computational node at each scheduling step also depends on the number of processor cores in that node. Experimental results show that the proposed approach performs better than previous schemes. Chao-Tung Yang 楊朝棟 2009 學位論文 ; thesis 60 en_US |
collection |
NDLTD |
language |
en_US |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 東海大學 === 資訊工程與科學系碩士在職專班 === 97 === Parallel loop self-scheduling on parallel and distributed systems has been a critical problem and it is becoming more difficult to deal with in the emerging heterogeneous cluster computing environments. In the past, some self-scheduling schemes have been proposed as applicable to heterogeneous cluster computing environments. In recent years, multicore computers have been widely included in cluster systems. However, previous researches on parallel loop self-scheduling did not consider the features of multicore computers. It is more suitable for shared-memory multiprocessors to adopt OpenMP for parallel programming. In this paper, we propose a performance-based approach, by using hybrid OpenMP and MPI parallel programming, that partitions loop iterations according to the performance weighting of multicore nodes in a cluster. Because the iterations assigned to one MPI process will be processed in parallel by OpenMP threads running by the processor cores in the same computational node, the number of loop iterations to be allocated to one computational node at each scheduling step also depends on the number of processor cores in that node. Experimental results show that the proposed approach performs better than previous schemes.
|
author2 |
Chao-Tung Yang |
author_facet |
Chao-Tung Yang Jen-Hsiang Chang 張壬相 |
author |
Jen-Hsiang Chang 張壬相 |
spellingShingle |
Jen-Hsiang Chang 張壬相 A Performance-based Parallel Loop Self-Scheduling Using Hybrid OpenMP and MPI Programming on Multicore SMP Clusters |
author_sort |
Jen-Hsiang Chang |
title |
A Performance-based Parallel Loop Self-Scheduling Using Hybrid OpenMP and MPI Programming on Multicore SMP Clusters |
title_short |
A Performance-based Parallel Loop Self-Scheduling Using Hybrid OpenMP and MPI Programming on Multicore SMP Clusters |
title_full |
A Performance-based Parallel Loop Self-Scheduling Using Hybrid OpenMP and MPI Programming on Multicore SMP Clusters |
title_fullStr |
A Performance-based Parallel Loop Self-Scheduling Using Hybrid OpenMP and MPI Programming on Multicore SMP Clusters |
title_full_unstemmed |
A Performance-based Parallel Loop Self-Scheduling Using Hybrid OpenMP and MPI Programming on Multicore SMP Clusters |
title_sort |
performance-based parallel loop self-scheduling using hybrid openmp and mpi programming on multicore smp clusters |
publishDate |
2009 |
url |
http://ndltd.ncl.edu.tw/handle/34506415836160772354 |
work_keys_str_mv |
AT jenhsiangchang aperformancebasedparallelloopselfschedulingusinghybridopenmpandmpiprogrammingonmulticoresmpclusters AT zhāngrénxiāng aperformancebasedparallelloopselfschedulingusinghybridopenmpandmpiprogrammingonmulticoresmpclusters AT jenhsiangchang yīgèzàiduōhéduìchēngshìchùlǐqìcóngjíshàngshǐyònghùnhéopenmpyǔmpichéngshìyǐxiàonéngwèizhǔndepíngxínghuíquānzìwǒpáichéngfǎ AT zhāngrénxiāng yīgèzàiduōhéduìchēngshìchùlǐqìcóngjíshàngshǐyònghùnhéopenmpyǔmpichéngshìyǐxiàonéngwèizhǔndepíngxínghuíquānzìwǒpáichéngfǎ AT jenhsiangchang performancebasedparallelloopselfschedulingusinghybridopenmpandmpiprogrammingonmulticoresmpclusters AT zhāngrénxiāng performancebasedparallelloopselfschedulingusinghybridopenmpandmpiprogrammingonmulticoresmpclusters |
_version_ |
1718132119488692224 |