Performance Analysis and Comparison of Parallel Computing Frameworks: Sips and Openmp

2019 International Conference on Intelligent Computing and Control Systems (ICCS) Pub Date : 2019-05-01 DOI:10.1109/ICCS45141.2019.9065343

Likhilpreet Kaur, Anil Kumar

{"title":"Performance Analysis and Comparison of Parallel Computing Frameworks: Sips and Openmp","authors":"Likhilpreet Kaur, Anil Kumar","doi":"10.1109/ICCS45141.2019.9065343","DOIUrl":null,"url":null,"abstract":"Parallel execution of jobs progresses towards becoming need of great importance as holding up time required to be decreased. To this perspective a few stages have been worked upon by the specialists. Jobs started by the clients are sequential in nature yet execution of jobs required parallel stage for better execution regarding complete time and holding up time. This paper proposed the execution correlation of two of the most recent stages utilized for parallel employment execution. First stage considered for assessment is SIPS and second stage is OpenMP. Sequential calculation in parallel framework (SIPS) stage decreases the system overhead and its execution should be possible both in Windows as well as UNIX condition. The proposed work is completed in Windows condition, Matrix and Pie calculation parallel jobs are displayed. In the second case OpenMP stage is utilized to execute Matrix and dart programs. The parameters considered for assessment incorporates organize overhead, execution time and scheduling overhead. Correlation result recommends SIPS/OpenMP performs essentially better as far as recorded parameters.","PeriodicalId":433980,"journal":{"name":"2019 International Conference on Intelligent Computing and Control Systems (ICCS)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 International Conference on Intelligent Computing and Control Systems (ICCS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCS45141.2019.9065343","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

Abstract

Parallel execution of jobs progresses towards becoming need of great importance as holding up time required to be decreased. To this perspective a few stages have been worked upon by the specialists. Jobs started by the clients are sequential in nature yet execution of jobs required parallel stage for better execution regarding complete time and holding up time. This paper proposed the execution correlation of two of the most recent stages utilized for parallel employment execution. First stage considered for assessment is SIPS and second stage is OpenMP. Sequential calculation in parallel framework (SIPS) stage decreases the system overhead and its execution should be possible both in Windows as well as UNIX condition. The proposed work is completed in Windows condition, Matrix and Pie calculation parallel jobs are displayed. In the second case OpenMP stage is utilized to execute Matrix and dart programs. The parameters considered for assessment incorporates organize overhead, execution time and scheduling overhead. Correlation result recommends SIPS/OpenMP performs essentially better as far as recorded parameters.

查看原文本刊更多论文

并行计算框架的性能分析与比较:Sips和Openmp

作业的并行执行逐渐成为非常重要的需求，因为所需的延迟时间将会减少。从这个角度来看，专家们已经研究了几个阶段。客户端启动的作业本质上是顺序的，但作业的执行需要并行阶段，以便在完成时间和保持时间方面更好地执行。本文提出了用于并行就业执行的两个最新阶段的执行相关性。第一阶段是SIPS，第二阶段是OpenMP。并行框架(SIPS)阶段的顺序计算减少了系统开销，并且在Windows和UNIX条件下都可以执行。所提出的工作在Windows环境下完成，显示了矩阵和饼计算并行作业。在第二种情况下，OpenMP阶段用于执行矩阵和dart程序。评估所考虑的参数包括组织开销、执行时间和调度开销。相关结果表明，就记录的参数而言，SIPS/OpenMP的性能基本上更好。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

2019 International Conference on Intelligent Computing and Control Systems (ICCS)

自引率

0.00%

发文量