Decoding Parallel Program Execution by using Java Interactive Visualization Environment (JIVE): Behavioral and Performance Analysis

2019 3rd International Conference on Computing Methodologies and Communication (ICCMC) Pub Date : 2019-03-27 DOI:10.1109/ICCMC.2019.8819754

Ajaz Abdul Aziz, Malavika Unny, S. Niranjana, M. Sanjana, J. Swaminathan

{"title":"Decoding Parallel Program Execution by using Java Interactive Visualization Environment (JIVE): Behavioral and Performance Analysis","authors":"Ajaz Abdul Aziz, Malavika Unny, S. Niranjana, M. Sanjana, J. Swaminathan","doi":"10.1109/ICCMC.2019.8819754","DOIUrl":null,"url":null,"abstract":"With the proliferation of multi-core systems in the last decade or so even the personal computers have acquired the capability of supporting parallel programs. However, most applications are simply not designed to take advantage of this capability. This is firstly due to the difficulty in comprehending parallel programs. Secondly, the speed-up achieved due to parallelism is diminished by the overhead incurred. We study both these aspects in the context of fork/join, the parallel programming framework supported by Java and Java Interactive Visualization Environment (JIVE), a dynamic analysis framework for debugging and visualizing Java programs. In this paper, we demonstrate how JIVE can be used to decode parallel program execution and their behavior on single, dual and quad core systems. We also present the results of the performance study undertaken to compare the performance of parallel programs against their sequential and multi-threaded counterparts for small, medium and large sized executions.","PeriodicalId":232624,"journal":{"name":"2019 3rd International Conference on Computing Methodologies and Communication (ICCMC)","volume":"80 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 3rd International Conference on Computing Methodologies and Communication (ICCMC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCMC.2019.8819754","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 5

Abstract

With the proliferation of multi-core systems in the last decade or so even the personal computers have acquired the capability of supporting parallel programs. However, most applications are simply not designed to take advantage of this capability. This is firstly due to the difficulty in comprehending parallel programs. Secondly, the speed-up achieved due to parallelism is diminished by the overhead incurred. We study both these aspects in the context of fork/join, the parallel programming framework supported by Java and Java Interactive Visualization Environment (JIVE), a dynamic analysis framework for debugging and visualizing Java programs. In this paper, we demonstrate how JIVE can be used to decode parallel program execution and their behavior on single, dual and quad core systems. We also present the results of the performance study undertaken to compare the performance of parallel programs against their sequential and multi-threaded counterparts for small, medium and large sized executions.

查看原文本刊更多论文

用Java交互式可视化环境(JIVE)解码并行程序执行:行为与性能分析

随着近十年来多核系统的普及，甚至个人计算机也获得了支持并行程序的能力。然而，大多数应用程序根本就没有设计成利用这种功能。这首先是由于理解并行程序的困难。其次，由于并行性而获得的加速被开销所减少。我们在fork/join (Java支持的并行编程框架)和Java交互式可视化环境(Java Interactive Visualization Environment, JIVE)(用于调试和可视化Java程序的动态分析框架)的背景下研究了这两个方面。在本文中，我们演示了如何使用JIVE解码并行程序执行及其在单核，双核和四核系统上的行为。我们还提供了性能研究的结果，以比较并行程序与顺序程序和多线程程序在小型、中型和大型执行中的性能。

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 3rd International Conference on Computing Methodologies and Communication (ICCMC)

自引率

0.00%

发文量