期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2018
卷号：7
期号：10
页码：10828-10839
DOI：10.15680/IJIRSET.2018.0710013
出版社：S&S Publications
摘要：Actually Analyzing and processing the SPARK in timely and cost effective manner is very prominent
approchement and its very hectic job. And as usually the SPARK makes the huge data to understand in easy way that
can be happened (visualized) in a snap deal. In the market so many tools available for making explored such things, but
one of the best tool is ‘SPARK’ which makes the data to be adopted for repository and managing the things of
‘BIGDATA’. So many methods were suggested and developed for analyzing and augmenting ‘SPARK’
accomplishment (performance).In my paper piercely being focused on tuning configuration parameter approachement.
Even though several parameters those affect the performance of ‘SPARK’, in which map reduced related parameters
made on effective impact. The main aim of this work to raise the overall performance of ‘SPARK’ by reducing the job
execution time. Reduction in time is attained through tuning some of map reduce associated parameters. To get
understood these parameters is paramount as different types of parameter with clumsy(improper) values which shows
negative input upon overall performance. In this paper what we proposed method is that saves the job execution time
and optimizes disk usage properly and effectively. It exactly improves the overall performance of the ‘SPARK’ by
38.53% over the base system in heterogeneous environment.
