Critical insight for MAPReduce optimization in Hadoop

Khan, Burhan ul Islam and Olanrewaju, Rashidah Funke and Altaf, Hunain and Shah, Asadullah (2014) Critical insight for MAPReduce optimization in Hadoop. International Journal of Computer Science and Control Engineering, 2 (1). pp. 1-7.

PDF Restricted to Repository staff only Download (171kB) | Request a copy

Abstract

In present day scenario cloud has become an inevitable need for majority of IT operational organization s. Cloud applications such as data storage, data retrieval and data portability have become significant requirements for cloud computing. Numerous applications are being developed for BigData. Achieving an optimal approach for higher performance in terms of efficient load balancing, load distribution, optimum resource utilization, minimum overheads and least possible delay has been the vital issue for cloud infrastructure. Apache Hadoop is one the most used cloud frame work for cloud infrastructure. The predominant philosophy behind Hadoop optimization is the optimization of MapReduce, which is a dominant programming platform effective in bringing a=bout many functional enhancements as per scheduling algorithms developed and implemented. MapReduce has emerged as the most significant part of Hadoop system that establishes itself as a framework that can effectively simplify the overall complexity of running parallel data processes across the network of computing nodes. A number of scheduling techniques have been advocated in the last couple of years for achieving enhanced load balancing in Hadoop. Unfortunately Hadoop still lacks a system model that could facilitate an ultimate solution for delivering optimized performance without creating much computational overhead. In order to pave a way for the development of an adept and decisive load balancing and job scheduling scheme for minimum execution time and optimum resource utilization in future, here in this paper a comprehensive review of some of the major works has been done to discuss the prominence of issues, which will be needed to be taken care of while developing the same.

Item Type:	Article (Journal)
Additional Information:	6796/36441
Uncontrolled Keywords:	Bigdata, Mapreduce, Hadoop Optimization, Load Balan cing, QoS
Subjects:	T Technology > T Technology (General) > T10.5 Communication of technical information
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Information and Communication Technology > Department of Information System Kulliyyah of Information and Communication Technology > Department of Information System Kulliyyah of Engineering Kulliyyah of Engineering > Department of Electrical and Computer Engineering Kulliyyah of Information and Communication Technology Kulliyyah of Information and Communication Technology
Depositing User:	Prof Asadullah Shah
Date Deposited:	29 Apr 2014 11:52
Last Modified:	02 Nov 2020 09:45
URI:	http://irep.iium.edu.my/id/eprint/36441

Actions (login required)

View Item