Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/207463
Title: Multi Objective Optimization Oriented Policy for Performance and Energy Efficient Resource Allocation in Cloud Environment
Researcher: Shrimali Bela
Guide(s): Bhadka H.B., Patel Hiren B.
University: C.U. Shah University
Completed Date: 2018
Abstract: Cloud computing is a hybrid paradigm which makes use of utility computing, high newlineperformance cluster computing and grid computing that offers various benefits newlinesuch as flexibility, expandability, little or almost no capital investment, disaster newlinerecovery, moveable work space and much more. However, due to constantly newlineincreasing number of data centers worldwide, the issue of energy consumption by newlinethese data centers has attracted attention of researchers. newlineResource allocation and resource utilization are the major criterion in which the newlineproblem of energy efficiency can be addressed. In this research, we aim to provide newlinean energy efficient resource allocation using Multi-Objective Optimization newline(MOO) technique. We propose MOO-based virtual machine (VM) allocation newlinepolicy and implement it in CloudSim simulation environment. newlineThe results are compared with the existing policies. The results depict that MOObased newlinepolicy leads to energy saving due to an efficient resource allocation, without newlinecompromising performance of data center operations. Moreover, MOO-based newlinepolicy uses weighted sum method in which coefficient is attached with each of the newlineobjectives as a user s preference to decide a priority of objective. Genetic newlinealgorithm and fuzzy logic are used to calculate the co-efficient to generate pareto newlineoptimal solutions. In our research, we use fuzzy logic to generate the random newlinevalue of objectives co-efficient. The proposed fuzzy based computing is newlineimplemented in Cloudsim. The experimental results show that the proposed newlinescheme efficiently generates a random coefficient that assigns priority by newlineconsidering characteristics of host. Further, it demonstrates that the generated newlineweights give pareto optimum solutions that point to strict pareto curve leading to newlineexploration of more optimal solutions in MOO. newline
Pagination: 171p.
URI: http://hdl.handle.net/10603/207463
Appears in Departments:Department of Computer Engineering

Files in This Item:
File Description SizeFormat 
certificate.pdfAttached File395.76 kBAdobe PDFView/Open
chapter 1.pdf492.88 kBAdobe PDFView/Open
chapter 2.pdf823 kBAdobe PDFView/Open
chapter 3.pdf976.45 kBAdobe PDFView/Open
chapter 4.pdf582.21 kBAdobe PDFView/Open
chapter 5.pdf935.23 kBAdobe PDFView/Open
chapter 6.pdf950.5 kBAdobe PDFView/Open
chapter 7.pdf558.3 kBAdobe PDFView/Open
original preliminaray.pdf878.39 kBAdobe PDFView/Open
title page.pdf389.15 kBAdobe PDFView/Open


Items in Shodhganga are protected by copyright, with all rights reserved, unless otherwise indicated.