Dynamic virtual machine allocation policy for load balancing using principal component analysis and clustering technique in cloud computing

Research output: Contribution to journalArticle

Abstract

The scalability and agility characteristics of cloudcomputing allow load balancing to reroute workload requestseasily and to enhance overall accessibility. One of the mostimportant services for cloud computing is Infrastructure as aService (IaaS). There is a large number of physical hosts in acloud data center for IaaS and it is quite difficult to arrange theallocation of the workload requests manually. Therefore,different load balancing methods have been proposed byresearchers to avoid overloaded physical hosts in the cloud datacenter. However, fewer works have used multivariate analysis incloud computing environment for considering the dynamicchanges of the computing resources. Thus, this work suggests anew Virtual Machine (VM) allocation policy for load balancingby using a multivariate technique, Principal ComponentAnalysis (PCA), and clustering technique. Moreover, PCA andclustering techniques were simulated on a cloud computingsimulator, CloudSim. In the proposed allocation policy, a groupof VMs were dynamically allocated to physical hosts. Theallocation was based on the clusters of hosts according to theirsimilar features in computing resources. The clusters wereformed using PCA and a clustering technique based on variablesrelated to the physical hosts such as Million Instructions PerSecond (MIPS), Random Access Memory (RAM), bandwidthand storage. The results show that the completion time for alltasks has decreased, and the resource utilization has increased.This will optimize the performance of cloud data centers byeffectively utilizing the available resources.

Original languageEnglish
Pages (from-to)47-52
Number of pages6
JournalJournal of Telecommunication, Electronic and Computer Engineering
Volume10
Issue number3-2
Publication statusPublished - 1 Jan 2018

Fingerprint

Cloud computing
Principal component analysis
Resource allocation
Scalability
Data storage equipment
Virtual machine
Multivariate Analysis

Keywords

  • Allocation Policy
  • Cloud Computing
  • CloudSim
  • Clustering
  • PCA Technique
  • Virtual Machines

ASJC Scopus subject areas

  • Hardware and Architecture
  • Computer Networks and Communications
  • Electrical and Electronic Engineering

Cite this

@article{703eb9b201b8443fb2c8b5d8bc3b1f87,
title = "Dynamic virtual machine allocation policy for load balancing using principal component analysis and clustering technique in cloud computing",
abstract = "The scalability and agility characteristics of cloudcomputing allow load balancing to reroute workload requestseasily and to enhance overall accessibility. One of the mostimportant services for cloud computing is Infrastructure as aService (IaaS). There is a large number of physical hosts in acloud data center for IaaS and it is quite difficult to arrange theallocation of the workload requests manually. Therefore,different load balancing methods have been proposed byresearchers to avoid overloaded physical hosts in the cloud datacenter. However, fewer works have used multivariate analysis incloud computing environment for considering the dynamicchanges of the computing resources. Thus, this work suggests anew Virtual Machine (VM) allocation policy for load balancingby using a multivariate technique, Principal ComponentAnalysis (PCA), and clustering technique. Moreover, PCA andclustering techniques were simulated on a cloud computingsimulator, CloudSim. In the proposed allocation policy, a groupof VMs were dynamically allocated to physical hosts. Theallocation was based on the clusters of hosts according to theirsimilar features in computing resources. The clusters wereformed using PCA and a clustering technique based on variablesrelated to the physical hosts such as Million Instructions PerSecond (MIPS), Random Access Memory (RAM), bandwidthand storage. The results show that the completion time for alltasks has decreased, and the resource utilization has increased.This will optimize the performance of cloud data centers byeffectively utilizing the available resources.",
keywords = "Allocation Policy, Cloud Computing, CloudSim, Clustering, PCA Technique, Virtual Machines",
author = "Xue, {Law Siew} and {Abd Majid}, {Nazatul Aini} and {A Sundararajan}, Elankovan",
year = "2018",
month = "1",
day = "1",
language = "English",
volume = "10",
pages = "47--52",
journal = "Journal of Telecommunication, Electronic and Computer Engineering",
issn = "2180-1843",
publisher = "Universiti Teknikal Malaysia Melaka",
number = "3-2",

}

TY - JOUR

T1 - Dynamic virtual machine allocation policy for load balancing using principal component analysis and clustering technique in cloud computing

AU - Xue, Law Siew

AU - Abd Majid, Nazatul Aini

AU - A Sundararajan, Elankovan

PY - 2018/1/1

Y1 - 2018/1/1

N2 - The scalability and agility characteristics of cloudcomputing allow load balancing to reroute workload requestseasily and to enhance overall accessibility. One of the mostimportant services for cloud computing is Infrastructure as aService (IaaS). There is a large number of physical hosts in acloud data center for IaaS and it is quite difficult to arrange theallocation of the workload requests manually. Therefore,different load balancing methods have been proposed byresearchers to avoid overloaded physical hosts in the cloud datacenter. However, fewer works have used multivariate analysis incloud computing environment for considering the dynamicchanges of the computing resources. Thus, this work suggests anew Virtual Machine (VM) allocation policy for load balancingby using a multivariate technique, Principal ComponentAnalysis (PCA), and clustering technique. Moreover, PCA andclustering techniques were simulated on a cloud computingsimulator, CloudSim. In the proposed allocation policy, a groupof VMs were dynamically allocated to physical hosts. Theallocation was based on the clusters of hosts according to theirsimilar features in computing resources. The clusters wereformed using PCA and a clustering technique based on variablesrelated to the physical hosts such as Million Instructions PerSecond (MIPS), Random Access Memory (RAM), bandwidthand storage. The results show that the completion time for alltasks has decreased, and the resource utilization has increased.This will optimize the performance of cloud data centers byeffectively utilizing the available resources.

AB - The scalability and agility characteristics of cloudcomputing allow load balancing to reroute workload requestseasily and to enhance overall accessibility. One of the mostimportant services for cloud computing is Infrastructure as aService (IaaS). There is a large number of physical hosts in acloud data center for IaaS and it is quite difficult to arrange theallocation of the workload requests manually. Therefore,different load balancing methods have been proposed byresearchers to avoid overloaded physical hosts in the cloud datacenter. However, fewer works have used multivariate analysis incloud computing environment for considering the dynamicchanges of the computing resources. Thus, this work suggests anew Virtual Machine (VM) allocation policy for load balancingby using a multivariate technique, Principal ComponentAnalysis (PCA), and clustering technique. Moreover, PCA andclustering techniques were simulated on a cloud computingsimulator, CloudSim. In the proposed allocation policy, a groupof VMs were dynamically allocated to physical hosts. Theallocation was based on the clusters of hosts according to theirsimilar features in computing resources. The clusters wereformed using PCA and a clustering technique based on variablesrelated to the physical hosts such as Million Instructions PerSecond (MIPS), Random Access Memory (RAM), bandwidthand storage. The results show that the completion time for alltasks has decreased, and the resource utilization has increased.This will optimize the performance of cloud data centers byeffectively utilizing the available resources.

KW - Allocation Policy

KW - Cloud Computing

KW - CloudSim

KW - Clustering

KW - PCA Technique

KW - Virtual Machines

UR - http://www.scopus.com/inward/record.url?scp=85054511606&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054511606&partnerID=8YFLogxK

M3 - Article

VL - 10

SP - 47

EP - 52

JO - Journal of Telecommunication, Electronic and Computer Engineering

JF - Journal of Telecommunication, Electronic and Computer Engineering

SN - 2180-1843

IS - 3-2

ER -