Open Access Journal

ISSN : 2394-2320 (Online)

International Journal of Engineering Research in Computer Science and Engineering (IJERCSE)

Monthly Journal for Computer Science and Engineering

Open Access Journal

International Journal of Engineering Research in Computer Science and Engineering (IJERCSE)

Monthly Journal for Computer Science and Engineering

ISSN : 2394-2320 (Online)

A Novel Study on Pre-copy Method for Classification of Live Migration of Virtual Machines in Cloud Computing

Author : Minal Patel 1

Date of Publication :22nd February 2018

Abstract: Cloud computing is the framework of recent technology and it can set the vision of computing utilities into a reality. The most promising part of cloud computing is that it provides dynamic provisioning and delivers facility for computation, networking, storage etc using datacenters. The working of cloud resources provided through the Internet and it gives services on a pay-per-use basis from different cloud vendors. IaaS services used to develop background processing, PaaS provides programming platforms for developing applications and SaaS provides benefit to cloud users from the elastic scalability without making processes such as software development, installation, configuration, maintenance etc. The migration of a virtual machine has advantages such as increasing energy efficiency, resource sharing and load balancing, fault resilience etc. The pre-copy algorithm of Xen hypervisor is very well known and it is also used in real applications to run migration activities. The migration is categorized into three major types to improve pre-copy: i) compressing memory pages ii) predicting dirty pages iii) reducing dirty pages (performance modelling). Each category has its own importance for the live migration. At first, the compression model is proposed to achieve efficient virtual machine migration. Time series based techniques are developed using historical analysis of past data in the second category. The time series is generated with consecutive iterations of memory pages. Here, two different regression-based models of time series are proposed. In the third category, the performance modelling of live migration is proposed by the measured approach. These models are evaluated for measuring the computation of virtual machine (VM) downtime, the total number of pages transferred and total migration time

Reference :

Will Updated soon

Recent Article