Virtual Machine-Based Task Scheduling Algorithm in a Cloud

ABSTRACT:

Virtualization technology has been widely used to virtualize single server into multiple servers, which not only creates an operating environment for a virtual machine-based cloud computing platform but also potentially improves its efficiency. The principle of virtualization technology is to virtualize computer hardware to run multiple independent operating systems in the same hardware environment. Consequently, each operating systemcan run multiple applications simultaneously inindependent physical spaces, which significantlyimproves the efficiency of the cloud computingplatform. As cloud computing environments need to scaleto a large number of users and tasks, designing ascheduling algorithm that can efficiently distribute the tasks and resources becomes a key point for research. This paper introduces a Greedy Particle SwarmOptimization (G&PSO) based algorithm to solve the task scheduling problem. It uses a greedy algorithm to quicklysolve the initial particle value of a particle swarm optimization algorithm derived from a virtual machine-basedcloud platform.

EXISTING SYSTEM

Currently, most task scheduling-based algorithms used in cloud computing environmentsare slow to convergence or easily fall into a local optimum. Existing task-scheduling algorithms aimed at thecloud platform are achieved by large-scale serverclusters and virtual machine clusters. Therefore,a highly-efficient virtual machine task-scheduling algorithm is required to improve the overall efficiencyand operation cost of such a cloud platform.

DISADVANTAGES:

  • Work Load on the machines.
  • Less Efficiency.
  • Poor Performance.

PROPOSED SYSTEM:

This paper introduces a Greedy Particle SwarmOptimization (G&PSO) based algorithm to solve the task scheduling problem. It uses a greedy algorithm to quicklysolve the initial particle value of a particle swarm optimization algorithm derived from a virtual machine-basedcloud platform. The archived experimental results show that the algorithm exhibits better performance such asa faster convergence rate, stronger local and global search capabilities, and a more balanced workload on eachvirtual machine. Therefore, the G&PSO algorithm demonstrates improved virtual machine efficiency and resourceutilization compared with the traditional particle swarm optimization algorithm. In this paper, the proposedcloud computing task-scheduling algorithm virtualizesa single server into multiple virtual machines, thenassigns T independent tasks toM heterogeneous virtualmachines for execution (i.e., one task cannot berun on two virtual machines, each virtual machinecan only handle one task at one time and each hasdifferent properties), thus minimizing the time requiredto complete all the tasks. Within a cloudenvironment deployed by a single server, using theproposed algorithm will not only reduce the total taskcompletion time but also will balance the system loadand improve the efficiency of task scheduling andresource utilization of the cloud computing platform.

ADVANTAGES:

  • Balanced Work Load.
  • Improves efficiency.
  • Best Performance.
  • Reduces Business cost.

SYSTEM REQUIREMENTS

H/W System Configuration:-

Processor - Pentium –III

RAM - 256 MB (min)

Hard Disk - 20 GB

Key Board - Standard Windows Keyboard

Mouse - Two or Three Button Mouse

Monitor - SVGA

S/W System Configuration:-

Operating System : Windows95/98/2000/XP

Application Server : Tomcat5.0/6.X

Front End : HTML, Jsp

Scripts : JavaScript.

Server side Script : Java Server Pages.

Database : MySQL 5.0

Database Connectivity : JDBC

Further Details Contact: A Vinay 9030333433, 08772261612, 9014123891 #301, 303 & 304, 3rd Floor,

AVR Buildings, Opp to SV Music College, Balaji Colony, Tirupati - 515702 Email:

|