NexComm 2014
February 23 - 27, 2014
Nice, France

DigitalWorld 2014
March 23 - 27, 2014
Barcelona, Spain

InfoSys 2014
April 20 - 24, 2014
Chamonix, France

BioSciencesWorld 2014
April 20 - 24, 2014
Chamonix, France

ComputationWorld 2014
May 25 - 29, 2014
Venice, Italy

InfoWare 2014
June 22 - 26, 2014
Seville, Spain

DataSys 2014
July 20 - 24, 2014
Paris, France

NexTech 2014
August 24 - 28, 2014
Rome, Italy

SoftNet 2014
October 12 - 16, 2014
Nice, France

NetWare 2014
November 16 - 20, 2014
Lisbon, Portugal


ThinkMind // CLOUD COMPUTING 2010, The First International Conference on Cloud Computing, GRIDs, and Virtualization // View article cloud_computing_2010_1_50_50078

An Efficient Job Scheduling Technique in Trusted Clusters for Load Balancing

Shakti Mishra
Dharmender Singh Kushwaha
Arun Kumar Misra

Keywords: Cluster;Load Balancing; Scheduling; Priority; Multilevel feedback Queue.

Although there has been tremendous increase in PC power and most of it is not fully harnessed, yet certain computation intensive application tend to migrate their process with the aim of reducing the response time. Cluster computing is the area that aims just at this. Clustering provides means to improve availability of services, sharing computational workload and performing computation intensive application. However, these benefits can only be achieved if the computing power of cluster is used efficiently and allocated fairly among all the available nodes. As is the case with our desktops, it is usually seen that clusters also suffer from underutilization. A number of approaches proposed in the past share only idle CPU cycles and not use the resources of systems when the machine has its own local processes to execute. We propose a priority-based scheduling approach for run queue and multilevel feedback queue scheduling approach for migrated tasks that doesn’t degrade the performance of local jobs too. Simulation and experimental results have been able to show that priority-based run queue management and multilevel feedback queue scheduling for migrated tasks can increase overall throughput by about 28-33 percent.

Pages: 26 to 31

Copyright: Copyright (c) IARIA, 2010

Publication date: November 21, 2010

Published in: conference

ISSN: 2308-4294

ISBN: 978-1-61208-106-9

Location: Lisbon, Portugal

Dates: from November 21, 2010 to November 26, 2010

2010 - 2014 © ThinkMind. All rights reserved.
Read Terms of Service and Privacy Policy.