(3.236.86.184)
[ij] [ij] [ij] 
Email id
 

Asian Journal of Research in Social Sciences and Humanities
Year : 2016, Volume : 6, Issue : 5
First page : ( 775) Last page : ( 790)
Online ISSN : 2249-7315.
Article DOI : 10.5958/2249-7315.2016.00152.0

A Hierarchical Multi-Constrained Cost Efficient Qos-Driven Scheduling Heuristic for Computational Tasks in Grid

Keerthika P*, Suresh P**

*Assistant Professor, Department of Computer Science and Engineering, Kongu Engineering College, Perundurai, Erode, Tamil Nadu, India

**Assistant Professor, Department of Information Technology, Kongu Engineering College, Perundurai, Erode, Tamil Nadu, India

Online published on 3 May, 2016.

Abstract

In a Grid environment, a plenty of heterogeneous resources are shared among the users to solve complex problems that requires massive computational capability. In this view, scheduling algorithms plays a vital role in provisioning a suitable resource for executing a complex computational task. Also, the scheduling algorithms should address the issues that arise when an allocated resource fails. Resource allocation is an important phenomenon where several issues should be taken for consideration such as balancing the load of resources, handling resource failure, satisfying the user's demand and finally the cost for task execution. In this paper, we propose a scheduling heuristic (HMC-QoS) that addresses all these scheduling issues and performs better resource allocation. This heuristic works at three different levels of hierarchy and finds an appropriate resource for a task. It considers the factors like execution time, resource load, communication time, cost, and task completion deadline and failure history of resources for scheduling at each level. The proposed HMC-QoS heuristic is implemented and its performance is measured based on various performance metrics which involves makespan, task deadline hit, resource utilization, cost, and communication time and fault tolerance. The analysis shows that the proposed heuristic results in improved makespan and other metrics when compared with other existing algorithms.

Top

Keywords

Fault tolerance, Load Balancing, Grid Scheduling, Meta-task, communication overhead, Resource Utilization, User satisfaction.

Top

  
║ Site map ║ Privacy Policy ║ Copyright ║ Terms & Conditions ║ Page Rank Tool
812,534,620 visitor(s) since 30th May, 2005.
All rights reserved. Site designed and maintained by DIVA ENTERPRISES PVT. LTD..
Note: Please use Internet Explorer (6.0 or above). Some functionalities may not work in other browsers.