Improving SLA-based job scheduling in economic Grid Using Heuristics

Abstract

In shared heterogeneous environment of Grid that jobs compete for the best QoS, selecting appropriate resources to run a job efficiently is a complex responsibility of a scheduler. Schedulers today establish a SLA between two participants to meet the considered QoS. SLA-based job scheduling and resource allocation are based on the concept of agreement between supplier and consumer. In addition, heuristic techniques can help the scheduler to select the best SLA. Therefore, this paper presents a resource scheduling approach called HPESA, Heuristic and Priority based Economic Scheduling Approach, to improve SLA-based job scheduling in economic Grid. The main purpose of this study is the evaluation of some simple heuristics to enhance the performance, utilization, and the total earnings of the resources as well as to reduce the average response time, average budget spent and the number of messages exchanged. This article also intends to enhance SLA-based contract negotiation between supplier and consumer and job migration conditions. Our simulations on GridSim toolkit explain the usefulness of this approach in providing QoS to the users and its impact on scheduling in economic Grid.