Case Study: Computational Grid Computing

Abstract

Grid computing is a next-generation high performance computing platform to solve complex and large-scale scientific problems. Computational Grid has emerged as a future generation computing environment which is a collection of heterogeneous computing resources connected by a network across dynamic and geographically distributed organizations, to form a distributed high performance computing infrastructure. Job scheduling in computational Grid is a vital and challenging work. The primary objective of scheduling is to maximize the resource utilization and to minimize processing time of the jobs. In this paper, an improved cuckoo search (ICS) optimization method has been proposed for scheduling user-application components to available

A computational Grid environment behaves like a virtual organization consisting of distributed resources. A virtual organization is a group of individuals and institutions defined by a definite set of sharing rules like what is shared, who is allowed to share, and the conditions under which the sharing takes place [2]. As Grid computing focuses on dynamic and inter-organizational sharing, it enhances the existing distributed computing technologies. In most organizations, there are large amount of under-utilized computing resources. Considering a normal employee works 7 hours a day, 42 hours a week and there are 168 hours per week, most of the desktop machines remain busy less than 25% of the time. Even the server machines can often be fairly idle. Grid computing provides a framework for exploiting these under-utilized resources and thus has the possibility of substantially increasing the efficiency of resource usages. The easiest use of Grid computing would be to run an existing application on several machines. The machine on which the application is normally run might be unusually busy, resulting the delay in the execution of the job. Grid computing will enable the job to be run on an idle machine

In this paper, it is assumed that a job (or task) represents a computational unit (typically a program and possibly associated data) to run on a Grid node (or resource) which is a basic computational entity (computational device like processor or service) where jobs are scheduled, allocated and processed accordingly.

Job scheduling is an integrated mechanism of Grid computing which effectively utilizes the idle time of resources. Efficient scheduling algorithm is needed to utilize the resources effectively and reduce the overall completion time. Because of the heterogeneous and the dynamic nature of the Grid, scheduling in Grid environment is significantly complicated. So job scheduling which is one of the NP-Complete problems becomes a focus of researchers in Grid computing area.

A. Resource