Document Abstract
This paper presents the results of a study of a heterogeneous computational grid using different scheduling algorithms. After a definition of several algorithms based on the concept of work completion, efficiency is discussed; a new algorithm in grid computing is presented. Two well known scheduling algorithms and our own algorithm are then compared against each other, and it is proved that our algorithm has the highest computing efficiency performer.
In a shared environment, Good schedules involve the integration of specific information. The application performance may suffer when some resources represent abnormal usage pattern during applications execution. We study a performance-prediction based task scheduling system, which provides scheduling based on system-level performance prediction. The efforts to construct a grid computing environment have brought unprecedented computing capacity. Exploiting this complex infrastructure requires efficient middleware to support the execution of a distributed application, composed of a set of subtasks, for best performance. This presents the challenge how to schedule these subtasks in shared heterogeneous systems. For, this we develop our own algorithm whose computational efficiency is very less than the existing algorithm