Reliability Optimization for Distributed Systems through Task Clustering

Abstract

A distributed system consists of a collection of autonomous computers linked by a computer network and equipped with distributed system software. Reliability optimization has been an area of research in distributed systems for years, and the reliability based task allocation problems in distributed systems have been studied by various researchers. Allocation of tasks to available processors using clustering is a commonly used approach in this context. Through traditional task clustering methods, optimal value of the measure of effectiveness is obtained but the process is lengthy and time consuming. Thus, a need has been felt to develop an efficient method to reduce the no. Of iterations to reduce the run time complexity of the algorithm for the distributed systems. In the paper, we propose a method that addresses the problem of assigning a set of "m" tasks of a program to a set of "n" processors (m > n) in a Distributed Computing Environment with the goal to maximize the overall reliability of the system.