Market Mechanism for Dynamic Resource Management in Computational Grid

Abstract

This paper presents a market mechanism for dynamic resource allocation in computational grid. Grid market is described that consists of two economic agent types; it allows agents representing various grid resources to coordinate their resource allocation decisions without assuming a priori cooperation. The grid task agents buy resources to complete tasks. Grid resource agents charge the task agents for the amount of resource capacity allocated. Grid resource allocation problem is presented as grid user utility optimization. Given grid resource agent's pricing policy, the task agent optimization problem is to complete its job as quickly as possible when spending the least possible amount of money. This paper provides a resource allocation and pricing algorithm. Experiments are made to compare the performance of the price-directed resource allocation with conventional Round-Robin allocation.