LEARNING AUTOMATA BASED METHOD FOR GRID COMPUTING RESOURCE VALUATION WITH RESOURCE SUITABILITY CRITERIA

Abstract In economic Grid environments, the producers (resource owners) and consumers (resource users) have different goals, objectives, strategies, and supply-and-demand patterns. Mechanism based on economic models is an effective approach to solve the problem of grid resources management. Grid resource valuation and allocation is one of the fundamental problems in grid resource management. The essence of this problem is how to allocate and valuation resources for achieving the goal of a highly efficient utilization of resources in response to current resource valuations. Pricing policies are based on the demand from the users and the supply of resources is the main driver in the competitive, economic market model. In this paper, we present a new method of resource allocation and valuation based on the learning automata algorithms in order to maximize the benefit for both grid providers and grid users. We formulate the problem as an environment that learning automata's allocate best resource based on its complete time for proffered application. After allocate of resource, valuation of it based on its complete time is done. With this method the valuation of resource is based on their suitability for jobs execution. Using computer simulations, it is shown that the proposed methodology have higher performance in comparing with existing methods