New Algorithm for Resource Selection in Economic Grid with the Aim of Cost Optimization Using Learning Automata

In economic computational grids, resources have prices and the users must pay for executing their applications. The user determines his deadline and budget and then requests cost or time optimization. A resource selection service that adopts cost optimization strategy should select heterogeneous grid resources for heterogeneous user applications so that their execution finishes in the specified deadline with minimum cost. In this paper, new algorithms based on learning automata are proposed for this purpose. Using computer simulations, it is shown that the proposed algorithms have higher performance comparing to the existing algorithm.