Resource allocation in grid systems using collective case based reasoning methods and learning automata

At this article, we study the solutions about resources allocation in grid systems by means of synthetic method of collective case based on reasoning and learning automata. The collective case based on reasoning method in relation to agents, solves problems according to main agent method as scheduler agent. In fact, scheduler agent chooses a method within all present methods and sends it to other agents. All other agents use this method to find identical cases. These methods can be homogeneous or inhomogeneous. We try to study the effect of homogeneous or inhomogeneous agents and also using an identical case base or several discrete case bases for final answer so that each base pertains to one of the agents. The agents use ICBR-LA procedure in homogeneous methods and different kinds of CBR procedures like CBR-LA and ICBR-LA and also Max-Min and Min-Min methods in inhomogeneous cases. The scheduler agent chooses a method for solving a problem according to mentioned methods and sends it to other agents.