An Adaptive Strategy for Resource Allocation Modeled as Minority Game

In a computer system, different agents need different resources to complete their tasks. The objective of resource allocation is to allocate the resources to the agents so that they can complete their tasks and at the same time make good utilization of resources. In this paper, we model the resource allocation problem as a minority game. In this way, the system has the advantages of a distributed system but with reduced overhead. Also, this approach is autonomous and adaptive. We design an adaptive strategy for agents to play in the game model. We perform simulations of a single-choice model, a multi-choice model in which agents have (and do not have) preferences over the resources. Simulations show that agents with the proposed adaptive strategy are able to make more right decisions and better resource utilization than previous work.