Agent-Based Adaptive Resource Allocation on the Cloud Computing Environment.

Since both consumers and data centers of a cloud service provider can be distributed in geographically, the provider needs to allocate each consumer request to an appropriate data center among the distributed data centers, so that the consumers can satisfy with the service in terms of fast allocation time and execution response time. In this paper, we propose an adaptive resource allocation model that allocates the consumer's job to an appropriate data center. The method to adaptively find a proper data center is based on two evaluations: 1) the geographical distance (network delay)between a consumer and data centers, and 2) the workload of each data center. The proposed model is implemented in an agent based test bed. The test bed simulates a cloud computing environment adopting the proposed adaptive resource allocation model. Empirical results were obtained from simulations using the test bed. The results suggest that the proposed model can successfully allocate consumers' requests to the data center closest to each consumer. Also, the proposed model shows a better response time for allocation than related resource allocation models.