A SLA-based Cloud Computing Framework: Workload and Location Aware Resource Allocation to Distributed Data Centers in a Cloud

As the number of users of cloud services increase all over the world, cloud service providers keep deploying geographically distributed data centers more. Since resource capacity of a data center is limited, cloud service providers need to distribute load to data centers. However, when the provider distributes load, SLAs (service level agreements) established with consumers should be guaranteed in an environment where a cloud provider operates geographically distributed data centers. Therefore, this paper proposes a SLA-based cloud computing framework to facilitate temporal and locational load-aware resource allocation. The contributions of this paper include 1) design of a cloud computing framework including an automated SLA negotiation mechanism and a workload and location aware resource allocation (WLARA) and 2) implementation of an agent-based cloud test bed of the proposed framework. Empirical results using the test bed show the proposed WLARA performs better than other related approaches in terms of guaranteed SLAs.