A Computing Resource Selection Approach Based on Genetic Algorithm for Inter-Cloud Workload Migration

Cloud computing has been one of the most important topics in IT which aims to assure scalable and reliable on-demand services over the Internet. The expansion of the application scope of cloud services would require cooperation between clouds from different providers that have heterogeneous functionalities. However, current cloud systems do not fully support inter-cloud interoperability and require more research work to provide sufficient functions to enable that seamless collaboration between cloud services. This paper proposes an efficient model for selecting appropriate computing resource from multi-cloud providers that is required to achieve inter-cloud interoperability in a heterogeneous Infrastructure as a Service (IaaS) cloud environment. The goal of the model is dispatching the workload on the most effective clouds available at runtime offering the best performance at the least cost. We consider that each job can have six requirements: CPU, memory, network bandwidth, serving time, maximum possible waiting time, and the priority based on the agreed Service Level Agreement (SLA) contract and service price. Additionally, we assume the SLA contract with suitable criteria between cloud-subscriber and multiple IaaS cloud-providers is signed beforehand. This computing resource selection model is based on Genetic Algorithm (GA). The resource selection model is evaluated using agent based model simulation.