Optimizing management of cloud resources towards best performance for applications execution

Cloud computing is considered as the sixth generation of computing. It is a kind of distributed and parallel systems based on a pay-as-you-go model that offers computing resources to users. These resources are used in such manner to fulfill workload demands of customers. However, one of the main challenges in cloud computing is how to manage in an optimal way the cloud resources in response of execution needs; so that we can reach the two performance objectives that we are targeting: reducing both the cost and the time of running users applications. In this paper, we propose an optimization-based solution which aims to achieve the aforementioned objectives. First, we improved an existing mathematical model based on integer programming formulation. Then, we proposed a metaheuristic solution based on genetic algorithms. To check effectiveness of our solution, we tested it over some real applications using a set of cloud resources available on the market. The obtained results demonstrate clearly that our algorithms succeeded to find reasonable solutions that ensure a fair dealing with each objective function.