S Ghasemi - Department of Engineering, Sepidan Branch, Islamic Azad University, Sepidan, Iran

Resource provisioning in cloud computing as an important challenge in this new information technology paradigm has been recently attracted researchers of the field. There are many cloud providers who offer different services in form of applications to end-users all around the world. Different resources are required for deployment of these services. Developing effective resource provisioning schemes improves functionality of providers; genetic algorithms (GAs) which can consider numerous states simultaneously while investigating their fitness are applied in our study to provision the resources. Our proposed approach, called GARP, finds the exact virtual resources for hosting requests of demanded applications by keeping costs at minimum; previous approaches focused on workloads and minimizing their make spans. The approach helps cloud providers select the most proper computing virtual resources for hosting the demanded applications while considering both cost and mapping of requirements of the applications. Results show the efficiency of each approach in comparison with other approaches.

cloud computing, cost, genetic algorithms, resource provisioning, virtual machine.