IJSRP, Volume 5, Issue 12, December 2015 Edition [ISSN 2250-3153]
M.Krishna Sudha, Dr. S. Sukumaran
Abstract:
Latest distributed computing paradigm is cloud computing and it offers tremendous opportunities to solve large-scale problems. It is critical to evaluate the performance of cloud environments as the adoption and deployment of cloud computing increases. The number of active machines are matched dynamically to resource demands inorder to increase resource utilization. However based on extensive studies, existing solutions have not fully considered the heterogeneity of both workload and resource utilization found in computing environments. Currently, effective and efficient scheduling algorithms are needed for the better use of tremendous capabilities in large scale distributed system. Many such algorithms have been designed and implemented. New scheduling algorithm based on dynamic conventional scheduling Heuristics is designed to use their cons and at the same time overcome their pros. The cloud simulator Cloudsim is used to evaluate the proposed scheduling algorithm by comparing to its performance against the basic heuristics. Variety of scenarios is taken into consideration to show that the simulation results of the new heuristic algorithm leads to significant performance gain. Maximization in resource utilization and reduction in makespan are the two basic criteria used for the performance evaluation of the proposed Algorithm. The proposed algorithm proves that it is efficient than the other heuristic algorithms by reducing the make span and increasing the resource utilization.