يك روش تركيبي با استفاده از الگوريتم‌هاي فاخته و جستجوي ممنوعه براي كاهش مصرف انرژي در محاسبات ابري

A hybrid energy efficient approach using Cuckoo Optimization and Tabu Search algorithms in cloud computing

Consolidating the VMs is an effective way to reduce energy consumption and costs and to maximize resource utilization while providing negotiable Quality Of Services (QOS). This can be achieved by dynamic migration and switching the idle physical machines to sleep or hibernate. this power switching status mode should not cause SLA violation as another important metric in cloud computing environments. So there should be a trade-off between minimizing the energy consumption and SLA violation. In this paper we have presented a hybrid approach using Cuckoo Optimization Algorithm (COA) for detecting over-utilized hosts and a Tabu Search (TS) algorithm to select the best node for migration. We have also used usual Minimum Migration Time (MMT) for VM selection policy. We used COA for its convergence speed and good performance in solving bin packing problems And TS because of running away from local optima and detecting the best host node. The simulation result which is generated by CloudSim simulator, demonstrates that our approach overtakes other famous energy aware algorithms and other cuckoo approaches in power consumption.

Consolidating the VMs is an effective way to reduce energy consumption and costs and to maximize resource utilization while providing negotiable Quality Of Services (QOS). This can be achieved by dynamic migration and switching the idle physical machines to sleep or hibernate. this power switching status mode should not cause SLA violation as another important metric in cloud computing environments. So there should be a trade-off between minimizing the energy consumption and SLA violation. In this paper we have presented a hybrid approach using Cuckoo Optimization Algorithm (COA) for detecting over-utilized hosts and a Tabu Search (TS) algorithm to select the best node for migration. We have also used usual Minimum Migration Time (MMT) for VM selection policy. We used COA for its convergence speed and good performance in solving bin packing problems And TS because of running away from local optima and detecting the best host node. The simulation result which is generated by CloudSim simulator, demonstrates that our approach overtakes other famous energy aware algorithms and other cuckoo approaches in power consumption.