Temperature aware resource scheduling in Green Clouds

Abstract

Green Cloud Computing is a very hot topic now days as it deals with efficient usage of resources. Cloud Computing uses the concept of virtualization for efficient usage of cloud resources. As number of jobs are submitted to virtual machines (VMs) on a physical machine, the temperature of the physical machine increases. This may destroy the system and affects the performance also. It also has negative impact on our environment because of CO2 emissions. So temperature aware resource scheduling can provide green enhancements within cloud computing environment. In this paper, it is proposed that, current and critical temperature of physical machines must be considered before distributing workload among nodes. Lot of research work has already been done on providing better cooling system. But these systems consume a lot of energy in cooling the nodes. This paper proposes the technique for temperature aware resource scheduling and compares it with other scheduling techniques i.e. FCFS and Round-Robin scheduling. Proposed algorithm is better than existing methods as it considers critical temperature of a physical node before making Virtual Machine (VM) allocation decision, thereby reducing the need for cooling systems in a cloud environment.