DESIGNING POWER MANAGEMENT AWARE PROXIMITY BASED TASK SCHEDULING ALGORITHM FOR CLOUD DATA CENTERS

Abstract

Cloud computing is an evolving technology providing computing, communication and storage resources as a service via Internet. In cloud computing environment, many tasks come to data centers from different sites for computation and storage which makes data centers consume large energy. Thus proper task scheduling is one of the best solutions to utilize cloud data center resources efficiently and reduce their power consumption.
Task scheduling algorithms employed previously to solve large power consumption problem are not much efficient from energy consumption efficiency perspective of cloud data centers due to the features that they haven’t included. To solve this problem, in this research, an appropriate and efficient proximity based task scheduling algorithm is designed and implemented for cloud computing data centers. The proposed proximity based task scheduling algorithm considers the task incoming frequency and proximity of computing and storage machines from the cloud data center scheduler which has a great effect on total energy consumption of cloud computing data centers due to the time latency effect. The proposed scheduling algorithm also includes both preemptive and non-preemptive scheduling features for improved efficiency. The algorithm sets priority for tasks based on their incoming frequency and if two or more tasks have the same frequency, task preemption concept is employed by setting a small time slice for task execution balancing. After the tasks are properly scheduled in this fashion, the data center scheduler enables them to identify the nearby resource for host assignment.
To show its effectiveness, we have evaluated and compared the proposed algorithm with three popular reference scheduling algorithms, namely Random, Round Robin and Heterogeneous Energy efficient Resource allocation Optimizing Scheduling. The evaluation results, found from the experiment are presented and showed more efficient outcome from energy consumption perspective of cloud computing data centers.