Energy Saving Algorithms for Workflow Scheduling in Cloud Computing

Cloud computing is one of the platforms able to provide green computing solutions nowadays. Moreover, it is able to provide a high performance environment for scientific applications. However, if there is dependency between tasks, there may be a low utilization of the cloud. To address the problem, this paper presents new task schedulers for clouds aiming to achieve energy savings. Results of experiments comparing the proposed schedulers with other existing ones show that it is possible to obtain energy savings of up to 22.7%, with no penalty in the make span. Moreover it is observed that the efficiency of the scheduling is dependent on how tasks are interconnected in the workflow.