A Framework for the Energy Aware Multi-Layered Task Scheduling in Heterogeneous Cloud
The Cloud Computing paradigm stands out as the latest technological trends today where the application based services are provided by the distributed resources located at the remote locations. Task scheduling and the load balancing remain the most vital areas to check and save the energy during routine processes. This research paper presents a heuristics inspired approach for energy saving job scheduling and resource allocation. Here the scheduler adapts to the optimal scheduling technique to map the available resources for the various scalable tasks in the heterogeneous cloud environment. Common parameters that are used to analyze the performance of standard scheduling techniques to compare and access different nature of tasks are Make-span, Average Response Time (ART) and Throughput. The previous standard algorithms could not yield better performance with respect to the above cited parameters and hence resulted in loss of energy. In this paper, we have proposed a Energy Efficient Multi layered Scheduling (EEMLS) techniques that outperforms the standard algorithms techniques for energy efficient workflow in large tasks i.e. Round Robin, Max-Min, Opportunistic Load Balancing (OLB), Minimum Completion Time (MRT) and Artificial Bee Colony (ABC) algorithms, based on Average Response Time, Throughput and Makespan parameter. The efficacy of the proposed EEMLS techniques has been analyzed using CloudSim.