An Energy and Deadline Aware Scheduling Using Greedy Algorithm for Cloud Computing

Abstract

Cloud computing has been providing various services to different users by means of an aid of large and scalable virtualized resources on the internet. Owing to all the recent and inventive developments that are found in the field, there are several scheduling algorithms which were developed in a cloud computing environment with the intention of decreasing the services given in cloud computing. For a very enormous gauge, assorted and the multiuser atmosphere in the cloud scheme where maximization of profit for that of the Cloud Service Provider (CSP) has been the primary objective. For the purpose of this work, the inclusive optimization problem in the operation of the cloud system by means of lowering the cost of procedure and by maximizing the efficiency of energy. At the same time, it satisfies the deadlines that are definite in Service Level Agreements (SLA) that has been addressed from a CSP perspective. The work proposes a Greedy algorithm for the environment of the cloud and this is compared to the scheduling of a First Come First Served (FCFS) and the Min-Min scheduling procedure. This system exploits the tasks and their heterogeneity and also the resources using a scheduler unit that schedules and allocates the tasks which are deadline-constrained which is delimited to the nodes that are energy conscious. After this, the CSP capitalizes on the parallelisms of data for every user workload and also effectively manages all collective user requests and also apply the custom optimization that creates a cost of global energy and a cloud platform which is deadline aware. The results of the experiment prove that this proposed Greedy algorithm which achieves a performance which is better (a guarantee ratio, utilization of resources and energy saving) compared to the FCFS and the Min-Min scheduling algorithm.