Energy and Deadline Aware Workflow Scheduling using Adaptive Remora Optimization in Cloud Computing

Abstract

Cloud computing has become a more popular and well-known computing paradigm for delivering services to different organizations. The main benefits of the cloud computing paradigm, including on-demand services, pay-as-per-use policy, rapid elasticity, and so on, make cloud computing a more emerging technology to lead with new methods. Cloud systems have become more challenging than other systems because of their wide range of clients and the variety of services in the system. The cloud data center consists of many physical machines (PM) with virtual machines (VM), load balancers, switches, storage etc. Because of the inappropriate use of resources and inefficient scheduling, these data centers consume a lot of energy. In this paper, a multi-objective optimization model called Adaptive Remora Optimization (AROA) is proposed, which comprises sub-models viz; priority calculation, task clustering, probability definition and task-VM mapping using search mode based on Remora optimization to optimize energy consumption and execution time. CloudSim is used for the implementation of the proposed optimization technique. Through simulation the energy consumption is 0.695kWh and the execution time is 179.14sec. The result obtained by AROA is compared with the existing approaches to prove the efficacy of the proposed approach.Experimental results show that the proposed AROA algorithm outperforms the existing approaches.