| Abstrakt: |
Existing multiple workflow scheduling techniques focus on traditional Quality of Service (QoS) parameters such as cost, deadline, and makespan to find optimal solutions by consuming a large amount of electrical energy. Higher energy consumption decreases system efficiency, increases operational cost, and generates more carbon footprint. These major problems can lead to several problems, such as economic strain, environmental degradation, resource depletion, energy dependence, health impacts, etc. In a cloud computing environment, scheduling multiple workflows is critical in developing a strategy for energy optimization, which is an NP-hard problem. This paper proposes a novel, bi-phase Energy-Efficient Fruit Fly-based Optimization (EFFO) algorithm for optimizing energy consumption for scheduling multiple workflows. In the first phase, the proposed EFFO algorithm uses first come, first serve, priority scheduling and a Genetic Algorithm to generate the initial workflow search space. In the second phase, the energy consumption is optimized by the proposed EFFO algorithm. Eight NAS benchmarks and five NAS classes (A, B, C, S & W) are employed as a case study. The simulation results are carried out on the WorkflowSim 1.0 platform to test the efficacy of the proposed EFFO algorithm. The experimental results are compared against energy-aware for workflow scheduling and virtual machine consolidation (EASVMC), Power-Efficient Scheduling for Virtual Machine Systems (PESVMS), Energy Efficiency Scheduler (EES), and heterogeneous earliest finish time (HEFT) algorithms and outperformed them with 10.518%, 16.302%, 26.154%, and 28.982%, respectively, based on average energy consumption on five scientific workflows comprised Montage, CyberShake, Laser Interferometer Gravitational-Wave Observatory (LIGO), Scripps Institution of Oceanography High-Throughput (SIPHT), and Epigenomics. [ABSTRACT FROM AUTHOR] |
