Scholarlink

<p>The prevalent adoption of IoT devices and the growth of scientific applications have led to a rapid increase in data volume and computational complexity, necessitating scalable computing infrastructure. As compute and data demands continue to escalate due to large, complex scientific applications, cloud computing has emerged as a key solution for executing scientific workflows within defined Quality of Service parameters. Nevertheless, it remains an exceptionally challenging endeavor to create an efficient and cost-effective solution, particularly for extensive-scale applications. To handle this challenge, this research presents a fully hybrid scheduling algorithm <b>P</b>article <b>E</b>nhanced <b>W</b>ater wave optimization <b>S</b>cheduler (PEWS) based on Particle Swarm Optimization (PSO) and Enhanced Water Wave Optimization (EWWO) that comprises of a clustered breaking operator, aiming to reduce the makespan and financial cost incurred in executing a workflow. The proposed approach utilizes PSO to enhance the global exploring caliber of WWO and is then refined by EWWO. The strength of the proposed method is assessed on the WorkflowSim tool using four different scientific workflows. The experimental results reveal significant improvements, a 17.61% reduction in makespan, a 10.83% decrease in financial costs, and a 14.14% increase in resource utilization. The Average Relative Deviation Index (ARDI) values for our approach are lower in all of the cases, and the results are further validated through Wilcoxon and Friedman statistical tests. Simulation outcomes demonstrate that PEWS surpasses five other established and well-recognized multi-objective optimization approaches.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Optimizing scientific workflow scheduling in cloud environments: a hybrid PSO-EWWO

  • 期刊论文

  • 发表日期：

    2025年8月19日

查看全文
  • Prasanth Kumar Bevara,
  • Ravi Shankar Singh,
  • Rambabu Medera,
  • V. V. D. Prasad Chelluri

摘要

The prevalent adoption of IoT devices and the growth of scientific applications have led to a rapid increase in data volume and computational complexity, necessitating scalable computing infrastructure. As compute and data demands continue to escalate due to large, complex scientific applications, cloud computing has emerged as a key solution for executing scientific workflows within defined Quality of Service parameters. Nevertheless, it remains an exceptionally challenging endeavor to create an efficient and cost-effective solution, particularly for extensive-scale applications. To handle this challenge, this research presents a fully hybrid scheduling algorithm Particle Enhanced Water wave optimization Scheduler (PEWS) based on Particle Swarm Optimization (PSO) and Enhanced Water Wave Optimization (EWWO) that comprises of a clustered breaking operator, aiming to reduce the makespan and financial cost incurred in executing a workflow. The proposed approach utilizes PSO to enhance the global exploring caliber of WWO and is then refined by EWWO. The strength of the proposed method is assessed on the WorkflowSim tool using four different scientific workflows. The experimental results reveal significant improvements, a 17.61% reduction in makespan, a 10.83% decrease in financial costs, and a 14.14% increase in resource utilization. The Average Relative Deviation Index (ARDI) values for our approach are lower in all of the cases, and the results are further validated through Wilcoxon and Friedman statistical tests. Simulation outcomes demonstrate that PEWS surpasses five other established and well-recognized multi-objective optimization approaches.