Scholarlink

<p>It is crucial to optimize task scheduling in cloud computing to enhance performance, particularly in scientific and resource-intensive applications. Many existing studies fail to balance key metrics such as makespan, resource utilization, and execution cost. This paper introduces an Enhanced Predator-Prey Optimization (EPPO) algorithm based on game theory to address these challenges. In addition to dynamic predator–prey interactions, EPPO incorporates objective functions that consider waiting time and task processing time. Comparative analyses with WOA, SSA, HHO, HGS, GWO, FOX, and GEO demonstrate EPPO’s superior performance, achieving improvements of 32% in makespan, 23% in load balancing, 25% in execution cost, and 4% in resource utilization. EPPO also converges faster than metaheuristic-based and game theory-based scheduling algorithms. By integrating game theory, EPPO effectively resolves conflicting objectives and allocates resources efficiently. Its robustness and scalability highlight its potential as a next-generation cloud scheduling solution. Future work will focus on parallelization, energy consumption, and security considerations.</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

Game theory-based framework for efficient task scheduling in cloud computing

  • 期刊论文

  • 发表日期：

    2026年1月9日

查看全文
  • Zahra Jalali Khalil Abadi,
  • Mohammad Masoud Javidi,
  • Najme Mansouri,
  • Behnam Mohammad Hasani Zade

摘要

It is crucial to optimize task scheduling in cloud computing to enhance performance, particularly in scientific and resource-intensive applications. Many existing studies fail to balance key metrics such as makespan, resource utilization, and execution cost. This paper introduces an Enhanced Predator-Prey Optimization (EPPO) algorithm based on game theory to address these challenges. In addition to dynamic predator–prey interactions, EPPO incorporates objective functions that consider waiting time and task processing time. Comparative analyses with WOA, SSA, HHO, HGS, GWO, FOX, and GEO demonstrate EPPO’s superior performance, achieving improvements of 32% in makespan, 23% in load balancing, 25% in execution cost, and 4% in resource utilization. EPPO also converges faster than metaheuristic-based and game theory-based scheduling algorithms. By integrating game theory, EPPO effectively resolves conflicting objectives and allocates resources efficiently. Its robustness and scalability highlight its potential as a next-generation cloud scheduling solution. Future work will focus on parallelization, energy consumption, and security considerations.