Scholarlink

<p>Cloud computing task scheduling presents a critical challenge in enhancing resource utilization, requiring both reduced completion times and balanced workloads. This paper proposes the Gauss Dual-Mode Competitive Swarm Optimizer (GDCSO), which integrates Gaussian mutation with a dual-mode update strategy. By enhancing population diversity and local search capabilities, GDCSO effectively avoids premature convergence. Comparisons with algorithms such as Particle Swarm Optimization (PSO), Dwarf Mongoose Optimization (DMO), and Improved Competitive Swarm Optimization (ICSO) in the CloudSim simulation environment demonstrate that GDCSO significantly reduces task completion time, improves virtual machine load balancing, and achieves superior overall performance. Furthermore, evaluations on the IEEE Congress on Evolutionary Computation (CEC) benchmark dataset show that GDCSO outperforms both traditional and emerging optimization algorithms in key performance metrics such as mean and standard deviation. The proposal and validation of GDCSO carry important implications for advancing the application of intelligent optimization algorithms in cloud computing.</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

Gauss dual-mode competitive swarm optimizer algorithm in cloud computing environment

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年12月1日

查看全文
  • Bin Liang,
  • Qiansi Chen

摘要

Cloud computing task scheduling presents a critical challenge in enhancing resource utilization, requiring both reduced completion times and balanced workloads. This paper proposes the Gauss Dual-Mode Competitive Swarm Optimizer (GDCSO), which integrates Gaussian mutation with a dual-mode update strategy. By enhancing population diversity and local search capabilities, GDCSO effectively avoids premature convergence. Comparisons with algorithms such as Particle Swarm Optimization (PSO), Dwarf Mongoose Optimization (DMO), and Improved Competitive Swarm Optimization (ICSO) in the CloudSim simulation environment demonstrate that GDCSO significantly reduces task completion time, improves virtual machine load balancing, and achieves superior overall performance. Furthermore, evaluations on the IEEE Congress on Evolutionary Computation (CEC) benchmark dataset show that GDCSO outperforms both traditional and emerging optimization algorithms in key performance metrics such as mean and standard deviation. The proposal and validation of GDCSO carry important implications for advancing the application of intelligent optimization algorithms in cloud computing.