Scholarlink

<p>With the significant growth of cloud computing, optimal task allocation to computational resources has become increasingly important. In this study, an approach based on the Adaptive Neuro-Fuzzy Inference System (ANFIS) and the Giza Pyramid Construction (GPC) metaheuristic algorithm is proposed to enhance task scheduling performance in cloud computing environments. In this model, the GPC algorithm is employed to train and optimize the parameters of ANFIS, enabling more accurate modeling of resource behavior and achieving more efficient task allocation. The GPC algorithm establishes a proper balance between exploration and exploitation through structured search and prevention of premature convergence. The performance of the proposed model is evaluated on two datasets: Synthetic and GoCJ. The results indicate a 15–25% reduction in makespan, a 10% to 20% decrease in energy consumption, and a 12% to 18% increase in throughput compared to baseline methods. These outcomes demonstrate the strong capability of the ANFIS-GPC algorithm in simultaneously optimizing key task scheduling criteria in cloud computing environments.</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

The art of scheduling: ANFIS-GPC synergy for energy-aware cloud optimization

  • 期刊论文

  • 发表日期：

    2026年1月10日

查看全文
  • Ali Abdulkhaleq Alwan Alattraqchi,
  • Madjid Khalilian,
  • Ali Alsalamy,
  • Mohammadreza Soltanaghaei

摘要

With the significant growth of cloud computing, optimal task allocation to computational resources has become increasingly important. In this study, an approach based on the Adaptive Neuro-Fuzzy Inference System (ANFIS) and the Giza Pyramid Construction (GPC) metaheuristic algorithm is proposed to enhance task scheduling performance in cloud computing environments. In this model, the GPC algorithm is employed to train and optimize the parameters of ANFIS, enabling more accurate modeling of resource behavior and achieving more efficient task allocation. The GPC algorithm establishes a proper balance between exploration and exploitation through structured search and prevention of premature convergence. The performance of the proposed model is evaluated on two datasets: Synthetic and GoCJ. The results indicate a 15–25% reduction in makespan, a 10% to 20% decrease in energy consumption, and a 12% to 18% increase in throughput compared to baseline methods. These outcomes demonstrate the strong capability of the ANFIS-GPC algorithm in simultaneously optimizing key task scheduling criteria in cloud computing environments.