Scholarlink

<p>One of the main challenges in cloud computing is efficient resource utilization and energy management without compromising service quality. This includes optimizing performance and energy consumption, managing physical and virtual machines, reducing server usage, prioritizing virtual machines appropriately, and ensuring centralized task assignment with balanced workload distribution. This research proposes a fuzzy hierarchical approach based on a combination of optimal feature selection and machine learning techniques for effective load and energy management in cloud environments. The proposed method consists of three main phases: feature selection, prioritization of virtual machines, and optimization using an improved Fruit Fly Optimization Algorithm (FFOA). The method is evaluated on the Azure dataset using multiple performance metrics. The results show that the proposed method significantly improves predictive accuracy, achieves efficient scheduling, and reduces energy consumption. In particular, the framework demonstrates notable improvements in both Makespan and energy efficiency compared with existing methods, confirming its effectiveness for large-scale cloud 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

Presenting a fuzzy hierarchical method based on the combination of optimal features and machine learning for load and energy management in the cloud environment

  • 期刊论文

  • 发表日期：

    2025年10月2日

查看全文
  • Saif Salam Ibrahim Al-witwit,
  • Esmaeil Bagheri,
  • Mehdi Hamidkhani,
  • Mezher H. Mezher

摘要

One of the main challenges in cloud computing is efficient resource utilization and energy management without compromising service quality. This includes optimizing performance and energy consumption, managing physical and virtual machines, reducing server usage, prioritizing virtual machines appropriately, and ensuring centralized task assignment with balanced workload distribution. This research proposes a fuzzy hierarchical approach based on a combination of optimal feature selection and machine learning techniques for effective load and energy management in cloud environments. The proposed method consists of three main phases: feature selection, prioritization of virtual machines, and optimization using an improved Fruit Fly Optimization Algorithm (FFOA). The method is evaluated on the Azure dataset using multiple performance metrics. The results show that the proposed method significantly improves predictive accuracy, achieves efficient scheduling, and reduces energy consumption. In particular, the framework demonstrates notable improvements in both Makespan and energy efficiency compared with existing methods, confirming its effectiveness for large-scale cloud environments.