Scholarlink

<p>Prior studies on task offloading in mobile edge computing have primarily focused on task scheduling methods, which assume that the task’s characteristics are known before actual execution. In real-world mobile edge computing, involving uncertainty in the computation and communication size of latency-sensitive tasks, it is difficult to find an ideal assignment of tasks. The primary objective of the present work is to maximize the number of tasks that meet the deadline and then to optimize the secondary objectives, overall completion time, total cost for execution of all tasks and total energy consumption of all computing devices as well. In this paper, an enhanced multi-objective approach is proposed to handle the deadline-aware bag-of-tasks scheduling problem in an edge computing environment to optimize multiple objectives, which considers benefits in terms of reduction in overall makespan, total cost, and energy consumption while considering the deadline constraints. In this work, an improved Strength Pareto Evolutionary Algorithm 2 ( SPEA2) is presented that can give the best trade-offs among multiple objectives. The outcomes of the experiments demonstrate that the proposed approaches can identify a collection of pareto optimal solutions of comparable quality to meet several QoS objectives with disparate user preferences.</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

Efficient deadline-aware Non-Clairvoyant bag-of-tasks scheduling for edge computing using an enhanced SPEA2

  • 期刊论文

  • 发表日期：

    2025年9月29日

查看全文
  • Varsha Kumari,
  • Chapram Sudhakar

摘要

Prior studies on task offloading in mobile edge computing have primarily focused on task scheduling methods, which assume that the task’s characteristics are known before actual execution. In real-world mobile edge computing, involving uncertainty in the computation and communication size of latency-sensitive tasks, it is difficult to find an ideal assignment of tasks. The primary objective of the present work is to maximize the number of tasks that meet the deadline and then to optimize the secondary objectives, overall completion time, total cost for execution of all tasks and total energy consumption of all computing devices as well. In this paper, an enhanced multi-objective approach is proposed to handle the deadline-aware bag-of-tasks scheduling problem in an edge computing environment to optimize multiple objectives, which considers benefits in terms of reduction in overall makespan, total cost, and energy consumption while considering the deadline constraints. In this work, an improved Strength Pareto Evolutionary Algorithm 2 ( SPEA2) is presented that can give the best trade-offs among multiple objectives. The outcomes of the experiments demonstrate that the proposed approaches can identify a collection of pareto optimal solutions of comparable quality to meet several QoS objectives with disparate user preferences.