Scholarlink

<p>Edge computing, as a key technology in the Industrial Internet of Things (IIoT), can meet the essential requirements of the IIoT in terms of real-time operations, intelligent applications, security, and privacy protection. With the continuous expansion of the IIoT, a single edge server will struggle to handle the complex task demands generated by numerous terminal devices. Therefore, this paper addresses the Multi-Server Collaborative Task Offloading and Resource Allocation (MSCTORA) optimization problem, using the weighted sum of task completion time and energy consumption as the performance metric. The proposed approach involves the joint optimization of task offloading decisions, uplink bandwidth allocation, and server computation resource allocation across multiple collaborating servers, while ensuring load balancing among the edge servers. The MSCTORA problem is decomposed into a Task Offloading and Load Balancing (TOLB) problem and a Resource Allocation (RA) problem, which are solved using a hierarchical optimization approach. For the upper-layer TOLB problem, a heuristic algorithm based on server workload thresholds and distance thresholds between terminals and Mobile Edge Computing (MEC) servers is proposed. For the lower-layer RA problem, the Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm is employed. Simulation results demonstrate that, compared to other methods, the proposed MSCTORA optimization scheme significantly reduces production time and energy consumption in industrial production.</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

Hierarchical optimization method for industrial task offloading and resource allocation in multi-server edge computing

  • 期刊论文

  • 发表日期：

    2025年10月9日

查看全文
  • Chunxiang Zheng,
  • Lin Chen,
  • Jiadong Dong,
  • Zhaoxiang Wang,
  • Qinghu Guo,
  • Feihu Sang

摘要

Edge computing, as a key technology in the Industrial Internet of Things (IIoT), can meet the essential requirements of the IIoT in terms of real-time operations, intelligent applications, security, and privacy protection. With the continuous expansion of the IIoT, a single edge server will struggle to handle the complex task demands generated by numerous terminal devices. Therefore, this paper addresses the Multi-Server Collaborative Task Offloading and Resource Allocation (MSCTORA) optimization problem, using the weighted sum of task completion time and energy consumption as the performance metric. The proposed approach involves the joint optimization of task offloading decisions, uplink bandwidth allocation, and server computation resource allocation across multiple collaborating servers, while ensuring load balancing among the edge servers. The MSCTORA problem is decomposed into a Task Offloading and Load Balancing (TOLB) problem and a Resource Allocation (RA) problem, which are solved using a hierarchical optimization approach. For the upper-layer TOLB problem, a heuristic algorithm based on server workload thresholds and distance thresholds between terminals and Mobile Edge Computing (MEC) servers is proposed. For the lower-layer RA problem, the Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm is employed. Simulation results demonstrate that, compared to other methods, the proposed MSCTORA optimization scheme significantly reduces production time and energy consumption in industrial production.