
Research Article
A Dichotomous Repair-Based Load-Balanced Task Allocation Strategy in Cloud-Edge Environment
@INPROCEEDINGS{10.1007/978-3-031-54521-4_8, author={Zekun Hu and Pengwei Wang and Peihai Zhao and Zhaohui Zhang}, title={A Dichotomous Repair-Based Load-Balanced Task Allocation Strategy in Cloud-Edge Environment}, proceedings={Collaborative Computing: Networking, Applications and Worksharing. 19th EAI International Conference, CollaborateCom 2023, Corfu Island, Greece, October 4-6, 2023, Proceedings, Part I}, proceedings_a={COLLABORATECOM}, year={2024}, month={2}, keywords={Load balancing Task allocation Dichotomous Edge computing Cloud-edge}, doi={10.1007/978-3-031-54521-4_8} }
- Zekun Hu
Pengwei Wang
Peihai Zhao
Zhaohui Zhang
Year: 2024
A Dichotomous Repair-Based Load-Balanced Task Allocation Strategy in Cloud-Edge Environment
COLLABORATECOM
Springer
DOI: 10.1007/978-3-031-54521-4_8
Abstract
Load balancing is a hot issue in the current cloud-edge environment. However, due to the characteristics of edge computing, load balancing needs to be better integrated with edge devices and edge networks to provide higher performance and reliability. The presence of a large number of overloaded nodes may lead to load imbalance and thus affect the efficiency of nodes. To solve this problem, the key is how to allocate tasks to the appropriate resources. To this end, this work proposes a dichotomous task allocation policy Dichotomous Repair (DREP) to achieve efficient task allocation and overall load balancing of edge nodes in cloud-edge environment. The proposed policy consists of five steps: grouping, adjustment, filtering, greed and repair. The dichotomous policy is adopted to generate the initial allocation scheme according to the number of edge nodes, and then the overloaded and underloaded nodes are repaired by the subsequent two-stage repair policy to maintain the load balance. Finally, through extensive experiments, we evaluate the proposed method and the results show that it outperforms other algorithms in terms of workload balancing.