The HEFT Algorithm Based on Node Merging
The task scheduling problem in power systems is complex, especially when communication costs exceed computation costs. This paper proposes an improved HEFT algorithm with a node merging mechanism, merging nodes through hierarchical traversal to assign high-communication-cost tasks to the same processor, reducing communication overhead. Experiments with 499-dimensional and 1614-dimensional matrices from the power system show that the improved algorithm reduces completion time by 5ms and 2ms, respectively, compared to standard HEFT.
Authors
- Ying Huang (ORCID: https://orcid.org/0000-0003-0543-0256)
- Jingyue Zhang (ORCID: https://orcid.org/0009-0004-8375-0380)
- J. Y. Zhang (ORCID: https://orcid.org/0009-0001-9587-9913)
- Yuanhong Lu (ORCID: https://orcid.org/0009-0003-0381-3971)
- Haiping Guo (ORCID: https://orcid.org/0009-0005-1675-7811)
Institutions
- China Southern Power Grid (China) (CN)
Publication Details
- Published
- 2024-11-01
- DOI
- https://doi.org/10.1109/iwceaa63616.2024.10823829
- Primary Topic
- Advanced Algorithms and Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00