HATS: HetTask scheduling
To handle task execution, modern supercomputers employ thousands (or millions) of processors. In such supercomputers, task scheduling has a meaningful impression on system performance. To improve efficiency, task scheduling algorithms aim to decrease the volume of communication and the number of mes...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Published: |
Institute of Electrical and Electronics Engineers
2022
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/101684/ |
| _version_ | 1848863609788039168 |
|---|---|
| author | Koohi, Sina Zangbari Abdul Hamid, Nor Asilah Wati Othman, Mohamed Ibragimov, Gafurjan |
| author_facet | Koohi, Sina Zangbari Abdul Hamid, Nor Asilah Wati Othman, Mohamed Ibragimov, Gafurjan |
| author_sort | Koohi, Sina Zangbari |
| building | UPM Institutional Repository |
| collection | Online Access |
| description | To handle task execution, modern supercomputers employ thousands (or millions) of processors. In such supercomputers, task scheduling has a meaningful impression on system performance. To improve efficiency, task scheduling algorithms aim to decrease the volume of communication and the number of message exchanges. These efforts, however, result in other bottlenecks, such as high-link congestion. In addition, the heterogeneity of processors and networks is another major challenge for schedulers. This paper presents a new algorithm for scheduling called Heterogeneity-Aware Task Scheduling (HATS). The proposed algorithm adopts an updated multi-level hyper-graph partitioning approach. It describes a new method of aggregation in the coarsening step that helps to accurately coarsen the hyper-graph of the task model. The Raccoon Optimization algorithm is then used in the initial partitioning phase, and in the un-coarsening phase, a novel refinement procedure optimises the initial partitions. The experiments on this approach showed that, compared to the other well-known algorithms, the proposed method offers better schedules with lower communication volume and imbalance ratio in a shorter time. |
| first_indexed | 2025-11-15T13:35:39Z |
| format | Article |
| id | upm-101684 |
| institution | Universiti Putra Malaysia |
| institution_category | Local University |
| last_indexed | 2025-11-15T13:35:39Z |
| publishDate | 2022 |
| publisher | Institute of Electrical and Electronics Engineers |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | upm-1016842024-08-05T07:43:42Z http://psasir.upm.edu.my/id/eprint/101684/ HATS: HetTask scheduling Koohi, Sina Zangbari Abdul Hamid, Nor Asilah Wati Othman, Mohamed Ibragimov, Gafurjan To handle task execution, modern supercomputers employ thousands (or millions) of processors. In such supercomputers, task scheduling has a meaningful impression on system performance. To improve efficiency, task scheduling algorithms aim to decrease the volume of communication and the number of message exchanges. These efforts, however, result in other bottlenecks, such as high-link congestion. In addition, the heterogeneity of processors and networks is another major challenge for schedulers. This paper presents a new algorithm for scheduling called Heterogeneity-Aware Task Scheduling (HATS). The proposed algorithm adopts an updated multi-level hyper-graph partitioning approach. It describes a new method of aggregation in the coarsening step that helps to accurately coarsen the hyper-graph of the task model. The Raccoon Optimization algorithm is then used in the initial partitioning phase, and in the un-coarsening phase, a novel refinement procedure optimises the initial partitions. The experiments on this approach showed that, compared to the other well-known algorithms, the proposed method offers better schedules with lower communication volume and imbalance ratio in a shorter time. Institute of Electrical and Electronics Engineers 2022 Article PeerReviewed Koohi, Sina Zangbari and Abdul Hamid, Nor Asilah Wati and Othman, Mohamed and Ibragimov, Gafurjan (2022) HATS: HetTask scheduling. IEEE Transactions on Cloud Computing, 11 (2). pp. 2071-2083. ISSN 2168-7161; ESSN: 2372-0018 https://ieeexplore.ieee.org/document/9800200 10.1109/tcc.2022.3184081 |
| spellingShingle | Koohi, Sina Zangbari Abdul Hamid, Nor Asilah Wati Othman, Mohamed Ibragimov, Gafurjan HATS: HetTask scheduling |
| title | HATS: HetTask scheduling |
| title_full | HATS: HetTask scheduling |
| title_fullStr | HATS: HetTask scheduling |
| title_full_unstemmed | HATS: HetTask scheduling |
| title_short | HATS: HetTask scheduling |
| title_sort | hats: hettask scheduling |
| url | http://psasir.upm.edu.my/id/eprint/101684/ http://psasir.upm.edu.my/id/eprint/101684/ http://psasir.upm.edu.my/id/eprint/101684/ |