File Download
There are no files associated with this item.
Supplementary
-
Citations:
- Appears in Collections:
Conference Paper: TapFinger: Task Placement and Fine-Grained Resource Allocation for Edge Machine Learning
Title | TapFinger: Task Placement and Fine-Grained Resource Allocation for Edge Machine Learning |
---|---|
Authors | |
Issue Date | 17-May-2023 |
Abstract | Machine learning (ML) tasks are one of the major workloads in today’s edge computing networks. Existing edge- cloud schedulers allocate the requested amounts of resources to each task, falling short of best utilizing the limited edge resources flexibly for ML task performance optimization. This paper proposes TapFinger, a distributed scheduler that minimizes the total completion time of ML tasks in a multi-cluster edge network, through co-optimizing task placement and fine-grained multi-resource allocation. To learn the tasks’ uncertain resource sensitivity and enable distributed online scheduling, we adopt multi-agent reinforcement learning (MARL), and propose sev- eral techniques to make it efficient for our ML-task resource allocation. First, TapFinger uses a heterogeneous graph attention network as the MARL backbone to abstract inter-related state features into more learnable environmental patterns. Second, the actor network is augmented through a tailored task selection phase, which decomposes the actions and encodes the opti- mization constraints. Third, to mitigate decision conflicts among agents, we novelly combine Bayes’ theorem and masking schemes to facilitate our MARL model training. Extensive experiments using synthetic and test-bed ML task traces show that TapFinger can achieve up to 28.6% reduction in the average task completion time and improve resource efficiency as compared to state-of-the- art resource schedulers. |
Persistent Identifier | http://hdl.handle.net/10722/333889 |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Li, Yihong | - |
dc.contributor.author | Zeng, Tianyu | - |
dc.contributor.author | Zhang, Xiaoxi | - |
dc.contributor.author | Duan, Jingpu | - |
dc.contributor.author | Wu, Chuan | - |
dc.date.accessioned | 2023-10-06T08:39:55Z | - |
dc.date.available | 2023-10-06T08:39:55Z | - |
dc.date.issued | 2023-05-17 | - |
dc.identifier.uri | http://hdl.handle.net/10722/333889 | - |
dc.description.abstract | <p>Machine learning (ML) tasks are one of the major workloads in today’s edge computing networks. Existing edge- cloud schedulers allocate the requested amounts of resources to each task, falling short of best utilizing the limited edge resources flexibly for ML task performance optimization. This paper proposes TapFinger, a distributed scheduler that minimizes the total completion time of ML tasks in a multi-cluster edge network, through co-optimizing task placement and fine-grained multi-resource allocation. To learn the tasks’ uncertain resource sensitivity and enable distributed online scheduling, we adopt multi-agent reinforcement learning (MARL), and propose sev- eral techniques to make it efficient for our ML-task resource allocation. First, TapFinger uses a heterogeneous graph attention network as the MARL backbone to abstract inter-related state features into more learnable environmental patterns. Second, the actor network is augmented through a tailored task selection phase, which decomposes the actions and encodes the opti- mization constraints. Third, to mitigate decision conflicts among agents, we novelly combine Bayes’ theorem and masking schemes to facilitate our MARL model training. Extensive experiments using synthetic and test-bed ML task traces show that TapFinger can achieve up to 28.6% reduction in the average task completion time and improve resource efficiency as compared to state-of-the- art resource schedulers.</p> | - |
dc.language | eng | - |
dc.relation.ispartof | IEEE International Conference on Computer Communications (INFOCOM) 2023 (17/05/2023-20/05/2023, New York) | - |
dc.title | TapFinger: Task Placement and Fine-Grained Resource Allocation for Edge Machine Learning | - |
dc.type | Conference_Paper | - |
dc.identifier.doi | 10.1109/INFOCOM53939.2023.10229031 | - |