SMDP-Based Dynamic Batching for Improving Responsiveness and Energy Efficiency of Batch Services
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866913635077455872 |
|---|---|
| author | Xu, Yaodan Zhou, Sheng Niu, Zhisheng |
| author_facet | Xu, Yaodan Zhou, Sheng Niu, Zhisheng |
| contents | For servers incorporating parallel computing resources, batching is a pivotal technique for providing efficient and economical services at scale. Parallel computing resources exhibit heightened computational and energy efficiency when operating with larger batch sizes. However, in the realm of online services, the adoption of a larger batch size may lead to longer response times. This paper aims to provide a dynamic batching scheme that delicately balances latency and efficiency. The system is modeled as a batch service queue with size-dependent service times. Then, the design of dynamic batching is formulated as a semi-Markov decision process (SMDP) problem, with the objective of minimizing the weighted sum of average response time and average power consumption. A method is proposed to derive an approximate optimal SMDP solution, representing the chosen dynamic batching policy. By introducing an abstract cost to reflect the impact of "tail" states, the space complexity and the time complexity of the procedure can decrease by 63.5% and 98%, respectively. Numerical results showcase the superiority of SMDP-based batching policies across various parameter setups. Additionally, the proposed scheme exhibits noteworthy flexibility in balancing power consumption and latency. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_02181 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SMDP-Based Dynamic Batching for Improving Responsiveness and Energy Efficiency of Batch Services Xu, Yaodan Zhou, Sheng Niu, Zhisheng Distributed, Parallel, and Cluster Computing Machine Learning Systems and Control For servers incorporating parallel computing resources, batching is a pivotal technique for providing efficient and economical services at scale. Parallel computing resources exhibit heightened computational and energy efficiency when operating with larger batch sizes. However, in the realm of online services, the adoption of a larger batch size may lead to longer response times. This paper aims to provide a dynamic batching scheme that delicately balances latency and efficiency. The system is modeled as a batch service queue with size-dependent service times. Then, the design of dynamic batching is formulated as a semi-Markov decision process (SMDP) problem, with the objective of minimizing the weighted sum of average response time and average power consumption. A method is proposed to derive an approximate optimal SMDP solution, representing the chosen dynamic batching policy. By introducing an abstract cost to reflect the impact of "tail" states, the space complexity and the time complexity of the procedure can decrease by 63.5% and 98%, respectively. Numerical results showcase the superiority of SMDP-based batching policies across various parameter setups. Additionally, the proposed scheme exhibits noteworthy flexibility in balancing power consumption and latency. |
| title | SMDP-Based Dynamic Batching for Improving Responsiveness and Energy Efficiency of Batch Services |
| topic | Distributed, Parallel, and Cluster Computing Machine Learning Systems and Control |
| url | https://arxiv.org/abs/2501.02181 |