DOPPLER: Dual-Policy Learning for Device Assignment in Asynchronous Dataflow Graphs
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915311998992384 |
|---|---|
| author | Yao, Xinyu Bourgeois, Daniel Jain, Abhinav Tang, Yuxin Yao, Jiawen Ding, Zhimin Silva, Arlei Jermaine, Chris |
| author_facet | Yao, Xinyu Bourgeois, Daniel Jain, Abhinav Tang, Yuxin Yao, Jiawen Ding, Zhimin Silva, Arlei Jermaine, Chris |
| contents | We study the problem of assigning operations in a dataflow graph to devices to minimize execution time in a work-conserving system, with emphasis on complex machine learning workloads. Prior learning-based methods often struggle due to three key limitations: (1) reliance on bulk-synchronous systems like TensorFlow, which under-utilize devices due to barrier synchronization; (2) lack of awareness of the scheduling mechanism of underlying systems when designing learning-based methods; and (3) exclusive dependence on reinforcement learning, ignoring the structure of effective heuristics designed by experts. In this paper, we propose \textsc{Doppler}, a three-stage framework for training dual-policy networks consisting of 1) a $\mathsf{SEL}$ policy for selecting operations and 2) a $\mathsf{PLC}$ policy for placing chosen operations on devices. Our experiments show that \textsc{Doppler} outperforms all baseline methods across tasks by reducing system execution time and additionally demonstrates sampling efficiency by reducing per-episode training time. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_23131 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DOPPLER: Dual-Policy Learning for Device Assignment in Asynchronous Dataflow Graphs Yao, Xinyu Bourgeois, Daniel Jain, Abhinav Tang, Yuxin Yao, Jiawen Ding, Zhimin Silva, Arlei Jermaine, Chris Machine Learning Distributed, Parallel, and Cluster Computing We study the problem of assigning operations in a dataflow graph to devices to minimize execution time in a work-conserving system, with emphasis on complex machine learning workloads. Prior learning-based methods often struggle due to three key limitations: (1) reliance on bulk-synchronous systems like TensorFlow, which under-utilize devices due to barrier synchronization; (2) lack of awareness of the scheduling mechanism of underlying systems when designing learning-based methods; and (3) exclusive dependence on reinforcement learning, ignoring the structure of effective heuristics designed by experts. In this paper, we propose \textsc{Doppler}, a three-stage framework for training dual-policy networks consisting of 1) a $\mathsf{SEL}$ policy for selecting operations and 2) a $\mathsf{PLC}$ policy for placing chosen operations on devices. Our experiments show that \textsc{Doppler} outperforms all baseline methods across tasks by reducing system execution time and additionally demonstrates sampling efficiency by reducing per-episode training time. |
| title | DOPPLER: Dual-Policy Learning for Device Assignment in Asynchronous Dataflow Graphs |
| topic | Machine Learning Distributed, Parallel, and Cluster Computing |
| url | https://arxiv.org/abs/2505.23131 |