AllReduce is a technique in distributed computing which saw use in many critical applications of deep learning. Existing methods of AllReduce scheduling oftentimes lack flexibility due to being topology-specific or relying on extensive handcrafted designs that require domain-specific knowledge. In this work, we aim to alleviate this inflexibility by proposing a deep-reinforcement-learning (DRL)-based pipeline that can generate AllReduce scheduling for various network topologies without topology-specific design features. The flow scheduling module of this pipeline consists of two hierarchically-structured DRL policies that work cooperatively to find optimal scheduling. We showcase the performance of our method compared to the baseline methods on three topologies: BCube, DCell, and Jellyfish. Finally, we contributed a Python-based simulation environment simulating AllReduce scheduling on these network topologies.
@article{arxiv.2503.21013,
title = {AllReduce Scheduling with Hierarchical Deep Reinforcement Learning},
author = {Yufan Wei and Mickel Liu and Wenfei Wu},
journal= {arXiv preprint arXiv:2503.21013},
year = {2025}
}