25898

HeterPS: Distributed Deep Learning With Reinforcement Learning Based Scheduling in Heterogeneous Environments

Ji Liu, Zhihua Wu, Dianhai Yu, Yanjun Ma, Danlei Feng, Minxu Zhang, Xinxuan Wu, Xuefeng Yao, Dejing Dou
Baidu Inc.
arXiv:2111.10635 [cs.DC], (20 Nov 2021)

@misc{liu2021heterps,

   title={HeterPS: Distributed Deep Learning With Reinforcement Learning Based Scheduling in Heterogeneous Environments},

   author={Ji Liu and Zhihua Wu and Dianhai Yu and Yanjun Ma and Danlei Feng and Minxu Zhang and Xinxuan Wu and Xuefeng Yao and Dejing Dou},

   year={2021},

   eprint={2111.10635},

   archivePrefix={arXiv},

   primaryClass={cs.DC}

}

Deep neural networks (DNNs) exploit many layers and a large number of parameters to achieve excellent performance. The training process of DNN models generally handles large-scale input data with many sparse features, which incurs high Input/Output (IO) cost, while some layers are compute-intensive. The training process generally exploits distributed computing resources to reduce training time. In addition, heterogeneous computing resources, e.g., CPUs, GPUs of multiple types, are available for the distributed training process. Thus, the scheduling of multiple layers to diverse computing resources is critical for the training process. To efficiently train a DNN model using the heterogeneous computing resources, we propose a distributed framework, i.e., Paddle-Heterogeneous Parameter Server (Paddle-HeterPS), composed of a distributed architecture and a Reinforcement Learning (RL)-based scheduling method. The advantages of Paddle-HeterPS are three-fold compared with existing frameworks. First, Paddle-HeterPS enables efficient training process of diverse workloads with heterogeneous computing resources. Second, Paddle-HeterPS exploits an RL-based method to efficiently schedule the workload of each layer to appropriate computing resources to minimize the cost while satisfying throughput constraints. Third, Paddle-HeterPS manages data storage and data communication among distributed computing resources. We carry out extensive experiments to show that Paddle-HeterPS significantly outperforms state-of-the-art approaches in terms of throughput (14.5 times higher) and monetary cost (312.3% smaller). The codes of the framework are publicly available.
No votes yet.
Please wait...

* * *

* * *

* * *

HGPU group © 2010-2022 hgpu.org

All rights belong to the respective authors

Contact us: