arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1809.02839cs.DCcs.LG
收藏

多GPU平台上基于模型并行训练的高效鲁棒并行DNN

Efficient and Robust Parallel DNN Training through Model Parallelism on Multi-GPU Platform

Chi-Chung Chen, Chia-Lin Yang, Hsiang-Yun Cheng

摘要:

深度神经网络(DNN)的训练过程是计算密集型的,通常需要数天到数周的时间来训练DNN模型。因此,在GPU上并行执行DNN训练是当今广泛采用的加速过程的方法。由于实现简单,数据并行是目前最常用的并行化方法。然而,由于GPU之间频繁的权重同步,数据并行遭受过多的GPU间通信开销。另一种方法是流水线模型并行,它在GPU之间划分DNN模型,并并发处理多个小批。与数据并行相比,该方法可以显著降低GPU间通信成本。然而,流水线模型并行面临着权重陈旧的问题;也就是说,梯度是用陈旧的权重计算的,导致训练不稳定和精度损失。在本文中,我们提出了一种流水线模型并行执行方法,使高GPU利用率,同时保持强大的训练精度,通过一种新的权重预测技术,SpecTrain。实验结果表明,我们的建议实现了高达8.91倍的加速比相比,在4-GPU平台上的数据并行,同时保持可比的模型精度。

英文摘要:

The training process of Deep Neural Network (DNN) is compute-intensive, often taking days to weeks to train a DNN model. Therefore, parallel execution of DNN training on GPUs is a widely adopted approach to speed up the process nowadays. Due to the implementation simplicity, data parallelism is currently the most commonly used parallelization method. Nonetheless, data parallelism suffers from excessive inter-GPU communication overhead due to frequent weight synchronization among GPUs. Another approach is pipelined model parallelism, which partitions a DNN model among GPUs, and processes multiple mini-batches concurrently. This approach can significantly reduce inter-GPU communication cost compared to data parallelism. However, pipelined model parallelism faces the weight staleness issue; that is, gradients are computed with stale weights, leading to training instability and accuracy loss. In this paper, we present a pipelined model parallel execution method that enables high GPU utilization while maintaining robust training accuracy via a novel weight prediction technique, SpecTrain. Experimental results show that our proposal achieves up to 8.91x speedup compared to data parallelism on a 4-GPU platform while maintaining comparable model accuracy.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑