arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2005.13247cs.DCcs.LGcs.NI
收藏

分布式深度学习中通信优化的定量研究

A Quantitative Survey of Communication Optimizations in Distributed Deep Learning

The Hong Kong University of Science and Technology(香港科技大学);Hong Kong Baptist University(香港浸会大学);Shenzhen Technology University(深圳技术大学)

Shaohuai Shi, Zhenheng Tang, Xiaowen Chu, Chengjian Liu, Wei Wang, Bo Li

摘要:

如今,大型复杂的深度学习(DL)模型越来越多地以分布式方式在多个工作机器上进行训练,其中工作机器之间的广泛通信带来了严重的扩展问题。在这篇文章中,我们提出了一个定量的调查数据并行分布式DL的通信优化技术。我们首先确定了主要的通信挑战,并将现有的解决方案分为三个层次,即学习算法,系统架构和网络基础设施。我们提出了最先进的通信优化技术,并进行了比较研究的7种常见的无损分布式DL方法上的32 GPU集群与100 Gbps的InfiniBand(IB)。我们发现:(1)低模型强度的DL模型(如BERT和BERT-Large)即使使用最好的无损算法也难以扩展到100 Gbps以上的IB;(2)系统架构和调度算法对扩展性能有关键影响。最后,我们对有待进一步研究的开放问题进行了讨论。

英文摘要:

Nowadays, large and complex deep learning (DL) models are increasingly trained in a distributed manner across multiple worker machines, in which extensive communications between workers pose serious scaling problems. In this article, we present a quantitative survey of communication optimization techniques for data parallel distributed DL. We first identify the major communication challenges and classify the existing solutions into three levels, namely the learning algorithm, the system architecture, and the network infrastructure. We present the state-of-the-art communication optimization techniques and conduct a comparative study of seven common lossless distributed DL methods on a 32-GPU cluster with 100Gbps InfiniBand (IB). We show that (1) the DL models with low model intensity (such as BERT and BERT-Large) are difficult to scale out even with the best available lossless algorithm over 100Gbps IB; (2) the system architecture and scheduling algorithms have a critical impact on the scaling property. We conclude the article with discussions on the open issues for further investigations.

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。

↑