arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1804.01138cs.DC
收藏

设计微型基准套件以评估用于TensorFlow的gRPC:早期经验

Designing a Micro-Benchmark Suite to Evaluate gRPC for TensorFlow: Early Experiences

Rajarshi Biswas, Xiaoyi Lu, Dhabaleswar K. Panda

摘要:

远程过程调用(RPC)是许多现代分布式系统的支柱。Google的gRPC是社区中最流行的开源RPC框架之一。gRPC是Google深度学习框架TensorFlow的主要通信引擎。TensorFlow主要使用gRPC在不同进程之间传输张量和管理任务。训练阶段的张量更新是通信密集型的,因此TensorFlow的性能在很大程度上取决于底层网络和通信引擎的效率。在TensorFlow上训练深度学习模型可能需要大量时间,从几分钟到几小时,甚至几天。因此,系统研究人员需要投入大量的时间来了解通信对整体性能的影响。显然,系统研究人员缺乏可用的基准。因此,我们提出了TF-gRPC-Bench微基准测试套件,使系统研究人员能够快速了解底层网络和通信运行时对深度学习工作负载的影响。为了实现这一点,我们首先通过训练流行的深度学习模型来分析gRPC上TensorFlow工作负载的特征。然后,我们提出了三个微基准,考虑到这些工作负载的特点。此外,我们全面评估gRPC与TF-gRPC-Bench微基准套件在不同的集群在以太网,IPoIB,和RDMA,并提出了结果。

英文摘要:

Remote procedure call (RPC) is the backbone of many modern distributed systems. Google's gRPC is one of the most popular open source RPC frameworks available in the community. gRPC is the main communication engine for Google's Deep Learning framework TensorFlow. TensorFlow primarily uses gRPC for communicating tensors and administrative tasks among different processes. Tensor updates during the training phase are communication intensive and thus TensorFlow's performance is heavily dependent on the underlying network and the efficacy of the communication engine. Training deep learning models on TensorFlow can take significant time ranging from several minutes to several hours, even several days. Thus system researchers need to devote a lot of time to understand the impact of communication on the overall performance. Clearly, there is lack of benchmarks available for system researchers. Therefore, we propose TF-gRPC-Bench micro-benchmark suite that enables system researchers to quickly understand the impact of the underlying network and communication runtime on deep learning workloads. To achieve this, we first analyze the characteristics of TensorFlow workload over gRPC by training popular deep learning models. Then, we propose three micro-benchmarks that take account these workload characteristics. In addition, we comprehensively evaluate gRPC with TF-gRPC-Bench micro-benchmark suite on different clusters over Ethernet, IPoIB, and RDMA, and present the results.

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。

↑