arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1704.04560cs.DC
收藏

用户透明的分布式TensorFlow

User-transparent Distributed TensorFlow

Abhinav Vishnu, Joseph Manzano, Charles Siegel, Jeff Daily

摘要:

深度学习(DL)算法已成为数据分析的{ \em de facto}选择。几个DL实现-主要限于单个计算节点-如Caffe,TensorFlow,Theano和Torch已经变得容易使用。能够在大规模系统上执行的分布式DL实现对于解决由科学模拟和实验产生的大数据的计算需求变得越来越重要。然而,分布式DL实现的采用面临重大障碍:1)大多数实现要求DL分析师显著修改其代码-这是一个展示停止器,2)一些分布式DL实现面向云计算系统-这不足以在超级计算机等大规模并行系统上执行。这项工作解决了所有这些问题。我们通过在TensorFlow运行时本身中合并所需的更改来提供分布式内存DL实现。这大大降低了使用分布式TensorFlow实现的进入门槛。我们使用消息传递接口(MPI)--它提供了性能可移植性,特别是因为MPI特定的更改是从用户那里抽象出来的。最后,也可以说是最重要的一点,我们在{ \texttt http://hpc.pnl.gov/matex}上的机器学习Toolkit for Extreme Scale(MaTEX)的保护下,使我们的实现可用于更广泛的用途。我们将我们的实现称为MaTEx-TensorFlow。

英文摘要:

Deep Learning (DL) algorithms have become the {\em de facto} choice for data analysis. Several DL implementations -- primarily limited to a single compute node -- such as Caffe, TensorFlow, Theano and Torch have become readily available. Distributed DL implementations capable of execution on large scale systems are becoming important to address the computational needs of large data produced by scientific simulations and experiments. Yet, the adoption of distributed DL implementations faces significant impediments: 1) most implementations require DL analysts to modify their code significantly -- which is a show-stopper, 2) several distributed DL implementations are geared towards cloud computing systems -- which is inadequate for execution on massively parallel systems such as supercomputers. This work addresses each of these problems. We provide a distributed memory DL implementation by incorporating required changes in the TensorFlow runtime itself. This dramatically reduces the entry barrier for using a distributed TensorFlow implementation. We use Message Passing Interface (MPI) -- which provides performance portability, especially since MPI specific changes are abstracted from users. Lastly -- and arguably most importantly -- we make our implementation available for broader use, under the umbrella of Machine Learning Toolkit for Extreme Scale (MaTEx) at {\texttt http://hpc.pnl.gov/matex}. We refer to our implementation as MaTEx-TensorFlow.

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。

↑