2017

PowerAI DDL

Cho, Minsik, Finkler, Ulrich, Kumar, Sameer et al.

Understand

As deep neural networks become more complex and input datasets grow larger, it can take days or even weeks to train a deep neural network to the desired accuracy.

  • Therefore, distributed Deep Learning at a massive scale is a critical capability, since it offers the potential to reduce the training time from weeks to hours.
  • In this paper, we present a software-hardware co-optimized distributed Deep Learning system that can achieve near-linear scaling up to hundreds of GPUs.
  • The core algorithm is a multi-ring communication pattern that provides a good tradeoff between latency and bandwidth and adapts to a variety of system configurations.

Built on

  • M. Barnett, R. Littlefield, D. Payne, and R. van de Geijn. Global combine on mesh architecture with wormhole routing

    1993

    Earlier work this paper cites.

  • DW Walker, JJ Dongarra. MPI: a standard message passing interface

    1996

    Earlier work this paper cites.

Similar

  • J. Dean, G. Corrado, R. Monga, K. Chen, M. Devin, M. Mao, M. Ranzato, A. Senior, P. Tucker, K. Yang, Q. Le, A. Ng. Large Scale Distributed Deep Networks

    2012

    Cited alongside, same era.

  • Priya Goyal, Piotr Dollar, Ross Girshick, Pieter Noordhuis, Lukasz Wesolowski, Aapo Kyrola, Andrew Tulloch, Yangqing Jia, Kaiming He. Accurate, Large Minibatch SGD: Training Imagenet in 1 Hour

    Cited in the paper.

  • Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun. Deep Residual Learning for Image Recognition

    Original

    Cited in the paper.

  • IBM Spectrum MPI: User’s Guide, GC27-8265-01

    Cited in the paper.

  • https://github.com/torch/torch7

    Cited in the paper.

  • Yangqing Jia, Evan Shelhamer, Jeff Donahue, Sergey Karayev, Jonathan Long, Ross Girshick, Sergio Guadarrama, Trevor Darrell. Caffe: Convolutional Architecture for Fast Feature Embedding

    Original

    Cited in the paper.

  • Mark Harris. NVIDIA DGX-1: The Fastest Deep Learning System. https://devblogs.nvidia.com/parallelforall/dgx-1-fastest-deep-learning-system

    Cited in the paper.

  • Trishul Chilimbi, Yutaka Suzue, Johnson Apacible, Karthik Kalyanaraman. Project Adam: building an efficient and scalable Deep Learning training system

    Cited in the paper.

  • https://github.com/PPC64/torch-distro

    Cited in the paper.

  • https://github.com/soumith/imagenet-multiGPU.torch

    Cited in the paper.

  • K. He, X. Zhang, S. Ren, and J. Sun. Deep residual learning for image recognition

    Cited in the paper.

  • Priya Goyal, Piotr Dollar, Ross Girshick, Pieter Noordhuis, Lukasz Wesolowski, Aapo Kyrola, Andrew Tulloch, Yangqing Jia, Kaiming He. Accurate, Large Minibatch SGD: Training Imagenet in 1 Hour. https://research.fb.com/publications/imagenet1kin1h

    Cited in the paper.

Then

  • Martin Abadi, Paul Barham, Jianmin Chen, Zhifeng Chen, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Geoffrey Irving, Michael Isard, Manjunath Kudlur, Josh Levenberg, Rajat Monga, Sherry Moore, Derek G. Murray, Benoit Steiner, Paul Tucker, Vijay Vasudevan, Pete Warden, Martin Wicke, Yuan Yu, Xiaoqiang Zheng. TensorFlow: A system for large-scale machine learning

    2015

    Later among the works it cites.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…