Sign up for our newsletter and get the latest HPC news and analysis.
Send me information from insideHPC:


DK Panda Team Launches High-Performance Deep Learning Project

hidlDeep learning is one of the hottest topics at SC16. Now, DK Panda and his team at Ohio State University have announced an exciting new High-Performance Deep Learning project that aims to bring HPC technologies to the DL field.

Welcome to the High-Performance Deep Learning project created by the Network-Based Computing Laboratory of The Ohio State University. Availability of large data sets like ImageNet and massively parallel computation support in modern HPC devices like NVIDIA GPUs have fueled a renewed interest in Deep Learning (DL) algorithms. This has triggered the development of DL frameworks like Caffe, Torch, TensorFlow, and CNTK. However, most DL frameworks have been limited to a single node. The objective of the HiDL project is to exploit modern HPC technologies and solutions to scale out and accelerate DL frameworks.

According to Dr. Panda, as a first step, we have co-designed the popular Caffe with CUDA-aware MPI libraries (specifically with MVAPICH2-GDR 2.2 release).

OSU-Caffe 0.9 Features:

  • Based on Nvidia’s Caffe fork (caffe-0.14)
  • MPI-based distributed training support
  • Efficient scale-out support for multi-GPU nodes systems
  • New workflow to overlap the compute layers and the communication
  • Efficient parallel file readers to optimize I/O and data movement
    • Takes advantage of Lustre Parallel File System
  • Exploits efficient large message collectives in MVAPICH2-GDR 2.2
  • Tested with
    • Various CUDA-aware MPI libraries
    • CUDA 7.5
    • Various HPC Clusters with K80 GPUs, varying number of GPUs/node, and InfiniBand (FDR and EDR) adapters
DK Panda, Ohio State University

DK Panda, Ohio State University

For downloading OSU-Caffe 0.9 library and the associated user guides, please visit the following URL: http://hidl.cse.ohio-state.edu

DK Panda will present more details at several talks at SC16.

Sign up for our insideHPC Newsletter

Leave a Comment

*

Resource Links: