Login / Signup

Efficient Large Message Broadcast using NCCL and CUDA-Aware MPI for Deep Learning.

A. A. AwanKhaled HamidoucheAkshay VenkateshDhabaleswar K. Panda
Published in: EuroMPI (2016)
Keyphrases
  • deep learning
  • general purpose
  • parallel implementation
  • email
  • parallel computing
  • unsupervised feature learning
  • feature extraction
  • viewpoint
  • image features
  • higher order
  • parallel algorithm