リングでクルクル廻すのね。
Baiduのサイト:Bringing HPC Techniques to Deep Learning
引用 The ring allreduce algorithm could speed up the training of an example neural network by 31x across 40 GPUs, compared to using a single GPU.
単一GPUに対して、40GPUで31倍になった例もあったって。
引用 Baidu’s SVAIL used the ring allreduce algorithm to train state of the art speech recognition models, and now it hopes that others will take advantage of the group’s implementation to do even more interesting things. The group released its ring allreduce implementation as both a standalone C++ library as well as a patch for TensorFlow.とあったので、Google君に聞いたら、ありました。