Poster
in
Workshop: Has it Trained Yet? A Workshop for Algorithmic Efficiency in Practical Neural Network Training

Breadth-first pipeline parallelism

Joel Lamy-Poirier

Project Page [ OpenReview]

Abstract

We introduce Breadth-First Pipeline Parallelism, a novel training schedule which optimizes the combination of pipeline and data parallelism. Breadth-First Pipeline Parallelism lowers the training time, cost and memory usage by combining a high GPU utilization with a small batch size per GPU, and by making use of fully sharded data parallelism. Experimentally, we observed increases of up to 53% in training speed.

Chat is not available.