HetSeq: Distributed GPU Training on Heterogeneous Infrastructure

TL;DR

HetSeq is a software tool that enables training large neural networks efficiently on heterogeneous GPU systems, overcoming infrastructure limitations common in many organizations.

Contribution

HetSeq extends PyTorch to support distributed training on heterogeneous hardware, allowing scalable training across diverse systems.

Findings

HetSeq successfully scales transformer translation models.

HetSeq enables training of BERT on heterogeneous infrastructure.

The package is publicly available for broader use.

Abstract

Modern deep learning systems like PyTorch and Tensorflow are able to train enormous models with billions (or trillions) of parameters on a distributed infrastructure. These systems require that the internal nodes have the same memory capacity and compute performance. Unfortunately, most organizations, especially universities, have a piecemeal approach to purchasing computer systems resulting in a heterogeneous infrastructure, which cannot be used to compute large models. The present work describes HetSeq, a software package adapted from the popular PyTorch package that provides the capability to train large neural network models on heterogeneous infrastructure. Experiments with transformer translation and BERT language model shows that HetSeq scales over heterogeneous systems. HetSeq can be easily extended to other models like image classification. Package with supported document is publicly available at https://github.com/yifding/hetseq.