Authors: Xiaohui Wang, Yang Wei, and Ying Xiong (ByteDance Ltd, AI Lab); Guyue Huang (University of California, Santa Barbara); Xian Qian (ByteDance Ltd, AI Lab); Yufei Ding (University of California, Santa Barbara); Mingxuan Wang (ByteDance Ltd, AI Lab); and Lei Li (University of California, Santa Barbara)
Abstract: Transformer-based neural models are used in many AI applications. Training these models is expensive, as it takes huge GPU resources and long duration. It is challenging because typical data like sentences have variable lengths, and Transformer's computation patterns are more complex than convolutional neural networks. Existing systems either only focus on model inference or optimization for only BERT-like encoder models.
In this paper, we present LightSeq2, a system to accelerate training for a general family of Transformer models on GPUs. We propose a series of GPU optimization techniques tailored to the specific computation flow and memory access patterns of Transformer models. LightSeq2 supports many model architectures, including BERT (encoder-only), GPT (decoder-only), Transformer (encoder-decoder), and vision Transformer. Our experiments for a variety of models and benchmarks show that LightSeq2 is consistently faster (1.4-3.5x) than previous systems on different GPUs. In particular, it gains 308% training speedup on WMT14 English-German benchmark.
Back to Technical Papers Archive Listing