BOOST: BOttleneck-Optimized Scalable Training Framework for Low-Rank Large Language Models
This work addresses the computational and communication bottlenecks in scaling low-rank large language model training, offering a domain-specific optimization for faster and more efficient pre-training.
The paper tackles the poor scalability of low-rank bottleneck architectures in transformer model pre-training by proposing BOOST, an efficient training framework that achieves 1.46-1.91× speedup over full-rank baselines and 1.87-2.27× speedup over naive low-rank methods with improved GPU utilization and reduced communication.
The scale of transformer model pre-training is constrained by the increasing computation and communication cost. Low-rank bottleneck architectures offer a promising solution to significantly reduce the training time and memory footprint with minimum impact on accuracy. Despite algorithmic efficiency, bottleneck architectures scale poorly under standard tensor parallelism. Simply applying 3D parallelism designed for full-rank methods leads to excessive communication and poor GPU utilization. To address this limitation, we propose BOOST, an efficient training framework tailored for large-scale low-rank bottleneck architectures. BOOST introduces a novel Bottleneck-aware Tensor Parallelism, and combines optimizations such as online-RMSNorm, linear layer grouping, and low-rank activation checkpointing to achieve end-to-end training speedup. Evaluations on different low-rank bottleneck architectures demonstrate that BOOST achieves 1.46-1.91$\times$ speedup over full-rank model baselines and 1.87-2.27$\times$ speedup over low-rank model with naively integrated 3D parallelism, with improved GPU utilization and reduced communication overhead.