Training transformer models requires substantial GPU compute and memory resources. In homogeneous clusters, distributed strategies allocate resources evenly, but this approach is inefficient for heterogeneous clusters, where GPUs differ in power and memory. As high-end GPUs are costly and limited in availability, heterogeneous clusters with diverse GPU types are becoming more common. Existing methods attempt to balance compute across GPUs based on capacity but often underutilize compute due to memory constraints. We present Cephalo, a system that optimizes compute and memory usage by decoupling compute distribution from training state assignment. Cephalo outperforms state-of-the-art methods by achieving significantly higher training throughput while supporting larger models and batch sizes.

翻译：训练Transformer模型需要大量的GPU计算和内存资源。在异构集群中，GPU的计算能力和内存各不相同，而现有同构集群的分布式策略采用均匀分配资源的方式，在此类场景下效率低下。由于高端GPU成本高昂且供应有限，由不同类型GPU组成的异构集群正变得越来越普遍。现有方法尝试根据GPU容量平衡计算负载，但常因内存限制导致计算资源利用不足。本文提出Cephalo系统，通过将计算分布与训练状态分配解耦，实现计算与内存使用的协同优化。实验表明，Cephalo在支持更大模型和批处理规模的同时，训练吞吐量显著优于现有最优方法。