AMP: Automatically Finding Model Parallel Strategies with Heterogeneity Awareness
This addresses the need for efficient distributed training without expert tuning, especially for complex models and heterogeneous clusters, though it is incremental as it builds on existing model-parallel systems.
The paper tackles the challenge of automatically finding model-parallel execution strategies for training large ML models, particularly in heterogeneous environments, and shows that AMP achieves up to 1.77x higher throughput compared to state-of-the-art systems.
Scaling up model sizes can lead to fundamentally new capabilities in many machine learning (ML) tasks. However, training big models requires strong distributed system expertise to carefully design model-parallel execution strategies that suit the model architectures and cluster setups. In this paper, we develop AMP, a framework that automatically derives such strategies. AMP identifies a valid space of model parallelism strategies and efficiently searches the space for high-performed strategies, by leveraging a cost model designed to capture the heterogeneity of the model and cluster specifications. Unlike existing methods, AMP is specifically tailored to support complex models composed of uneven layers and cluster setups with more heterogeneous accelerators and bandwidth. We evaluate AMP on popular models and cluster setups from public clouds and show that AMP returns parallel strategies that match the expert-tuned strategies on typical cluster setups. On heterogeneous clusters or models with heterogeneous architectures, AMP finds strategies with 1.54x and 1.77x higher throughput than state-of-the-art model-parallel systems, respectively.