DeepCEE: Efficient Cross-Region Model Distributed Training System under Heterogeneous GPUs and Networks

TL;DR

DeepCEE is a geo-distributed training system designed for heterogeneous GPUs and networks in cross-region environments, achieving significant throughput improvements by adaptive strategies and communication-centric design.

Contribution

It introduces a novel system tailored for cross-region, heterogeneous GPU environments with adaptive strategies and a communication-centric approach for efficient training.

Findings

Achieves 1.3-2.8x higher throughput than existing systems.

Effectively handles network variability with dynamic adaptation.

Utilizes device profiling for optimal parallel strategy derivation.

Abstract

Most existing training systems focus on a single region. In contrast, we envision that cross-region training offers more flexible GPU resource allocation and yields significant potential. However, the hierarchical cluster topology and unstable networks in the cloud-edge-end (CEE) environment, a typical cross-region scenario, pose substantial challenges to building an efficient and autonomous model training system. We propose DeepCEE, a geo-distributed model training system tailored for heterogeneous GPUs and networks in CEE environments. DeepCEE adopts a communication-centric design philosophy to tackle challenges arising from slow and unstable inter-region networks. It begins with a heterogeneous device profiler that identifies and groups devices based on both network and compute characteristics. Leveraging device groups, DeepCEE implements compact, zero-bubble pipeline parallelism, automatically deriving optimal parallel strategies. To further adapt to runtime variability, DeepCEE integrates a dynamic environment adapter that reacts to network fluctuations. Extensive evaluations demonstrate that DeepCEE achieves 1.3-2.8x higher training throughput compared to widely used and SOTA training systems.