Merak: An Efficient Distributed DNN Training Framework with Automated 3D Parallelism for Giant Foundation Models

Abstract

Foundation models are becoming the dominant deep learning technologies. Pretraining a foundation model is always time-consumed due to the large scale of both the model parameter and training dataset. Besides being computing-intensive, the training process is extremely memory-intensive and communication-intensive. These features make it necessary to apply 3D parallelism, which integrates data parallelism, pipeline model parallelism and tensor model parallelism, to achieve high training efficiency. To achieve this goal, some custom software frameworks such as Megatron-LM and DeepSpeed are developed. However, current 3D parallelism frameworks still meet two issues: i) they are not transparent to model developers, which need to manually modify the model to parallelize training. ii) their utilization of computation, GPU memory and network bandwidth are not sufficient. We propose Merak, an automated 3D parallelism deep learning training framework with high resource utilization. Merak automatically deploys with an automatic model partitioner, which uses a graph sharding algorithm on a proxy representation of the model. Merak also presents the non-intrusive API for scaling out foundation model training with minimal code modification. In addition, we design a high-performance 3D parallel runtime engine in Merak. It uses several techniques to exploit available training resources, including shifted critical path pipeline schedule that brings a higher computation utilization, stage-aware recomputation that makes use of idle worker memory, and sub-pipelined tensor model parallelism that overlaps communication and computation. Experiments on 64 GPUs show Merak can speedup the training performance over the state-of-the-art 3D parallelism frameworks of models with 1.5, 2.5, 8.3, and 20 billion parameters by up to 1.42X, 1.39X, 1.43X, and 1.61X, respectively.

References in corpus (9)

• On the Opportunities and Risks of Foundation Models
• Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, A Large-Scale Generative Language Model
• Chimera: Efficiently Training Large-Scale Neural Networks with Bidirectional Pipelines
• PanGu-$\alpha$: Large-scale Autoregressive Pretrained Chinese Language Models with Auto-parallel Computation
• PatrickStar: Parallel Training of Pre-trained Models via Chunk-based Memory Management
• OneFlow: Redesign the Distributed Deep Learning Framework from Scratch
• Colossal-AI: A Unified Deep Learning System For Large-Scale Parallel Training
• Maximizing Parallelism in Distributed Training for Huge Neural Networks
• Amazon SageMaker Model Parallelism: A General and Flexible Framework for Large Model Training

Cited by in corpus (4)

• Oobleck: Resilient Distributed Training of Large Models Using Pipeline Templates
• Reducing Energy Bloat in Large Model Training
• Mist: Efficient Distributed Training of Large Language Models via Memory-Parallelism Co-Optimization
• Oases: Efficient Large-Scale Model Training on Commodity Servers via Overlapped and Automated Tensor Model Parallelism