# FlexMoE: Scaling Large-scale Sparse Pre-trained Model Training via Dynamic Device Placement

arXiv: https://arxiv.org/abs/2304.03946 · HTML: https://mlsanity.com/paper/2304.03946/

## Key facts

- Cited by 53 papers in the map (data through 14 Oct 2026).
- Placed in Expert routing strategies, one of 12 directions in Mixture-of-experts.
- Builds on 27 of its 48 references that are in the map.
- First posted on arXiv 8 Apr 2023.

## Placed in

- Mixture-of-experts / [Expert routing strategies](https://mlsanity.com/moe/expert-routing-strategies/)

## Builds on (most cited first, 27 of 27)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systems](https://mlsanity.com/paper/1905.00537/) (2019)
- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [GLaM: Efficient Scaling of Language Models with Mixture-of-Experts](https://mlsanity.com/paper/2112.06905/) (2021)
- [Scaling Vision with Sparse Mixture of Experts](https://mlsanity.com/paper/2106.05974/) (2021)
- [Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM](https://mlsanity.com/paper/2104.04473/) (2021)
- [DeepSpeed-MoE: Advancing Mixture-of-Experts Inference and Training to Power Next-Generation AI Scale](https://mlsanity.com/paper/2201.05596/) (2022)
- [ST-MoE: Designing Stable and Transferable Sparse Expert Models](https://mlsanity.com/paper/2202.08906/) (2022)
- [BASE Layers: Simplifying Training of Large, Sparse Models](https://mlsanity.com/paper/2103.16716/) (2021)
- [Hash Layers For Large Sparse Models](https://mlsanity.com/paper/2106.04426/) (2021)
- [Tutel: Adaptive Mixture-of-Experts at Scale](https://mlsanity.com/paper/2206.03382/) (2022)
- [StableMoE: Stable Routing Strategy for Mixture of Experts](https://mlsanity.com/paper/2204.08396/) (2022)
- [M6: A Chinese Multimodal Pretrainer](https://mlsanity.com/paper/2103.00823/) (2021)
- [Galvatron: Efficient Transformer Training over Multiple GPUs Using Automatic Parallelism](https://mlsanity.com/paper/2211.13878/) (2022)
- [HetuMoE: An Efficient Trillion-scale Mixture-of-Expert Distributed Training System](https://mlsanity.com/paper/2203.14685/) (2022)
- [HET: Scaling out Huge Embedding Model Training via Cache-enabled Distributed Framework](https://mlsanity.com/paper/2112.07221/) (2021)
- [Towards Communication-efficient Vertical Federated Learning Training via Cache-enabled Local Updates](https://mlsanity.com/paper/2207.14628/) (2022)
- [Angel-PTM: A Scalable and Economical Large-scale Pre-training System in Tencent](https://mlsanity.com/paper/2303.02868/) (2023)
- [An Experimental Evaluation of Large Scale GBDT Systems](https://mlsanity.com/paper/1907.01882/) (2019)

## Built on by (most cited first, 30 of 53)

- [Towards Efficient Generative Large Language Model Serving: A Survey from Algorithms to Systems](https://mlsanity.com/paper/2312.15234/) (2023)
- [SpotServe: Serving Generative Large Language Models on Preemptible Instances](https://mlsanity.com/paper/2311.15566/) (2023)
- [Advancing Transformer Architecture in Long-Context Large Language Models: A Comprehensive Survey](https://mlsanity.com/paper/2311.12351/) (2023)
- [KeyVideoLLM: Towards Large-scale Video Keyframe Selection](https://mlsanity.com/paper/2407.03104/) (2024)
- [A Survey on Inference Optimization Techniques for Mixture of Experts Models](https://mlsanity.com/paper/2412.14219/) (2024)
- [LocMoE: A Low-Overhead MoE for Large Language Model Training](https://mlsanity.com/paper/2401.13920/) (2024)
- [Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity](https://mlsanity.com/paper/2309.10285/) (2023)
- [FSMoE: A Flexible and Scalable Training System for Sparse Mixture-of-Experts Models](https://mlsanity.com/paper/2501.10714/) (2025)
- [FlashMoE: Fast Distributed MoE in a Single Kernel](https://mlsanity.com/paper/2506.04667/) (2025)
- [GRACE-MoE: Grouping and Replication with Locality-Aware Routing for Efficient Distributed MoE Inference](https://mlsanity.com/paper/2509.25041/) (2025)
- [Experts Weights Averaging: A New General Training Scheme for Vision Transformers](https://mlsanity.com/paper/2308.06093/) (2023)
- [Pangu Ultra MoE: How to Train Your Big MoE on Ascend NPUs](https://mlsanity.com/paper/2505.04519/) (2025)
- [Prediction Is All MoE Needs: Expert Load Distribution Goes from Fluctuating to Stabilizing](https://mlsanity.com/paper/2404.16914/) (2024)
- [Synth-Empathy: Towards High-Quality Synthetic Empathy Data](https://mlsanity.com/paper/2407.21669/) (2024)
- [Semantic Parallelism: Redefining Efficient MoE Inference via Model-Data Co-Scheduling](https://mlsanity.com/paper/2503.04398/) (2025)
- [UltraEP: Unleash MoE Training and Inference on Rack-Scale Nodes with Near-Optimal Load Balancing](https://mlsanity.com/paper/2606.04101/) (2026)
- [PAS: Data-Efficient Plug-and-Play Prompt Augmentation System](https://mlsanity.com/paper/2407.06027/) (2024)
- [Designing Large Foundation Models for Efficient Training and Inference: A Survey](https://mlsanity.com/paper/2409.01990/) (2024)
- [Optimizing Mixture-of-Experts Inference Time Combining Model Deployment and Communication Scheduling](https://mlsanity.com/paper/2410.17043/) (2024)
- [HeterMoE: Efficient Training of Mixture-of-Experts Models on Heterogeneous GPUs](https://mlsanity.com/paper/2504.03871/) (2025)
- [Accelerating Edge Inference for Distributed MoE Models with Latency-Optimized Expert Placement](https://mlsanity.com/paper/2508.12851/) (2025)
- [SynthVLM: Towards High-Quality and Efficient Synthesis of Image-Caption Datasets for Vision-Language Models](https://mlsanity.com/paper/2407.20756/) (2024)
- [EPS-MoE: Expert Pipeline Scheduler for Cost-Efficient MoE Inference](https://mlsanity.com/paper/2410.12247/) (2024)
- [LAER-MoE: Load-Adaptive Expert Re-layout for Efficient Mixture-of-Experts Training](https://mlsanity.com/paper/2602.11686/) (2026)
- [SYMI: Efficient Mixture-of-Experts Training via Model and Optimizer State Decoupling](https://mlsanity.com/paper/2504.19925/) (2025)
- [Efficient MoE Serving in the Memory-Bound Regime: Balance Activated Experts, Not Tokens](https://mlsanity.com/paper/2512.09277/) (2025)
- [Grouter: Decoupling Routing from Representation for Accelerated MoE Training](https://mlsanity.com/paper/2603.06626/) (2026)
- [MoE-GPS: Guidlines for Prediction Strategy for Dynamic Expert Duplication in MoE Load Balancing](https://mlsanity.com/paper/2506.07366/) (2025)
- [HierMoE: Accelerating MoE Training with Hierarchical Token Deduplication and Expert Swap](https://mlsanity.com/paper/2508.09591/) (2025)
- [Sparse Checkpointing for Fast and Reliable MoE Training](https://mlsanity.com/paper/2412.15411/) (2024)

## Abstract (arXiv)

With the increasing data volume, there is a trend of using large-scale pre-trained models to store the knowledge into an enormous number of model parameters. The training of these models is composed of lots of dense algebras, requiring a huge amount of hardware resources. Recently, sparsely-gated Mixture-of-Experts (MoEs) are becoming more popular and have demonstrated impressive pretraining scalability in various downstream tasks. However, such a sparse conditional computation may not be effective as expected in practical systems due to the routing imbalance and fluctuation problems. Generally, MoEs are becoming a new data analytics paradigm in the data life cycle and suffering from unique challenges at scales, complexities, and granularities never before possible. In this paper, we propose a novel DNN training framework, FlexMoE, which systematically and transparently address the inefficiency caused by dynamic dataflow. We first present an empirical analysis on the problems and opportunities of training MoE models, which motivates us to overcome the routing imbalance and fluctuation problems by a dynamic expert management and device placement mechanism. Then we introduce a novel scheduling module over the existing DNN runtime to monitor the data flow, make the scheduling plans, and dynamically adjust the model-to-hardware mapping guided by the real-time data traffic. A simple but efficient heuristic algorithm is exploited to dynamically optimize the device placement during training. We have conducted experiments on both NLP models (e.g., BERT and GPT) and vision models (e.g., Swin). And results show FlexMoE can achieve superior performance compared with existing systems on real-world workloads -- FlexMoE outperforms DeepSpeed by 1.70x on average and up to 2.10x, and outperforms FasterMoE by 1.30x on average and up to 1.45x.