# ReCycle: Resilient Training of Large DNNs using Pipeline Adaptation

arXiv: https://arxiv.org/abs/2405.14009 · HTML: https://mlsanity.com/paper/2405.14009/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Builds on 40 of its 75 references that are in the map.
- First posted on arXiv 22 May 2024.

## Builds on (most cited first, 30 of 40)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [On the Opportunities and Risks of Foundation Models](https://mlsanity.com/paper/2108.07258/) (2021)
- [OPT: Open Pre-trained Transformer Language Models](https://mlsanity.com/paper/2205.01068/) (2022)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [BLOOM: A 176B-Parameter Open-Access Multilingual Language Model](https://mlsanity.com/paper/2211.05100/) (2022)
- [Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism](https://mlsanity.com/paper/1909.08053/) (2019)
- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour](https://mlsanity.com/paper/1706.02677/) (2017)
- [ZeRO: Memory Optimizations Toward Training Trillion Parameter Models](https://mlsanity.com/paper/1910.02054/) (2019)
- [PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel](https://mlsanity.com/paper/2304.11277/) (2023)
- [Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM](https://mlsanity.com/paper/2104.04473/) (2021)
- [ZeRO-Offload: Democratizing Billion-Scale Model Training](https://mlsanity.com/paper/2101.06840/) (2021)
- [Horovod: fast and easy distributed deep learning in TensorFlow](https://mlsanity.com/paper/1802.05799/) (2018)
- [One weird trick for parallelizing convolutional neural networks](https://mlsanity.com/paper/1404.5997/) (2014)
- [ZeRO-Infinity: Breaking the GPU Memory Wall for Extreme Scale Deep Learning](https://mlsanity.com/paper/2104.07857/) (2021)
- [TPU v4: An Optically Reconfigurable Supercomputer for Machine Learning with Hardware Support for Embeddings](https://mlsanity.com/paper/2304.01433/) (2023)
- [Beyond Data and Model Parallelism for Deep Neural Networks](https://mlsanity.com/paper/1807.05358/) (2018)
- [Mesh-TensorFlow: Deep Learning for Supercomputers](https://mlsanity.com/paper/1811.02084/) (2018)
- [Revisiting Distributed Synchronous SGD](https://mlsanity.com/paper/1702.05800/) (2017)
- [MegaScale: Scaling Large Language Model Training to More Than 10,000 GPUs](https://mlsanity.com/paper/2402.15627/) (2024)
- [PyTorch Distributed: Experiences on Accelerating Data Parallel Training](https://mlsanity.com/paper/2006.15704/) (2020)
- [Analysis of Large-Scale Multi-Tenant GPU Clusters for DNN Training Workloads](https://mlsanity.com/paper/1901.05758/) (2019)
- [Chimera: Efficiently Training Large-Scale Neural Networks with Bidirectional Pipelines](https://mlsanity.com/paper/2107.06925/) (2021)
- [Bamboo: Making Preemptible Instances Resilient for Affordable Training of Large DNNs](https://mlsanity.com/paper/2204.12013/) (2022)
- [Supporting Very Large Models using Automatic Dataflow Graph Partitioning](https://mlsanity.com/paper/1807.08887/) (2018)
- [Themis: Fair and Efficient GPU Cluster Scheduling](https://mlsanity.com/paper/1907.01484/) (2019)

## Built on by (most cited first, 14 of 14)

- [Mycroft: Tracing Dependencies in Collective Communication Towards Reliable LLM Training](https://mlsanity.com/paper/2509.03018/) (2025)
- [Training LLMs with Fault Tolerant HSDP on 100,000 GPUs](https://mlsanity.com/paper/2602.00277/) (2026)
- [Reliable and Resilient Collective Communication Library for LLM Training and Serving](https://mlsanity.com/paper/2512.25059/) (2025)
- [Sparse Checkpointing for Fast and Reliable MoE Training](https://mlsanity.com/paper/2412.15411/) (2024)
- [DIP: Efficient Large Multimodal Model Training with Dynamic Interleaved Pipeline](https://mlsanity.com/paper/2504.14145/) (2025)
- [Towards Resiliency in Large Language Model Serving with KevlarFlow](https://mlsanity.com/paper/2601.22438/) (2026)
- [OctoPipe: Reducing Pipeline Bubbles for Heterogeneous Models via Co-Optimizing Partitioning, Placement, and Scheduling](https://mlsanity.com/paper/2509.23722/) (2025)
- [Role-Based Fault Tolerance System for LLM RL Post-Training](https://mlsanity.com/paper/2512.22492/) (2025)
- [TrainMover: An Interruption-Resilient Runtime for ML Training](https://mlsanity.com/paper/2412.12636/) (2024)
- [LUMEN: Coordinated Failure Recovery for Distributed LLM Serving](https://mlsanity.com/paper/2606.17787/) (2026)
- [Nonuniform-Tensor-Parallelism: Mitigating GPU failure impact for Scaled-up LLM Training](https://mlsanity.com/paper/2504.06095/) (2025)
- [Rethinking Dynamic Networks and Heterogeneous Computing with Automatic Parallelization](https://mlsanity.com/paper/2506.02787/) (2025)
- [ReCoVer: Resilient LLM Pre-Training System via Fault-Tolerant Collective and Versatile Workload](https://mlsanity.com/paper/2605.11215/) (2026)
- [Leto: Fast In-Place Recovery for LLM Training on Surviving Hardware](https://mlsanity.com/paper/2610.00687/) (2026)

## Abstract (arXiv)

Training large Deep Neural Network (DNN) models requires thousands of GPUs over the course of several days or weeks. At this scale, failures are frequent and can have a big impact on training throughput. Utilizing spare GPU servers to mitigate performance loss becomes increasingly costly as model sizes grow. ReCycle is a system designed for efficient DNN training in the presence of failures, without relying on spare servers. It exploits the inherent functional redundancy in distributed training systems -- where servers across data-parallel groups store the same model parameters -- and pipeline schedule bubbles within each data-parallel group. When servers fails, ReCycle dynamically re-routes micro-batches to data-parallel peers, allowing for uninterrupted training despite multiple failures. However, this re-routing can create imbalances across pipeline stages, leading to reduced training throughput. To address this, ReCycle introduces two key optimizations that ensure re-routed micro-batches are processed within the original pipeline schedule's bubbles. First, it decouples the backward pass into two phases: one for computing gradients for the input and another for calculating gradients for the parameters. Second, it avoids synchronization across pipeline stages by staggering the optimizer step. Together, these optimizations enable adaptive pipeline schedules that minimize or even eliminate training throughput degradation during failures. We describe a prototype for ReCycle and show that it achieves high training throughput under multiple failures, outperforming recent proposals for fault-tolerant training such as Oobleck and Bamboo by up to 1.46× and 1.64×, respectively.