English
Related papers

Related papers: CrossPipe: Towards Optimal Pipeline Schedules for …

200 papers

Large model training often uses recomputation to alleviate memory pressure and pipelines to exploit the parallelism of data, tensors, and devices. However, existing recomputation approaches may incur high overhead when training real-world…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-03-31 Ping Chen , Wenjie Zhang , Shuibing He , Weijian Chen , Siling Yang , Kexin Huang , Yanlong Yin , Xuan Zhan , Yingjie Gu , Zhuwei Peng , Yi Zheng , Zhefeng Wang , Gang Chen

In the machine learning system, the hybrid model parallelism combining tensor parallelism (TP) and pipeline parallelism (PP) has become the dominant solution for distributed training of Large Language Models~(LLMs) and Multimodal LLMs…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-11-03 Mengshi Qi , Jiaxuan Peng , Jie Zhang , Juan Zhu , Yong Li , Huadong Ma

Training large language models (LLMs) is known to be challenging because of the huge computational and memory capacity requirements. To address these issues, it is common to use a cluster of GPUs with 3D parallelism, which splits a model…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-05-29 Jinkyu Yim , Jaeyong Song , Yerim Choi , Jaebeen Lee , Jaewon Jung , Hongsun Jang , Jinho Lee

Pipeline parallelism is one of the key components for large-scale distributed training, yet its efficiency suffers from pipeline bubbles which were deemed inevitable. In this work, we introduce a scheduling strategy that, to our knowledge,…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-01-22 Penghui Qi , Xinyi Wan , Guangxing Huang , Min Lin

Large-language-models (LLMs) demonstrate enormous utility in long-context tasks which require processing prompts that consist of tens to hundreds of thousands of tokens. However, existing LLM training libraries do not provide easy to use…

Machine Learning · Computer Science 2026-05-01 Ahan Gupta , Zhihao Wang , Neel Dani , Masahiro Tanaka , Olatunji Ruwase , Minjia Zhang

The rapid growth of deep learning models has increased the demand for efficient distributed training strategies. Fully sharded approaches like ZeRO-3 and FSDP partition model parameters across GPUs and apply optimizations such as…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-02-20 Masahiro Tanaka , Du Li , Umesh Chand , Ali Zafar , Haiying Shen , Olatunji Ruwase

Large-scale deep learning models contribute to significant performance improvements on varieties of downstream tasks. Current data and model parallelism approaches utilize model replication and partition techniques to support the…

Distributed, Parallel, and Cluster Computing · Computer Science 2023-05-19 Youhe Jiang , Fangcheng Fu , Xupeng Miao , Xiaonan Nie , Bin Cui

Large-scale deep learning models contribute to significant performance improvements on varieties of downstream tasks. Current data and model parallelism approaches utilize model replication and partition techniques to support the…

Distributed, Parallel, and Cluster Computing · Computer Science 2023-05-22 Youhe Jiang , Fangcheng Fu , Xupeng Miao , Xiaonan Nie , Bin Cui

Scaling long-context ability is essential for Large Language Models (LLMs). To amortize the memory consumption across multiple devices in long-context training, inter-data partitioning (a.k.a. Data Parallelism) and intra-data partitioning…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-09-01 Hao Ge , Junda Feng , Qi Huang , Fangcheng Fu , Xiaonan Nie , Lei Zuo , Haibin Lin , Bin Cui , Xin Liu

Frontier models increasingly adopt Mixture-of-Experts (MoE) architectures to achieve large-model performance at reduced cost. However, training MoE models on HPC platforms is hindered by large memory footprints, frequent large-scale…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-05-07 Sajal Dash , Feiyi Wang

Asynchronous pipeline model parallelism with a "1F1B" (one forward, one backward) schedule generates little bubble overhead and always provides quite a high throughput. However, the "1F1B" schedule inevitably leads to weight inconsistency…

Machine Learning · Computer Science 2025-02-18 Lei Guan , Dongsheng Li , Yongle Chen , Jiye Liang , Wenjian Wang , Xicheng Lu

Data and pipeline parallelism are key strategies for scaling neural network training across distributed devices, but their high communication cost necessitates co-located computing clusters with fast interconnects, limiting their…

Automated data preparation is crucial for democratizing machine learning, yet existing reinforcement learning (RL) based approaches suffer from inefficient exploration in the vast space of possible preprocessing pipelines. We present…

Databases · Computer Science 2025-07-21 Jing Chang , Chang Liu , Jinbin Huang , Rui Mao , Jianbin Qin

Modern recommendation models have increased to trillions of parameters. As cluster scales expand to O(1k), distributed training bottlenecks shift from computation and memory to data movement, especially lookup and communication latency…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-04-09 Zhida Jiang , Zhaolong Xing , Huichao Chai , Tianxing Sun , Qiang Peng , Baopeng Yuan , Jiaxing Wang , Hua Du , Zhixin Wu , Xuemiao Li , Yikui Cao , Xinyu Liu , Yongxiang Feng , Zhen Chen , Ke Zhang

Deep neural networks (DNNs) continue to grow rapidly in size, making them infeasible to train on a single device. Pipeline parallelism is commonly used in existing DNN systems to support large-scale DNN training by partitioning a DNN into…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-10-29 Byungsoo Jeon , Mengdi Wu , Shiyi Cao , Sunghyun Kim , Sunghyun Park , Neeraj Aggarwal , Colin Unger , Daiyaan Arfeen , Peiyuan Liao , Xupeng Miao , Mohammad Alizadeh , Gregory R. Ganger , Tianqi Chen , Zhihao Jia

We introduce Breadth-First Pipeline Parallelism, a novel training schedule which optimizes the combination of pipeline and data parallelism. Breadth-First Pipeline Parallelism lowers training time, cost and memory usage by combining a high…

Distributed, Parallel, and Cluster Computing · Computer Science 2023-07-10 Joel Lamy-Poirier

Collaborative machine learning (CML) techniques, such as federated learning, have been proposed to train deep learning models across multiple mobile devices and a server. CML techniques are privacy-preserving as a local model that is…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-06-26 Zihan Zhang , Philip Rodgers , Peter Kilpatrick , Ivor Spence , Blesson Varghese

Diffusion models have emerged as dominant performers for image generation. To support training large diffusion models, this paper studies pipeline parallel training of diffusion models and proposes DiffusionPipe, a synchronous pipeline…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-05-03 Ye Tian , Zhen Jia , Ziyue Luo , Yida Wang , Chuan Wu

Consider a device that is connected to an edge processor via a communication channel. The device holds local data that is to be offloaded to the edge processor so as to train a machine learning model, e.g., for regression or classification.…

Machine Learning · Computer Science 2019-06-13 Nicolas Skatchkovsky , Osvaldo Simeone

Pipeline parallelism enables training models that exceed single-device memory, but practical throughput remains limited by pipeline bubbles. Although parameter freezing can improve training throughput by adaptively skipping backward…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-02-09 Seonghye Cho , Jaemin Han , Hyunjin Kim , Euisoo Jung , Jae-Gil Lee