English
Related papers

Related papers: Seq1F1B: Efficient Sequence-Level Pipeline Paralle…

200 papers

Many state-of-the-art ML results have been obtained by scaling up the number of parameters in existing models. However, parameters and activations for such large models often do not fit in the memory of a single accelerator device; this…

Machine Learning · Computer Science 2021-07-23 Deepak Narayanan , Amar Phanishayee , Kaiyu Shi , Xie Chen , Matei Zaharia

Model parallelism has become a necessity for training modern large-scale deep language models. In this work, we identify a new and orthogonal dimension from existing model parallel approaches: it is possible to perform pipeline parallelism…

Machine Learning · Computer Science 2021-09-29 Zhuohan Li , Siyuan Zhuang , Shiyuan Guo , Danyang Zhuo , Hao Zhang , Dawn Song , Ion Stoica

Training language models on long sequence data is a demanding requirement for enhancing the model's capability on complex tasks, e.g., long-chain reasoning. However, as the sequence length scales up, the memory cost for storing activation…

Machine Learning · Computer Science 2025-06-04 Qijun Luo , Mengqi Li , Lei Zhao , Xiao Li

Large language models (LLMs) have revolutionized the field of AI, demonstrating unprecedented capacity across various tasks. However, the inference process for LLMs comes with significant computational costs. In this paper, we propose an…

Computation and Language · Computer Science 2023-05-30 Zangwei Zheng , Xiaozhe Ren , Fuzhao Xue , Yang Luo , Xin Jiang , Yang You

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

Recent advancements in large language models (LLMs) with billions of parameters have improved performance in various applications, but their inference processes demand significant energy and computational resources. In contrast, the human…

Machine Learning · Computer Science 2025-04-11 Xingrun Xing , Boyan Gao , Zheng Zhang , David A. Clifton , Shitao Xiao , Li Du , Guoqi Li , Jiajun Zhang

With the rapid adoption of large language models (LLMs) in recommendation systems, the computational and communication bottlenecks caused by their massive parameter sizes and large data volumes have become increasingly prominent. This paper…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-06-25 Haowei Yang , Yu Tian , Zhongheng Yang , Zhao Wang , Chengrui Zhou , Dannier Li

Long context training is crucial for LLM's context extension. Existing schemes, such as sequence parallelism, incur substantial communication overhead. Pipeline parallelism (PP) reduces this cost, but its effectiveness hinges on…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-04-28 Shiju Wang , Yujie Wang , Ao Sun , Fangcheng Fu , Zijian Zhu , Bin Cui , Xu Han , Kaisheng Ma

Pipeline parallelism is a crucial paradigm for large-scale model training. However, imbalances in memory footprint across stages can lead to significant GPU memory wastage, limiting the model sizes that pipeline parallelism can effectively…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-05-12 Xuan Peng , Xuanhua Shi , Haolin Zhang , Yunfei Zhao , Xuehai Qian

Large Language Models (LLMs) have exhibited exceptional performance across a spectrum of natural language processing tasks. However, their substantial sizes pose considerable challenges, particularly in computational demands and inference…

Computation and Language · Computer Science 2025-06-03 Guoxuan Chen , Han Shi , Jiawei Li , Yihang Gao , Xiaozhe Ren , Yimeng Chen , Xin Jiang , Zhenguo Li , Weiyang Liu , Chao Huang

Currently, training large-scale deep learning models is typically achieved through parallel training across multiple GPUs. However, due to the inherent communication overhead and synchronization delays in traditional model parallelism…

Computer Vision and Pattern Recognition · Computer Science 2024-11-21 Xiuyuan Guo , Chengqi Xu , Guinan Guo , Feiyu Zhu , Changpeng Cai , Peizhe Wang , Xiaoming Wei , Junhao Su , Jialin Gao

Large Language Models (LLMs) with long context capabilities are integral to complex tasks in natural language processing and computational biology, such as text generation and protein sequence analysis. However, training LLMs directly on…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-05-14 Jinghan Yao , Sam Ade Jacobs , Masahiro Tanaka , Olatunji Ruwase , Hari Subramoni , Dhabaleswar K. Panda

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

Pipeline parallelism enables efficient training of Large Language Models (LLMs) on large-scale distributed accelerator clusters. Yet, pipeline bubbles during startup and tear-down reduce the utilization of accelerators. Although efficient…

Machine Learning · Computer Science 2023-05-16 Kazuki Osawa , Shigang Li , Torsten Hoefler

Large-scale language models have become increasingly challenging and expensive to train. Among various methods addressing this issue, Pipeline Parallelism has been widely employed to accommodate massive model weights within limited GPU…

Distributed, Parallel, and Cluster Computing · Computer Science 2023-08-31 Ziming Liu , Shenggan Cheng , Haotian Zhou , Yang You

With the increasing scale of models, the need for efficient distributed training has become increasingly urgent. Recently, many synchronous pipeline parallelism approaches have been proposed to improve training throughput. However, these…

Machine Learning · Computer Science 2024-10-28 Houming Wu , Ling Chen , Wenjie Yu

Efficiently training LLMs with long sequences is important yet challenged by the massive computation and memory requirements. Sequence parallelism has been proposed to tackle these problems, but existing methods suffer from scalability or…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-06-27 Diandian Gu , Peng Sun , Qinghao Hu , Ting Huang , Xun Chen , Yingtong Xiong , Guoteng Wang , Qiaoling Chen , Shangchun Zhao , Jiarui Fang , Yonggang Wen , Tianwei Zhang , Xin Jin , Xuanzhe Liu

Transformer achieves promising results on various tasks. However, self-attention suffers from quadratic memory requirements with respect to the sequence length. Existing work focuses on reducing time and space complexity from an algorithm…

Machine Learning · Computer Science 2022-05-24 Shenggui Li , Fuzhao Xue , Chaitanya Baranwal , Yongbin Li , Yang You

Fueled by their remarkable ability to tackle diverse tasks across multiple domains, large language models (LLMs) have grown at an unprecedented rate, with some recent models containing trillions of parameters. This growth is accompanied by…

Machine Learning · Computer Science 2025-05-30 Athanasios Glentis , Jiaxiang Li , Qiulin Shang , Andi Han , Ioannis Tsaknakis , Quan Wei , Mingyi Hong

Pipeline parallelism is a key technique for scaling large-model training, but modern workloads exhibit runtime variability in computation and communication. Existing pipeline systems typically consume static, profiled, or adaptively…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-05-19 Ruitao Liu , Xinyang Tian , Shuo Chen , Tingrui Zhang , Guang Yang , Alan Zhao , Wei Xu