English
Related papers

Related papers: Breadth-First Pipeline Parallelism

200 papers

Training and deploying large-scale machine learning models is time-consuming, requires significant distributed computing infrastructures, and incurs high operational costs. Our analysis, grounded in real-world large model training on…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-06-12 Samuel Hsia , Alicia Golden , Bilge Acun , Newsha Ardalani , Zachary DeVito , Gu-Yeon Wei , David Brooks , Carole-Jean Wu

This paper presents a low-cost network architecture for training large language models (LLMs) at hyperscale. We study the optimal parallelization strategy of LLMs and propose a novel datacenter network design tailored to LLM's unique…

Networking and Internet Architecture · Computer Science 2024-09-17 Weiyang Wang , Manya Ghobadi , Kayvon Shakeri , Ying Zhang , Naader Hasani

Heavy communication, in particular, collective operations, can become a critical performance bottleneck in scaling the training of billion-parameter neural networks to large-scale parallel systems. This paper introduces a four-dimensional…

Machine Learning · Computer Science 2024-05-15 Siddharth Singh , Prajwal Singhania , Aditya K. Ranjan , Zack Sating , Abhinav Bhatele

As Deep Neural Networks (DNNs) grow in size and complexity, they often exceed the memory capacity of a single accelerator, necessitating the sharding of model parameters across multiple accelerators. Pipeline parallelism is a commonly used…

Machine Learning · Computer Science 2024-05-29 Christopher Rae , Joseph K. L. Lee , James Richings

The increasing complexity of modern deep neural network models and the expanding sizes of datasets necessitate the development of optimized and scalable training methods. In this white paper, we addressed the challenge of efficiently…

Machine Learning · Computer Science 2024-04-29 Raphael Ruschel , A. S. M. Iftekhar , B. S. Manjunath , Suya You

What is a systematic way to efficiently apply a wide spectrum of advanced ML programs to industrial scale problems, using Big Models (up to 100s of billions of parameters) on Big Data (up to terabytes or petabytes)? Modern parallelization…

Hybrid parallelism underpins large-scale LLM training across tens of thousands of GPUs. At such scale, hardware failures on individual devices lead to performance skew across devices, diminishing overall training efficiency. Existing…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-05-12 Tenghui Ma , Jihu Guo , Wei Gao , Sitian Lu , Zhisheng Ye , Hanjing Wang , Dahua Lin

Solving inverse problems and achieving statistical rigour in landscape evolution models requires running many model realizations. Parallel computation is necessary to achieve this in a reasonable time. However, no previous algorithm is…

Computational Engineering, Finance, and Science · Computer Science 2019-01-23 Richard Barnes

Data parallelism has emerged as a necessary technique to accelerate the training of deep neural networks (DNN). In a typical data parallelism approach, the local workers push the latest updates of all the parameters to the parameter server…

Distributed, Parallel, and Cluster Computing · Computer Science 2017-09-28 Shizhao Sun , Wei Chen , Jiang Bian , Xiaoguang Liu , Tie-Yan Liu

The increasing demand for computational resources of training neural networks leads to a concerning growth in energy consumption. While parallelization has enabled upscaling model and dataset sizes and accelerated training, its impact on…

State-of-the-art machine learning frameworks support a wide variety of design features to enable a flexible machine learning programming interface and to ease the programmability burden on machine learning developers. Identifying and using…

Machine Learning · Computer Science 2020-07-01 Yu Emma Wang , Carole-Jean Wu , Xiaodong Wang , Kim Hazelwood , David Brooks

Many techniques in program synthesis, superoptimization, and array programming require parallel rollouts of general-purpose programs. GPUs, while capable targets for domain-specific parallelism, are traditionally underutilized by such…

Programming Languages · Computer Science 2026-04-15 Breandan Considine

Transformer models have emerged as potent solutions to a wide array of multidisciplinary challenges. The deployment of Transformer architectures is significantly hindered by their extensive computational and memory requirements,…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-07-03 Zhengxian Lu , Fangyu Wang , Zhiwei Xu , Fei Yang , Tao Li

More and more large data collections are gathered worldwide in various IT systems. Many of them possess the networked nature and need to be processed and analysed as graph structures. Due to their size they require very often usage of…

Distributed, Parallel, and Cluster Computing · Computer Science 2013-06-04 Tomasz Kajdanowicz , Przemyslaw Kazienko , Wojciech Indyk

Pre-training large neural networks at scale imposes heavy memory demands on accelerators and often requires costly communication. We introduce Subnetwork Data Parallelism (SDP), a distributed training framework that partitions a model into…

Machine Learning · Computer Science 2025-10-06 Vaibhav Singh , Zafir Khalid , Edouard Oyallon , Eugene Belilovsky

Dynamic parallelism on GPUs allows GPU threads to dynamically launch other GPU threads. It is useful in applications with nested parallelism, particularly where the amount of nested parallelism is irregular and cannot be predicted…

Distributed, Parallel, and Cluster Computing · Computer Science 2022-01-11 Mhd Ghaith Olabi , Juan Gómez Luna , Onur Mutlu , Wen-mei Hwu , Izzat El Hajj

Recently, a new paradigm, meta learning, has been widely applied to Deep Learning Recommendation Models (DLRM) and significantly improves statistical performance, especially in cold-start scenarios. However, the existing systems are not…

Machine Learning · Computer Science 2024-04-16 Youshao Xiao , Shangchun Zhao , Zhenglei Zhou , Zhaoxin Huan , Lin Ju , Xiaolu Zhang , Lin Wang , Jun Zhou

We present JaxPP, a system for efficiently scaling the training of large deep learning models with flexible pipeline parallelism. We introduce a seamless programming model that allows implementing user-defined pipeline schedules for…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-12-20 Anxhelo Xhebraj , Sean Lee , Hanfeng Chen , Vinod Grover

Among the many possible approaches for the parallelization of self-organizing networks, and in particular of growing self-organizing networks, perhaps the most common one is producing an optimized, parallel implementation of the standard…

Distributed, Parallel, and Cluster Computing · Computer Science 2015-03-31 Giacomo Parigi , Angelo Stramieri , Danilo Pau , Marco Piastra

Pipeline Parallelism (PP) enables large neural network training on small, interconnected devices by splitting the model into multiple stages. To maximize pipeline utilization, asynchronous optimization is appealing as it offers 100%…

Machine Learning · Computer Science 2025-05-05 Thalaiyasingam Ajanthan , Sameera Ramasinghe , Yan Zuo , Gil Avraham , Alexander Long