English
Related papers

Related papers: Characterizing Compute-Communication Overlap in GP…

200 papers

How can we optimally trade extra computing power to reduce the communication load in distributed computing? We answer this question by characterizing a fundamental tradeoff between computation and communication in distributed computing,…

Information Theory · Computer Science 2017-09-26 Songze Li , Mohammad Ali Maddah-Ali , Qian Yu , A. Salman Avestimehr

Most parallel neural network training methods assume homogeneous computing resources. For example, synchronous data-parallel SGD suffers from significant synchronization overhead under heterogeneous workloads, often forcing practitioners to…

Machine Learning · Computer Science 2026-02-24 Jihyun Lim , Junhyuk Jo , Chanhyeok Ko , Young Min Go , Jimin Hwa , Sunwoo Lee

The widely-adopted practice is to train deep learning models with specialized hardware accelerators, e.g., GPUs or TPUs, due to their superior performance on linear algebra operations. However, this strategy does not employ effectively the…

Distributed, Parallel, and Cluster Computing · Computer Science 2020-04-21 Yujing Ma , Florin Rusu

In recent processor development, we have witnessed the integration of GPU and CPUs into a single chip. The result of this integration is a reduction of the data communication overheads. This enables an efficient collaboration of both…

Distributed, Parallel, and Cluster Computing · Computer Science 2016-08-07 Francisco Corbera , Andrés Rodríguez , Rafael Asenjo , Angeles Navarro , Antonio Vilches , María J. Garzarán

With the rapid growth of large language models (LLMs), a wide range of methods have been developed to distribute computation and memory across hardware devices for efficient training and inference. While existing surveys provide descriptive…

Machine Learning · Computer Science 2026-02-11 Hossam Amer , Rezaul Karim , Ali Pourranjbar , Weiwei Zhang , Walid Ahmed , Boxing Chen

Nowadays the number of available processing cores within computing nodes which are used in recent clustered environments, are growing up with a rapid rate. Despite this trend, the number of available network interfaces in such computing…

Distributed, Parallel, and Cluster Computing · Computer Science 2012-07-13 Mohsen Soryani , Morteza Analoui , Ghobad Zarrinchian

Deep Learning Training (DLT) is a growing workload in shared GPU/CPU clusters due to its high computational cost and increasing number of jobs. This contributes to significant energy consumption in GPU clusters, further exacerbated by GPU…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-12-12 Kawsar Haghshenas , Mona Hashemi

Hardware compute power has been growing at an unprecedented rate in recent years. The utilization of such advancements plays a key role in producing better results in less time -- both in academia and industry. However, merging the existing…

Machine Learning · Computer Science 2021-10-19 Vineeth S

Energy system optimization models are increasing in scope and resolution, yielding large and challenging linear programs. For a long time, the standard way to address such problems has relied on shared-memory interior-point methods (IPM),…

Optimization and Control · Mathematics 2026-05-07 Janina Zittel , Annika Buchholz , Michael Bussieck , Frederik Fiand , Thorsten Koch , Lukas Mehl , Niels Lindner , Manuel Wetzel

To reduce computational complexity and delay in randomized network coded content distribution, and for some other practical reasons, coding is not performed simultaneously over all content blocks, but over much smaller, possibly overlapping…

Information Theory · Computer Science 2016-11-17 Yao Li , Emina Soljanin , Predrag Spasojevic

The rapid advancement of embedded multicore and many-core systems has revolutionized computing, enabling the development of high-performance, energy-efficient solutions for a wide range of applications. As models scale up in size, data…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-10-15 Ruhai Lin , Rui-Jie Zhu , Jason K. Eshraghian

Given the popularity of generative AI, Large Language Models (LLMs) often consume hundreds or thousands of GPUs for parallelizing and accelerating the training process. Communication overhead becomes more pronounced when training LLMs at…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-09-24 Guanhua Wang , Chengming Zhang , Zheyu Shen , Ang Li , Olatunji Ruwase

Graph neural network (GNN) has been demonstrated to be a powerful model in many domains for its effectiveness in learning over graphs. To scale GNN training for large graphs, a widely adopted approach is distributed training which…

Distributed, Parallel, and Cluster Computing · Computer Science 2022-04-19 Haiyang Lin , Mingyu Yan , Xiaocheng Yang , Mo Zou , Wenming Li , Xiaochun Ye , Dongrui Fan

Mixture-of-experts (MoE) architectures enable trillion-parameter LLMs with sparsely activated experts. Expert parallelism (EP) is a widely adopted MoE training strategy, but it suffers from severe all-to-all communication bottlenecks, which…

With rapidly increasing distributed deep learning workloads in large-scale data centers, efficient distributed deep learning framework strategies for resource allocation and workload scheduling have become the key to high-performance deep…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-06-13 Feng Liang , Zhen Zhang , Haifeng Lu , Chengming Li , Victor C. M. Leung , Yanyi Guo , Xiping Hu

Distributed artificial intelligence (AI) has recently accomplished tremendous breakthroughs in various communication services, ranging from fault-tolerant factory automation to smart cities. When distributed learning is run over a set of…

Networking and Internet Architecture · Computer Science 2022-08-03 Milad Ganjalizadeh , Hossein S. Ghadikolaei , Johan Haraldson , Marina Petrova

We present a new algorithmic paradigm for the decentralized solution of graph-structured optimization problems that arise in the estimation and control of network systems. A key and novel design concept of the proposed approach is that it…

Optimization and Control · Mathematics 2020-04-01 Sungho Shin , Victor M. Zavala , Mihai Anitescu

One of the major bottlenecks for efficient deployment of neural network based recommendation systems is the memory footprint of their embedding tables. Although many neural network based recommendation systems could benefit from the faster…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-11-05 Lutfi Eren Erdogan , Vijay Anand Raghava Kanakagiri , Kurt Keutzer , Zhen Dong

The last decade has seen a shift in the computer systems industry where heterogeneous computing has become prevalent. Graphics Processing Units (GPUs) are now present in supercomputers to mobile phones and tablets. GPUs are used for…

Distributed, Parallel, and Cluster Computing · Computer Science 2019-09-04 Yehia Arafa , Abdel-Hameed Badawy , Gopinath Chennupati , Nandakishore Santhi , Stephan Eidenbenz

Distributed multi-task learning (DMTL) effectively improves model generalization performance through the collaborative training of multiple related models. However, in large-scale learning scenarios, communication bottlenecks severely limit…

Information Theory · Computer Science 2025-07-25 Minquan Cheng , Yongkang Wang , Lingyu Zhang , Youlong Wu