paper-with-me

Papers

Exploring GPU-to-GPU Communication: Insights into Supercomputer Interconnects

2024-08-26 · Daniele De Sensi, Lorenzo Pichetti, Flavio Vella, Tiziano De Matteis, Zebin Ren, Luigi Fusco, Matteo Turisini, Daniele Cesarini, Kurt Lust, Animesh Trivedi, Duncan Roweth, Filippo Spiga, Salvatore Di Girolamo, Torsten Hoefler

Multi-GPU nodes are increasingly common in the rapidly evolving landscape of exascale supercomputers. On these systems, GPUs on the same node are connected through dedicated networks, with bandwidths up to a few terabits per second. However, gauging performance expectations and maximizing system efficiency is challenging due to different technologies, design options, and software layers. This paper comprehensively characterizes three supercomputers - Alps, Leonardo, and LUMI - each with a unique architecture and design. We focus on performance evaluation of intra-node and inter-node interconnects on up to 4096 GPUs, using a mix of intra-node and inter-node benchmarks. By analyzing its limitations and opportunities, we aim to offer practical guidance to researchers, system architects, and software developers dealing with multi-GPU supercomputing. Our results show that there is untapped bandwidth, and there are still many opportunities for optimization, ranging from network to software optimization.

📄 PDF Abstract BibTeX arXiv:2408.14090

Code (0)

등록된 구현이 없습니다.

Tasks

GPU

Methods 이 논문이 사용한 방법론

Focus 설명 없음

Similar Papers 제목 키워드 기반

Understanding and Improving Communication Performance in Multi-node LLM Inference

2025-11-12 · Prajwal Singhania, Siddharth Singh, Lannie Dalton Hough, Akarsh Srivastava 외 arxiv

As large language models (LLMs) continue to grow in size, distributed inference has become increasingly important. Model-parallel strategies must now efficiently scale not only across multiple GPUs but also across multip…

FLUX: Fast Software-based Communication Overlap On GPUs Through Kernel Fusion

2024-06-11 · Li-Wen Chang, Wenlei Bao, Qi Hou, Chengquan Jiang 외

Large deep learning models have demonstrated strong ability to solve many tasks across a wide range of applications. Those large models typically require training and inference to be distributed. Tensor parallelism is a …

GPU

Distributed Low-Communication Training with Decoupled Momentum Optimization

2025-10-03 · Sasho Nedelkoski, Alexander Acker, Odej Kao, Soeren Becker 외 arxiv

The training of large models demands substantial computational resources, typically available only in data centers with high-bandwidth interconnects. However, reducing the reliance on high-bandwidth interconnects between…

Wireless Interconnection Network (WINE) for Post-Exascale High-Performance Computing

2024-09-20 · Hong Ki Kim, Yong Hun Jang, Hee Soo Kim, Won Young Kang 외

Interconnection networks, or `interconnects,' play a crucial role in administering the communication among computing units of high-performance computing (HPC) systems. Efficient provisioning of interconnects minimizes th…

Exploring the Potential of Wireless-enabled Multi-Chip AI Accelerators

2025-01-29 · Emmanuel Irabor, Mariam Musavi, Abhijit Das, Sergi Abadal

The insatiable appetite of Artificial Intelligence (AI) workloads for computing power is pushing the industry to develop faster and more efficient accelerators. The rigidity of custom hardware, however, conflicts with th…