logo
科言猫学术研究的AI总结
language
手机端手机端
  • Phone Feature微信小程序
简体中文简体中文
notice消息通知
博客博客
delete
arrow
返回

Scaling GPU-to-CPU Migration for Efficient Distributed Execution on CPU Clusters

delete2026-01-01
delete0
PRE
AI
R
Ruobing Han *
H
Hyesoon Kim
DOI：10.1145/3774934.3786435delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
GPU资源需求的日益增长导致数据中心出现广泛短缺，促使探索使用CPU作为执行GPU程序的替代方案。尽管先前研究支持在单个CPU上执行GPU程序，但这些方法因CPU与GPU之间的计算能力差距而难以实现有竞争力的性能。为进一步提升性能，我们提出CuCC框架，将GPU到CPU的迁移扩展至CPU集群，并利用分布式CPU节点执行GPU程序。与单CPU执行相比，CPU集群执行需要跨节点通信以维护数据一致性。我们介绍了CuCC的执行工作流和通信优化，旨在减少网络开销。评估结果表明，CuCC在大规模CPU集群上实现了高可扩展性，运行时间接近GPU。在集群级吞吐量方面，CuCC使CPU的平均吞吐量比GPU高出2.59倍。
Keyword:
compiler optimization
GPU-to-CPU migration
CPU cluster

期刊

P
PROCEEDINGS OF THE 31ST ACM SIGPLAN ANNUAL SYMPOSIUM ON PRINCIPLES AND PRACTICE OF PARALLEL PROGRAMMING, PPOPP 2026
IF:
0
论文数:
43
被引数:
0

机构

U
university system of georgia
学者数:
7.3W
论文数: 6.6W
被引数: 101
引用论文

引用论文

Introducing OpenSHMEM
errProceedings of the Fourth Conference on Partitioned Global Address Space Programming Model
IF0
err2010-10-12
err0
PREAI
errBarbara Chapman; Tony Curtis; Swaroop Pophale; Stephen Poole; Jeff Kuehn; Chuck Koelbel; Lauren Smith
err分享
err收藏
Falcon: A Timestamp-based Protocol to Maximize the Cache Efficiency in the Distributed Shared Memory
err2022 IEEE International Parallel and Distributed Processing Symposium (IPDPS)
IF0
err2022-05-01
err0
PREAI
errZhang,Jin; Yu,Xiangyao; Qi,Zhengwei; Guan,Haibing
err分享
err收藏
Chunking parallel loops in the presence of synchronization
errProceedings of the 23rd international conference on Supercomputing
IF0
err2009-06-08
err0
PREAI
errShirako,Jun; Zhao,Jisheng M.; Nandivada,V. Krishna; Sarkar,Vivek N.
err分享
err收藏
Memory access patterns
errProceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis
IF0
err2015-11-15
err0
PREAI
errTal Ben-Nun; Ely Levy; Amnon Barak; Eri Rubin
err分享
err收藏
MCUDA: An Efficient Implementation of CUDA Kernels for Multi-core CPUs
errLecture Notes in Computer Science
IF0
err2008-01-01
err0
PREAI
errJohn A. Stratton; Sam S. Stone; Wen-mei W. Hwu
err分享
err收藏
Physics-informed dynamic mode decomposition
errProceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences
IF0
err2023-03-01
err0
errOAAI
errPeter J. Baddoo; Benjamin Herrmann; Beverley J. McKeon; J. Nathan Kutz; Steven L. Brunton
err分享
err收藏
err
IF0
err
err0
PREAI
err
err分享
err收藏
Efficient compilation of fine-grained SPMD-threaded programs for multicore CPUs
errProceedings of the 8th annual IEEE/ACM international symposium on Code generation and optimization
IF0
err2010-04-24
err0
PREAI
errStratton,John A.; Grover,Vinod; Marathe,Jaydeep; Aarts,Bastiaan; Murphy,Mike; Hu,Ziang; Hwu,Wen-mei W.
err分享
err收藏
Automatic Parallelization of Kernels in Shared-Memory Multi-GPU Nodes
errProceedings of the 29th ACM on International Conference on Supercomputing
IF0
err2015-06-08
err0
PREAI
errJavier Cabezas; Lluís Vilanova; Isaac Gelado; Thomas B. Jablin; Nacho Navarro; Wen-mei W. Hwu
err分享
err收藏
CuPBoP: Making CUDA a Portable Language
errACM Transactions on Design Automation of Electronic Systems
IF0
err2024-07-31
err0
PREAI
errHan,Ruobing; Chen,Jun; Garg,Bhanu; Zhou,Xule; Lu,John; Young,Jeffrey; Sim,Jaewoong; Kim,Hyesoon
err分享
err收藏
学者 查看更多内容