ACCL: Architecting Highly Scalable Distributed Training Systems With Highly Efficient Collective Communication Library-Reference-Cited by-同舟云学术

ACCL: Architecting Highly Scalable Distributed Training Systems With Highly Efficient Collective Communication Library

Published:2021-09-01 Issue:5 Volume:41 Page:85-92
ISSN:0272-1732
Container-title:IEEE Micro
language:
Short-container-title:IEEE Micro

Author:

Dong Jianbo^ORCID,Wang Shaochuang,Feng Fei,Cao Zheng^ORCID,Pan Heng,Tang Lingbo^ORCID,Li Pengcheng,Li Hao,Ran Qianyuan,Guo Yiqun,Gao Shanyuan,Long Xin,Zhang Jie,Li Yong,Xia Zhisheng,Song Liuyihan^ORCID,Zhang Yingya,Pan Pan,Wang Guohui,Jiang Xiaowei

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Subject

Electrical and Electronic Engineering,Hardware and Architecture,Software

Link

http://xplorestaging.ieee.org/ielx7/40/9536946/09462480.pdf?arnumber=9462480

Cited by 7 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. TCCL: Co-optimizing Collective Communication and Traffic Routing for GPU-centric Clusters;Proceedings of the 2024 SIGCOMM Workshop on Networks for AI Computing;2024-08-04

2. Network Load Balancing with Parallel Flowlets for AI Training Clusters;Proceedings of the 2024 SIGCOMM Workshop on Networks for AI Computing;2024-08-04

3. Crux: GPU-Efficient Communication Scheduling for Deep Learning Training;Proceedings of the ACM SIGCOMM 2024 Conference;2024-08-04

4. AdapCC: Making Collective Communication in Distributed Machine Learning Adaptive;2024 IEEE 44th International Conference on Distributed Computing Systems (ICDCS);2024-07-23

5. FlowStar: Fast Convergence Per-Flow State Accurate Congestion Control for InfiniBand;IEEE/ACM Transactions on Networking;2024-06