vPipe: A Virtualized Acceleration System for Achieving Efficient and Scalable Pipeline Parallel DNN Training-Reference-Cited by-同舟云学术

vPipe: A Virtualized Acceleration System for Achieving Efficient and Scalable Pipeline Parallel DNN Training

Published:2022-03-01 Issue:3 Volume:33 Page:489-506
ISSN:1045-9219
Container-title:IEEE Transactions on Parallel and Distributed Systems
language:
Short-container-title:IEEE Trans. Parallel Distrib. Syst.

Author:

Zhao Shixiong^ORCID,Li Fanxin,Chen Xusheng^ORCID,Guan Xiuxian,Jiang Jianyu^ORCID,Huang Dong,Qing Yuhao,Wang Sen,Wang Peng,Zhang Gong,Li Cheng^ORCID,Luo Ping,Cui Heming^ORCID

Funder

Huawei Innovation Research Program

Croucher Innovation Award

National Natural Science Foundation of China

USTC Research Funds of Double First-Class Initiative

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Subject

Computational Theory and Mathematics,Hardware and Architecture,Signal Processing

Link

http://xplorestaging.ieee.org/ielx7/71/9497774/09472938.pdf?arnumber=9472938

Cited by 20 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. A snapshot of parallelism in distributed deep learning training;Revista Colombiana de Computación;2024-06-30

2. Advances of Pipeline Model Parallelism for Deep Learning Training: An Overview;Journal of Computer Science and Technology;2024-05

3. AdaPipe: Optimizing Pipeline Parallelism with Adaptive Recomputation and Partitioning;Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3;2024-04-27

4. LaANIL: ANIL with Look-Ahead Meta-Optimization and Data Parallelism;Electronics;2024-04-22

5. HCPipe: Hidden Computation Pipeline Model Parallelism for Large-scale DCNN Model Training;2024 5th International Conference on Computer Engineering and Application (ICCEA);2024-04-12