A data backup optimization mechanism for the virtual FPGA cluster

Wei Liu; Yu Fang; Jingdong Zhang; Yanmei Shen; Kun Zhao; Dongdong Su

doi:10.1117/12.3037443

27 September 2024 A data backup optimization mechanism for the virtual FPGA cluster

Wei Liu, Yu Fang, Jingdong Zhang, Yanmei Shen, Kun Zhao, Dongdong Su

Author Affiliations +

Proceedings Volume 13275, Sixth International Conference on Information Science, Electrical, and Automation Engineering (ISEAE 2024); 132751K (2024) https://doi.org/10.1117/12.3037443
Event: 6th International Conference on Information Science, Electrical, and Automation Engineering (ISEAE 2024), 2024, Wuhan, China

Abstract

As a heterogeneous computing architecture, FPGA accelerators are widely used for AI and other fields. Compared with traditional CPU clusters, the architecture of the virtual FPGA cluster is relatively simple, and thus lacks the design of relevant mechanisms for data backup. In our work, a data backup optimization mechanism for the virtual FPGA cluster is proposed to select local sub-kernels and remote sub-kernels of other FPGA accelerators as data backups. During the selection of remote sub-kernel backups, the data transmission delay between the sub-kernel under evaluation and the subkernel to be backed up, the statistical probability of the historical errors of the sub-kernel under evaluation, and the internal environment influence factors in its FPGA accelerator are fully considered. The simulation results demonstrate that the proposed data backup optimization mechanism has shown good data disaster recovery performance.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Wei Liu, Yu Fang, Jingdong Zhang, Yanmei Shen, Kun Zhao, and Dongdong Su "A data backup optimization mechanism for the virtual FPGA cluster", Proc. SPIE 13275, Sixth International Conference on Information Science, Electrical, and Automation Engineering (ISEAE 2024), 132751K (27 September 2024); https://doi.org/10.1117/12.3037443

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

;

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE