IEICE TRANSACTIONS on Information
A Multi-Tenant Resource Management System for Multi-FPGA Systems
Summary :
FPGA (Field Programmable Gate Array) based accelerators are attracting significant interest in cloud computing systems. Combining multi-FPGA systems with cloud computing brings a new perspective to the reconfigurable computing research. However, the multi-tenancy of a multi-FPGA system has not been fully discussed in the previous researches. In this paper, we propose a multi-tenant resource management system, named FiC-RM, for a multi-FPGA cloud system. FiC-RM provides users with a set of FPGA resources according to their requirements and allows them to exclusively access FPGA boards and the interconnection network. To achieve this, we propose a placement algorithm which is a key to efficiently share the limited resources. We demonstrate FiC-RM controls a practical scale multi-FPGA system. Moreover, Our simulation study shows that our placement algorithm achieved 3 to 4% improvement in the average resource usage and a 20-second reduction in the response time, compared to other existing naive algorithms.
- Publication
- IEICE TRANSACTIONS on Information Vol.E104-D No.12 pp.2078-2088
- Publication Date
- 2021/12/01
- Publicized
- 2021/10/08
- Online ISSN
- 1745-1361
- DOI
- 10.1587/transinf.2021PAP0005
- Type of Manuscript
- Special Section PAPER (Special Section on Parallel, Distributed, and Reconfigurable Computing, and Networking)
- Category
Authors
Miho YAMAKURA
Keio University
Ryousei TAKANO
National Institute of Advanced Industrial Science and Technology
Akram BEN AHMED
National Institute of Advanced Industrial Science and Technology
Midori SUGAYA
Shibaura Institute of Technology
Hideharu AMANO
Keio University
Keyword
Latest Issue
Copyrights notice of machine-translated contents
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Cite this
Copy
Miho YAMAKURA, Ryousei TAKANO, Akram BEN AHMED, Midori SUGAYA, Hideharu AMANO, "A Multi-Tenant Resource Management System for Multi-FPGA Systems" in IEICE TRANSACTIONS on Information,
vol. E104-D, no. 12, pp. 2078-2088, December 2021, doi: 10.1587/transinf.2021PAP0005.
Abstract: FPGA (Field Programmable Gate Array) based accelerators are attracting significant interest in cloud computing systems. Combining multi-FPGA systems with cloud computing brings a new perspective to the reconfigurable computing research. However, the multi-tenancy of a multi-FPGA system has not been fully discussed in the previous researches. In this paper, we propose a multi-tenant resource management system, named FiC-RM, for a multi-FPGA cloud system. FiC-RM provides users with a set of FPGA resources according to their requirements and allows them to exclusively access FPGA boards and the interconnection network. To achieve this, we propose a placement algorithm which is a key to efficiently share the limited resources. We demonstrate FiC-RM controls a practical scale multi-FPGA system. Moreover, Our simulation study shows that our placement algorithm achieved 3 to 4% improvement in the average resource usage and a 20-second reduction in the response time, compared to other existing naive algorithms.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2021PAP0005/_p
Copy
@ARTICLE{e104-d_12_2078,
author={Miho YAMAKURA, Ryousei TAKANO, Akram BEN AHMED, Midori SUGAYA, Hideharu AMANO, },
journal={IEICE TRANSACTIONS on Information},
title={A Multi-Tenant Resource Management System for Multi-FPGA Systems},
year={2021},
volume={E104-D},
number={12},
pages={2078-2088},
abstract={FPGA (Field Programmable Gate Array) based accelerators are attracting significant interest in cloud computing systems. Combining multi-FPGA systems with cloud computing brings a new perspective to the reconfigurable computing research. However, the multi-tenancy of a multi-FPGA system has not been fully discussed in the previous researches. In this paper, we propose a multi-tenant resource management system, named FiC-RM, for a multi-FPGA cloud system. FiC-RM provides users with a set of FPGA resources according to their requirements and allows them to exclusively access FPGA boards and the interconnection network. To achieve this, we propose a placement algorithm which is a key to efficiently share the limited resources. We demonstrate FiC-RM controls a practical scale multi-FPGA system. Moreover, Our simulation study shows that our placement algorithm achieved 3 to 4% improvement in the average resource usage and a 20-second reduction in the response time, compared to other existing naive algorithms.},
keywords={},
doi={10.1587/transinf.2021PAP0005},
ISSN={1745-1361},
month={December},}
Copy
TY - JOUR
TI - A Multi-Tenant Resource Management System for Multi-FPGA Systems
T2 - IEICE TRANSACTIONS on Information
SP - 2078
EP - 2088
AU - Miho YAMAKURA
AU - Ryousei TAKANO
AU - Akram BEN AHMED
AU - Midori SUGAYA
AU - Hideharu AMANO
PY - 2021
DO - 10.1587/transinf.2021PAP0005
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E104-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2021
AB - FPGA (Field Programmable Gate Array) based accelerators are attracting significant interest in cloud computing systems. Combining multi-FPGA systems with cloud computing brings a new perspective to the reconfigurable computing research. However, the multi-tenancy of a multi-FPGA system has not been fully discussed in the previous researches. In this paper, we propose a multi-tenant resource management system, named FiC-RM, for a multi-FPGA cloud system. FiC-RM provides users with a set of FPGA resources according to their requirements and allows them to exclusively access FPGA boards and the interconnection network. To achieve this, we propose a placement algorithm which is a key to efficiently share the limited resources. We demonstrate FiC-RM controls a practical scale multi-FPGA system. Moreover, Our simulation study shows that our placement algorithm achieved 3 to 4% improvement in the average resource usage and a 20-second reduction in the response time, compared to other existing naive algorithms.
ER -
English
