Improving the Performance of Circuit-Switched Interconnection Network for a Multi-FPGA System

被引:4
|
作者
Ito, Kohei [1 ]
Iizuka, Kensuke [1 ]
Hironaka, Kazuei [1 ]
Hu, Yao [2 ]
Koibuchi, Michihiro [2 ]
Amano, Hideharu [1 ]
机构
[1] Keio Univ, Yokohama, Kanagawa 2238522, Japan
[2] Natl Inst Informat, Tokyo 1018430, Japan
来源
关键词
multi-FPGA; multi-FPGA communication; circuit-switched network; STDM switch; TOPOLOGY;
D O I
10.1587/transinf.2021PAP0002
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Multi-FPGA systems have gained attention because of their high performance and power efficiency. A multi-FPGA system called Flow-in-Cloud (FiC) is currently being developed as an accelerator of multi-access edge computing (MEC). FiC consists of multiple mid-range FPGAs tightly connected by high-speed serial links. Since time-critical jobs are assumed in MEC, a circuit-switched network with static timedivision multiplexing (STDM) switches has been implemented on FiC. This paper investigates techniques of enhancing the interconnection performance of FiC. Unlike switching fabrics for Network on Chips or parallel machines, economical multi-FPGA systems, such as FiC, use Xilinx Aurora IP and FireFly cables with multiple lanes. We adopted the link aggregation and the slot distribution for using multiple lanes. To mitigate the bottleneck between an STDM switch and user logic, we also propose a multi-ejection STDM switch. We evaluated various combinations of our techniques by using three practical applications on an FiC prototype with 24 boards. When the number of slots is large and transferred data size is small, the slot distribution was sometimes more effective, while the link aggregation was superior for other most cases. Our multi-ejection STDM switch mitigated the bottleneck in ejection ports and successfully reduced the number of time slots. As a result, by combining the link aggregation and multi-ejection STDM switch, communication performance improved up to 7.50 times with few additional resources. Although the performance of the fast Fourier transform with the highest communication ratio could not be enhanced by using multiple boards when a lane was used, 1.99 times performance improvement was achieved by using 8 boards with four lanes and our multi-ejection switch compared with a board.
引用
收藏
页码:2029 / 2039
页数:11
相关论文
共 50 条
  • [31] A NETWORK FLOW MODEL FOR LOAD BALANCING IN CIRCUIT-SWITCHED MULTICOMPUTERS
    BOKHARI, SH
    IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 1993, 4 (06) : 649 - 657
  • [32] Mapping and Scheduling for Circuit-Switched Network-on-Chip Architecture
    Wu, Chia-Ming
    Chi, Hsin-Chou
    Chang, Ruay-Shiung
    ETRI JOURNAL, 2009, 31 (02) : 111 - 120
  • [33] A circuit partitioning algorithm with replication capability for multi-FPGA systems
    Togawa, N
    Sato, M
    Ohtsuki, T
    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES, 1995, E78A (12) : 1765 - 1776
  • [34] PERFORMANCE ISSUES IN THE DESIGN OF DYNAMICALLY CONTROLLED CIRCUIT-SWITCHED NETWORKS
    CHEMOUIL, P
    FILIPIAK, J
    GAUTHIER, P
    IEEE COMMUNICATIONS MAGAZINE, 1990, 28 (10) : 90 - 95
  • [35] Toward a training of CNNs on a multi-FPGA system
    Hiruma, Aoi
    Iizuka, Kensuke
    Amano, Hideharu
    2022 TENTH INTERNATIONAL SYMPOSIUM ON COMPUTING AND NETWORKING WORKSHOPS, CANDARW, 2022, : 229 - 235
  • [36] COMPARATIVE PERFORMANCE OF CIRCUIT-SWITCHED NETWORKS BASED ON BLOCKING PROBABILITY
    HEIN, C
    PROCEEDINGS : SUPERCOMPUTING 89, 1989, : 704 - 712
  • [37] COMBINED PACKET AND CIRCUIT-SWITCHED PROCESSING SATELLITE SYSTEM.
    Ozarow, Lawrence H.
    DeRosa, Joseph K.
    Conference Record - International Conference on Communications, 1979, 2 (05):
  • [38] Integrated mapping and scheduling for circuit-switched network-on-chip architectures
    Chi, Hsin-Chou
    Wu, Chia-Ming
    Lee, Jun-Hui
    DELTA 2008: FOURTH IEEE INTERNATIONAL SYMPOSIUM ON ELECTRONIC DESIGN, TEST AND APPLICATIONS, PROCEEDINGS, 2008, : 415 - 420
  • [39] NON-BLOCKING CIRCUIT-SWITCHED LOCAL AREA NETWORK.
    Anon
    IBM technical disclosure bulletin, 1986, 28 (12): : 5254 - 5258
  • [40] Multi-FPGA Communication Interface for Electric Circuit Co-Simulation
    Lemaire, Michel
    Massicotte, Daniel
    Belanger, Jean
    2020 IEEE ELECTRIC POWER AND ENERGY CONFERENCE (EPEC), 2020,