TY - GEN
T1 - Power Analysis of Directly-connected FPGA Clusters
AU - Iizuka, Kensuke
AU - Takagi, Haruna
AU - Kamei, Aika
AU - Hironaka, Kazuei
AU - Amano, Hideharu
N1 - Funding Information:
ACKNOWLEDGMENT This work was supported by JST CREST Grant Number JPMJCR19K1, Japan and JST SPRING, Grant Number JP-
Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - Although low power consumption is a significant advantage of FPGA clusters, almost no power analyses with real systems have been reported. This study reports the detailed power consumption analyses of two FPGA clusters, namely, M-KUBOS and FiC, with power measurement tools and real applications. In both clusters, the type of logic design shells determines the base power consumption. For building clusters, the power for node communication links is mainly determined by the number of activated links and not influenced by the number of actually used links. Therefore, applying the link aggregation technique does not affect the power consumption. Increasing the clock frequency of the application logic mildly increases the power consumption. The obtained results suggest that the best way to reduce the total power consumption of an FPGA cluster and improve its performance is to use the minimum number of links for the application, apply link aggregation, and aggressively increase the clock frequency.
AB - Although low power consumption is a significant advantage of FPGA clusters, almost no power analyses with real systems have been reported. This study reports the detailed power consumption analyses of two FPGA clusters, namely, M-KUBOS and FiC, with power measurement tools and real applications. In both clusters, the type of logic design shells determines the base power consumption. For building clusters, the power for node communication links is mainly determined by the number of activated links and not influenced by the number of actually used links. Therefore, applying the link aggregation technique does not affect the power consumption. Increasing the clock frequency of the application logic mildly increases the power consumption. The obtained results suggest that the best way to reduce the total power consumption of an FPGA cluster and improve its performance is to use the minimum number of links for the application, apply link aggregation, and aggressively increase the clock frequency.
UR - http://www.scopus.com/inward/record.url?scp=85130850256&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85130850256&partnerID=8YFLogxK
U2 - 10.1109/COOLCHIPS54332.2022.9772675
DO - 10.1109/COOLCHIPS54332.2022.9772675
M3 - Conference contribution
AN - SCOPUS:85130850256
T3 - 25th IEEE Symposium on Low-Power and High-Speed Chips and Systems, COOL Chips 2022 - Proceedings
BT - 25th IEEE Symposium on Low-Power and High-Speed Chips and Systems, COOL Chips 2022 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 25th IEEE Symposium on Low-Power and High-Speed Chips and Systems, COOL Chips 2022
Y2 - 20 April 2022 through 22 April 2022
ER -