Deep learning on high performance FPGA switching boards: Flow-in-cloud

Kazusa Musha, Tomohiro Kudoh, Hideharu Amano

研究成果: Conference contribution

6 被引用数 (Scopus)

抄録

FiC (Flow-in-Cloud)-SW is an FPGA-based switching node for an efficient AI computing system. It is equipped with a number of serial links directly connected to other nodes. Unlike other multi-FPGA systems, the circuit switching fabric with the STDM (Static Time Division Multiplexing) is implemented on the FPGA for predictable communication and cost-efficient data broadcasting. Parallel convolution modules for AlexNet are implemented on FiC-SW1 prototype boards consisting of Kintex Ultrascale FPGA, and evaluation results show that the parallel execution with 20 boards achieved 4.6 times better performance than the state of art implementation on a single Virtex 7 FPGA board.

本文言語English
ホスト出版物のタイトルApplied Reconfigurable Computing
ホスト出版物のサブタイトルArchitectures, Tools, and Applications - 14th International Symposium, ARC 2018, Proceedings
編集者Nikolaos Voros, Georgios Keramidas, Christos Antonopoulos, Michael Huebner, Pedro C. Diniz, Diana Goehringer
出版社Springer Verlag
ページ43-54
ページ数12
ISBN(印刷版)9783319788890
DOI
出版ステータスPublished - 2018
イベント14th International Symposium on Applied Reconfigurable Computing, ARC 2018 - Santorini, Greece
継続期間: 2018 5月 22018 5月 4

出版物シリーズ

名前Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
10824 LNCS
ISSN(印刷版)0302-9743
ISSN(電子版)1611-3349

Other

Other14th International Symposium on Applied Reconfigurable Computing, ARC 2018
国/地域Greece
CitySantorini
Period18/5/218/5/4

ASJC Scopus subject areas

  • 理論的コンピュータサイエンス
  • コンピュータ サイエンス(全般)

フィンガープリント

「Deep learning on high performance FPGA switching boards: Flow-in-cloud」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル