M-KUBOS/PYNQ Cluster for multi-Access edge computing

Takumi Inage, Kazuei Hironaka, Kensuke Iizuka, Kohei Ito, Yasuyu Fukushima, Mitaro Namiki, Hideharu Amano

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The M-KUBOS/PYNQ cluster is proposed as an energy efficient computing server for MEC(Multi-Access Edge Computing) which treats various tasks including time-critical jobs at 5G base stations. M-KUBOS boards each of which provides the largest model of Zynq UltraScale+ are connected with high speed serial GTH links and STDM (static time division multiplexing) switches. PYNQ(Python productivity for Zynq), an open-source software platform by Xilinx was introduced so that the multi-Zynq system can be used as a multitenant server. The PYNQ cluster can be programmed by extension of a single PYNQ system, and RESTful APIs are introduced for runtime management of multiple Zynq boards from a remote server. The design examples and evaluation results demonstrate that the M-KUBOS/PYNQ cluster achieved 1.0-17.2 times the performance and 1.75-85.8 times the performance per watt of the recent Ryzen-7 server.

Original languageEnglish
Title of host publicationProceedings - 2021 9th International Symposium on Computing and Networking, CANDAR 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages95-101
Number of pages7
ISBN (Electronic)9781665442466
DOIs
Publication statusPublished - 2021
Event9th International Symposium on Computing and Networking, CANDAR 2021 - Virtual, Online, Japan
Duration: 2021 Nov 232021 Nov 26

Publication series

NameProceedings - 2021 9th International Symposium on Computing and Networking, CANDAR 2021

Conference

Conference9th International Symposium on Computing and Networking, CANDAR 2021
Country/TerritoryJapan
CityVirtual, Online
Period21/11/2321/11/26

Keywords

  • FPGA cluster
  • PYNQ
  • Zynq

ASJC Scopus subject areas

  • Computer Networks and Communications

Fingerprint

Dive into the research topics of 'M-KUBOS/PYNQ Cluster for multi-Access edge computing'. Together they form a unique fingerprint.

Cite this