Task level pipelining with PEACH2

An FPGA switching fabric for high performance computing

Takaaki Miyajima, Takuya Kuhara, Toshihiro Hanawa, Hideharu Amano, Taisuke Boku

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

1 Citation (Scopus)

Abstract

We demonstrate task level pipelining on multiple accelerators with PEACH2. PEACH2 is implmented on FPGA, and enables ultra low latency direct communication among multiple accelerators over computational nodes. By installing PEACH2, typical high performance computation nodes are tightly coupled. In this environment, application can be accelerated by exploiting not only data level parallelism, but also task level pipelined operation. Furthermore, we can processe multiple task on multiple accelerators in a pipelined manner. In our demonstration, application achieves 44% speed up compared to a single GPU.

Original languageEnglish
Title of host publicationFPT 2013 - Proceedings of the 2013 International Conference on Field Programmable Technology
Pages466-469
Number of pages4
DOIs
Publication statusPublished - 2013
Event2013 12th International Conference on Field-Programmable Technology, FPT 2013 - Kyoto, Japan
Duration: 2013 Dec 92013 Dec 11

Other

Other2013 12th International Conference on Field-Programmable Technology, FPT 2013
CountryJapan
CityKyoto
Period13/12/913/12/11

Fingerprint

Particle accelerators
Field programmable gate arrays (FPGA)
Demonstrations
Communication
Graphics processing unit

ASJC Scopus subject areas

  • Software

Cite this

Miyajima, T., Kuhara, T., Hanawa, T., Amano, H., & Boku, T. (2013). Task level pipelining with PEACH2: An FPGA switching fabric for high performance computing. In FPT 2013 - Proceedings of the 2013 International Conference on Field Programmable Technology (pp. 466-469). [6718416] https://doi.org/10.1109/FPT.2013.6718416

Task level pipelining with PEACH2 : An FPGA switching fabric for high performance computing. / Miyajima, Takaaki; Kuhara, Takuya; Hanawa, Toshihiro; Amano, Hideharu; Boku, Taisuke.

FPT 2013 - Proceedings of the 2013 International Conference on Field Programmable Technology. 2013. p. 466-469 6718416.

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

Miyajima, T, Kuhara, T, Hanawa, T, Amano, H & Boku, T 2013, Task level pipelining with PEACH2: An FPGA switching fabric for high performance computing. in FPT 2013 - Proceedings of the 2013 International Conference on Field Programmable Technology., 6718416, pp. 466-469, 2013 12th International Conference on Field-Programmable Technology, FPT 2013, Kyoto, Japan, 13/12/9. https://doi.org/10.1109/FPT.2013.6718416
Miyajima T, Kuhara T, Hanawa T, Amano H, Boku T. Task level pipelining with PEACH2: An FPGA switching fabric for high performance computing. In FPT 2013 - Proceedings of the 2013 International Conference on Field Programmable Technology. 2013. p. 466-469. 6718416 https://doi.org/10.1109/FPT.2013.6718416
Miyajima, Takaaki ; Kuhara, Takuya ; Hanawa, Toshihiro ; Amano, Hideharu ; Boku, Taisuke. / Task level pipelining with PEACH2 : An FPGA switching fabric for high performance computing. FPT 2013 - Proceedings of the 2013 International Conference on Field Programmable Technology. 2013. pp. 466-469
@inproceedings{5fb38450268e4a2895e9ef95090279bf,
title = "Task level pipelining with PEACH2: An FPGA switching fabric for high performance computing",
abstract = "We demonstrate task level pipelining on multiple accelerators with PEACH2. PEACH2 is implmented on FPGA, and enables ultra low latency direct communication among multiple accelerators over computational nodes. By installing PEACH2, typical high performance computation nodes are tightly coupled. In this environment, application can be accelerated by exploiting not only data level parallelism, but also task level pipelined operation. Furthermore, we can processe multiple task on multiple accelerators in a pipelined manner. In our demonstration, application achieves 44{\%} speed up compared to a single GPU.",
author = "Takaaki Miyajima and Takuya Kuhara and Toshihiro Hanawa and Hideharu Amano and Taisuke Boku",
year = "2013",
doi = "10.1109/FPT.2013.6718416",
language = "English",
isbn = "9781479921990",
pages = "466--469",
booktitle = "FPT 2013 - Proceedings of the 2013 International Conference on Field Programmable Technology",

}

TY - GEN

T1 - Task level pipelining with PEACH2

T2 - An FPGA switching fabric for high performance computing

AU - Miyajima, Takaaki

AU - Kuhara, Takuya

AU - Hanawa, Toshihiro

AU - Amano, Hideharu

AU - Boku, Taisuke

PY - 2013

Y1 - 2013

N2 - We demonstrate task level pipelining on multiple accelerators with PEACH2. PEACH2 is implmented on FPGA, and enables ultra low latency direct communication among multiple accelerators over computational nodes. By installing PEACH2, typical high performance computation nodes are tightly coupled. In this environment, application can be accelerated by exploiting not only data level parallelism, but also task level pipelined operation. Furthermore, we can processe multiple task on multiple accelerators in a pipelined manner. In our demonstration, application achieves 44% speed up compared to a single GPU.

AB - We demonstrate task level pipelining on multiple accelerators with PEACH2. PEACH2 is implmented on FPGA, and enables ultra low latency direct communication among multiple accelerators over computational nodes. By installing PEACH2, typical high performance computation nodes are tightly coupled. In this environment, application can be accelerated by exploiting not only data level parallelism, but also task level pipelined operation. Furthermore, we can processe multiple task on multiple accelerators in a pipelined manner. In our demonstration, application achieves 44% speed up compared to a single GPU.

UR - http://www.scopus.com/inward/record.url?scp=84894223759&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84894223759&partnerID=8YFLogxK

U2 - 10.1109/FPT.2013.6718416

DO - 10.1109/FPT.2013.6718416

M3 - Conference contribution

SN - 9781479921990

SP - 466

EP - 469

BT - FPT 2013 - Proceedings of the 2013 International Conference on Field Programmable Technology

ER -