Presentation 2010-05-13
A datapath classification method for efficient arithmetic pipeline combining on FPGAs
Yui OGAWA, Tomonori OOYA, Yasunori OSANA, Masato YOSHIMI, Yuri NISHIKAWA, Akira FUNAHASHI, Noriko HIROI, Hideharu AMANO, Yuichiro SHIBATA, Kiyoshi OGURI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Resource reduction design techniques also plays an important role to implement large-scale FPGA-based accelerator systems in floating point applications since available resources on FPGAs are limited. This paper proposes a datapath classification method which makes groups of datapaths based on their similarity in order to bring out efficient datapath combining. Aiming at finding effective parameters for our k-means based algorithm, various combinations of parameters are evaluated and compared in terms of performance and resource reduction effects. The experimental results using an FPGA-based biochemical simulator reveal that the datapath clustering that uses information on the maximum common subgraphs achieved 73.3% of resource reduction rate while alleviating the performance degradation.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) pipeline / datapath / FPGA / clustering / common subgraph
Paper # RECONF2010-8
Date of Issue

Conference Information
Committee RECONF
Conference Date 2010/5/6(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Reconfigurable Systems (RECONF)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A datapath classification method for efficient arithmetic pipeline combining on FPGAs
Sub Title (in English)
Keyword(1) pipeline
Keyword(2) datapath
Keyword(3) FPGA
Keyword(4) clustering
Keyword(5) common subgraph
1st Author's Name Yui OGAWA
1st Author's Affiliation Nagasaki University()
2nd Author's Name Tomonori OOYA
2nd Author's Affiliation Nagasaki University
3rd Author's Name Yasunori OSANA
3rd Author's Affiliation Seikei University
4th Author's Name Masato YOSHIMI
4th Author's Affiliation Doshisha University
5th Author's Name Yuri NISHIKAWA
5th Author's Affiliation Keio University
6th Author's Name Akira FUNAHASHI
6th Author's Affiliation Keio University
7th Author's Name Noriko HIROI
7th Author's Affiliation Keio University
8th Author's Name Hideharu AMANO
8th Author's Affiliation Keio University
9th Author's Name Yuichiro SHIBATA
9th Author's Affiliation Nagasaki University
10th Author's Name Kiyoshi OGURI
10th Author's Affiliation Nagasaki University
Date 2010-05-13
Paper # RECONF2010-8
Volume (vol) vol.110
Number (no) 32
Page pp.pp.-
#Pages 6
Date of Issue