Presentation 2015-01-29
Inter-Cube Data-Exchanging for Custom Fluid Computing Machine Based on Building-Cube Method
Tomoya UENO, Tomohiro UENO, Kentaro SANO, Satoru YAMAMOTO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In our research project, we have developed an FPGA-based custom computing machine for parallel fluid computation based on Building-Cube Method(BCM). In BCM, fluid computation is parallelized with cubes with various sizes, which are each composed of an orthogonal computational lattice with the same resolution. Although BCM is expected to provide good load balancing with cubes, the sustained performance is limited for an incompressive fluid computing scheme due to its low operational intensity. Moreover, it is also reported that parallel processing efficiency is decreased by the overhead of data transfer and interpolation for exchanging ghost cells between adjacent cubes. This paper describes an algorithm and its hardware mechanism for data-transfer and interpolation to achieve efficient inter-cube data exchange.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) FPGA / custom computing machines / parallel computational fluid dynamics / building-cube method / data exchange
Paper # VLD2014-124,CPSY2014-133,RECONF2014-57
Date of Issue

Conference Information
Committee RECONF
Conference Date 2015/1/22(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) Inter-Cube Data-Exchanging for Custom Fluid Computing Machine Based on Building-Cube Method
Sub Title (in English)
Keyword(1) FPGA
Keyword(2) custom computing machines
Keyword(3) parallel computational fluid dynamics
Keyword(4) building-cube method
Keyword(5) data exchange
1st Author's Name Tomoya UENO
1st Author's Affiliation Graduate school of information sciences, Tohoku university()
2nd Author's Name Tomohiro UENO
2nd Author's Affiliation Graduate school of information sciences, Tohoku university
3rd Author's Name Kentaro SANO
3rd Author's Affiliation Graduate school of information sciences, Tohoku university
4th Author's Name Satoru YAMAMOTO
4th Author's Affiliation Graduate school of information sciences, Tohoku university
Date 2015-01-29
Paper # VLD2014-124,CPSY2014-133,RECONF2014-57
Volume (vol) vol.114
Number (no) 428
Page pp.pp.-
#Pages 6
Date of Issue