FPGA-based streaming computation for lattice Boltzmann method

Kentaro Sano; Oliver Pell; Wayne Luk; Satoru Yamamoto

doi:10.1109/FPT.2007.4439254

FPGA-based streaming computation for lattice Boltzmann method

Kentaro Sano, Oliver Pell, Wayne Luk, Satoru Yamamoto

INF - Computer and Mathematical Sciences

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

30 Citations (Scopus)

Abstract

This paper presents an FPGA-based streaming computation for the lattice Boltzmann method (LBM) to simulate fluid flow with floating-point calculations. LBM is suitable for streaming computation because of its parallelism and regularity. We optimize the equations of LBM, and then formulate a streaming computation. To design an efficient data-path for throughput and hardware resource utilization, we introduce multiple cycle inputs and computing-unit sharing to the streaming data-path. The streaming accelerator implemented on a Virtex-4 FPGA with PCI-Express x8 interface achieves 2.93 and 2.46 times faster computation than a 3.4GHz Pentium4 processor and a 2.2GHz Opteron processor, respectively, for 2-dimensional time-dependent fluid dynamics problems.

Original language	English
Title of host publication	ICFPT 2007 - International Conference on Field Programmable Technology
Pages	233-236
Number of pages	4
DOIs	https://doi.org/10.1109/FPT.2007.4439254
Publication status	Published - 2007
Event	International Conference on Field Programmable Technology, ICFPT 2007 - Kitakyushu, Japan Duration: 2007 Dec 12 → 2007 Dec 14

Publication series

Name	ICFPT 2007 - International Conference on Field Programmable Technology

Conference

Conference	International Conference on Field Programmable Technology, ICFPT 2007
Country/Territory	Japan
City	Kitakyushu
Period	07/12/12 → 07/12/14

Access to Document

10.1109/FPT.2007.4439254

Cite this

@inproceedings{d2f9a05931524b87887f940b20374e7a,

title = "FPGA-based streaming computation for lattice Boltzmann method",

abstract = "This paper presents an FPGA-based streaming computation for the lattice Boltzmann method (LBM) to simulate fluid flow with floating-point calculations. LBM is suitable for streaming computation because of its parallelism and regularity. We optimize the equations of LBM, and then formulate a streaming computation. To design an efficient data-path for throughput and hardware resource utilization, we introduce multiple cycle inputs and computing-unit sharing to the streaming data-path. The streaming accelerator implemented on a Virtex-4 FPGA with PCI-Express x8 interface achieves 2.93 and 2.46 times faster computation than a 3.4GHz Pentium4 processor and a 2.2GHz Opteron processor, respectively, for 2-dimensional time-dependent fluid dynamics problems.",

author = "Kentaro Sano and Oliver Pell and Wayne Luk and Satoru Yamamoto",

year = "2007",

doi = "10.1109/FPT.2007.4439254",

language = "English",

isbn = "1424414725",

series = "ICFPT 2007 - International Conference on Field Programmable Technology",

pages = "233--236",

booktitle = "ICFPT 2007 - International Conference on Field Programmable Technology",

note = "International Conference on Field Programmable Technology, ICFPT 2007 ; Conference date: 12-12-2007 Through 14-12-2007",

}

Sano, K, Pell, O, Luk, W & Yamamoto, S 2007, FPGA-based streaming computation for lattice Boltzmann method. in ICFPT 2007 - International Conference on Field Programmable Technology., 4439254, ICFPT 2007 - International Conference on Field Programmable Technology, pp. 233-236, International Conference on Field Programmable Technology, ICFPT 2007, Kitakyushu, Japan, 07/12/12. https://doi.org/10.1109/FPT.2007.4439254

FPGA-based streaming computation for lattice Boltzmann method. / Sano, Kentaro; Pell, Oliver; Luk, Wayne et al.
ICFPT 2007 - International Conference on Field Programmable Technology. 2007. p. 233-236 4439254 (ICFPT 2007 - International Conference on Field Programmable Technology).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - FPGA-based streaming computation for lattice Boltzmann method

AU - Sano, Kentaro

AU - Pell, Oliver

AU - Luk, Wayne

AU - Yamamoto, Satoru

PY - 2007

Y1 - 2007

N2 - This paper presents an FPGA-based streaming computation for the lattice Boltzmann method (LBM) to simulate fluid flow with floating-point calculations. LBM is suitable for streaming computation because of its parallelism and regularity. We optimize the equations of LBM, and then formulate a streaming computation. To design an efficient data-path for throughput and hardware resource utilization, we introduce multiple cycle inputs and computing-unit sharing to the streaming data-path. The streaming accelerator implemented on a Virtex-4 FPGA with PCI-Express x8 interface achieves 2.93 and 2.46 times faster computation than a 3.4GHz Pentium4 processor and a 2.2GHz Opteron processor, respectively, for 2-dimensional time-dependent fluid dynamics problems.

AB - This paper presents an FPGA-based streaming computation for the lattice Boltzmann method (LBM) to simulate fluid flow with floating-point calculations. LBM is suitable for streaming computation because of its parallelism and regularity. We optimize the equations of LBM, and then formulate a streaming computation. To design an efficient data-path for throughput and hardware resource utilization, we introduce multiple cycle inputs and computing-unit sharing to the streaming data-path. The streaming accelerator implemented on a Virtex-4 FPGA with PCI-Express x8 interface achieves 2.93 and 2.46 times faster computation than a 3.4GHz Pentium4 processor and a 2.2GHz Opteron processor, respectively, for 2-dimensional time-dependent fluid dynamics problems.

UR - http://www.scopus.com/inward/record.url?scp=50149121349&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=50149121349&partnerID=8YFLogxK

U2 - 10.1109/FPT.2007.4439254

DO - 10.1109/FPT.2007.4439254

M3 - Conference contribution

AN - SCOPUS:50149121349

SN - 1424414725

SN - 9781424414727

T3 - ICFPT 2007 - International Conference on Field Programmable Technology

SP - 233

EP - 236

BT - ICFPT 2007 - International Conference on Field Programmable Technology

T2 - International Conference on Field Programmable Technology, ICFPT 2007

Y2 - 12 December 2007 through 14 December 2007

ER -

FPGA-based streaming computation for lattice Boltzmann method

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this