Accelerating an Iterative Helmholtz Solver with FPGAs

A Petrenko; T. van Leeuwen; D Oriato; S. Tilbury; F.J. Herrmann

doi:10.3997/2214-4609.20141141

Accelerating an Iterative Helmholtz Solver with FPGAs

A Petrenko, T. van Leeuwen, D Oriato, S. Tilbury, F.J. Herrmann

Sub Fundamental Mathematics

CWI

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

Abstract

We implement the Kaczmarz row-projection algorithm (Kaczmarz (1937)) on a CPU host + FPGA accelerator platform using techniques of dataflow programming. This algorithm is then used as the preconditioning step in CGMN, a modified version of the conjugate gradients method (Bjorck and Elfving (1979)) that we use to solve the time-harmonic acoustic isotropic constant density wave equation. Using one accelerator we achieve a speed-up of over 2 times compared with one Intel core.

Original language	English
Title of host publication	76th EAGE Conference and Exhibition
Publisher	European Association of Geoscientists and Engineers, EAGE
Pages	16-19
Number of pages	4
DOIs	https://doi.org/10.3997/2214-4609.20141141
Publication status	Published - 2014

Keywords

expanded

Access to Document

10.3997/2214-4609.20141141

https://www.slim.eos.ubc.ca/Publications/Public/Conferences/EAGE/2014/petrenko2014EAGEaih.pdf

Cite this

@inproceedings{6919dc1b73bf4c59957224bc1d1497b8,

title = "Accelerating an Iterative Helmholtz Solver with FPGAs",

abstract = "We implement the Kaczmarz row-projection algorithm (Kaczmarz (1937)) on a CPU host + FPGA accelerator platform using techniques of dataflow programming. This algorithm is then used as the preconditioning step in CGMN, a modified version of the conjugate gradients method (Bjorck and Elfving (1979)) that we use to solve the time-harmonic acoustic isotropic constant density wave equation. Using one accelerator we achieve a speed-up of over 2 times compared with one Intel core.",

keywords = "expanded",

author = "A Petrenko and {van Leeuwen}, T. and D Oriato and S. Tilbury and F.J. Herrmann",

year = "2014",

doi = "10.3997/2214-4609.20141141",

language = "English",

pages = "16--19",

booktitle = "76th EAGE Conference and Exhibition",

publisher = "European Association of Geoscientists and Engineers, EAGE",

}

TY - GEN

T1 - Accelerating an Iterative Helmholtz Solver with FPGAs

AU - Petrenko, A

AU - van Leeuwen, T.

AU - Oriato, D

AU - Tilbury, S.

AU - Herrmann, F.J.

PY - 2014

Y1 - 2014

N2 - We implement the Kaczmarz row-projection algorithm (Kaczmarz (1937)) on a CPU host + FPGA accelerator platform using techniques of dataflow programming. This algorithm is then used as the preconditioning step in CGMN, a modified version of the conjugate gradients method (Bjorck and Elfving (1979)) that we use to solve the time-harmonic acoustic isotropic constant density wave equation. Using one accelerator we achieve a speed-up of over 2 times compared with one Intel core.

AB - We implement the Kaczmarz row-projection algorithm (Kaczmarz (1937)) on a CPU host + FPGA accelerator platform using techniques of dataflow programming. This algorithm is then used as the preconditioning step in CGMN, a modified version of the conjugate gradients method (Bjorck and Elfving (1979)) that we use to solve the time-harmonic acoustic isotropic constant density wave equation. Using one accelerator we achieve a speed-up of over 2 times compared with one Intel core.

KW - expanded

U2 - 10.3997/2214-4609.20141141

DO - 10.3997/2214-4609.20141141

M3 - Conference contribution

SP - 16

EP - 19

BT - 76th EAGE Conference and Exhibition

PB - European Association of Geoscientists and Engineers, EAGE

ER -

Accelerating an Iterative Helmholtz Solver with FPGAs

Abstract

Keywords

Access to Document

Fingerprint

Cite this