Accelerating an iterative Helmholtz solver with FPGAs

TitleAccelerating an iterative Helmholtz solver with FPGAs
Publication TypeConference
Year of Publication2014
AuthorsArt Petrenko, Tristan van Leeuwen, Diego Oriato, Simon Tilbury, Felix J. Herrmann
Conference NameEAGE Annual Conference Proceedings
KeywordsCGMN, EAGE, FPGA, Helmholtz equation, Kaczmarz, reconfigurable computing

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 (Björck 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× compared with one Intel core.

Citation Keypetrenko2014EAGEaih