Computational Science Technical Note CSTN-151

CSTN Home

Field Programmable Gate Arrays for Computational Acceleration of Lattice-Oriented Simulation Models

A. Gilman and K. A. Hawick

Archived April 2012

Abstract

Field Programmable Gate Arrays (FPGAs) have attracted recent attention as accelerators for a range of scientific applications which had previously been only practicable on conventional general purpose programming platforms. We report on the performance scalability and software engineering considerations when FPGAs are used to accelerate performance of lattice-oriented simulation models such as complex systems models. We report on the specifics of an FPGA implementation of cellular automaton models like the Game-of-Life. Typical FPGA toolkits and approaches are well suited to the localised updating of the regular data structures of models like these. We review ideas for more widespread uptake of FPGA technology in complex systems simulation applications.

Keywords: FPGA-based design; simulation; complex systems; implementation development; performance evaluation.

Full Document Text: PDF version.

Citation Information: BiBTeX database for CSTN Notes.

BiBTeX reference:

@INPROCEEDINGS{CSTN-151,
  author = {A. Gilman and K.A. Hawick},
  title = {Field Programmable Gate Arrays for Computational Acceleration of
	Lattice-Oriented Simulation Models},
  booktitle = {Proc. Int. Conf. on Computer Design (CDES'12)},
  year = {2012},
  pages = {91-97},
  address = {Las Vegas, USA},
  month = {16-19 July},
  publisher = {CSREA},
  institution = {Computer Science, Massey University},
  timestamp = {2012.05.03}
}


[ CSTN Index | CSTN BiBTeX ]