High Performance Computing

Massively parallel processors

Modern computing hardware is capable of parallelisation at multiple granularities. This includes course grain parallelisation such as distributed computing and multicore processing, and fine gran parallelisation such as single instruction multiple data (SIMD) processing units available in many CPUs and GPUs. Through computing software frameworks such as OpenMP, OpenCL or OpenACC it is possible to exploit modern hardware to achieve substantial performance improvements.

Publications

DJ Warne, SA Sisson, C Drovandi. (2021) Vector operations for accelerating Bayesian computation -- A tutorial guide. Bayesian Analysis, 17(2):593-622 DOI arXiv.org
AS Hurn, KA Lindsay, DJ Warne. (2016) A heterogeneous computing approach to maximum likelihood parameter estimation for the Heston model of stochastic volatility. The ANZIAM Journal, 57:C364-381 DOI
HJ Macintosh, DJ Warne, NA Kelson, JE Banks, TW Farrell. (2015) Implementation of parallel tridiagonal solvers for a heterogeneous computing environment. The ANZIAM Journal 56, C446-C462 DOI

Low power high performance computing with field programmable gate arrays

Field programmable gate arrays are reconfigurable digital circuits that enable customised architectures to be implemented at runtime. This enables algorithms to be encoded directly into the the hardware for high performance computing with incredibly low power compared to CPU and GPU counterparts.

Publications

HJ Macintosh, DJ Warne, NA Kelson, JE Banks, TW Farrell. (2015) Implementation of parallel tridiagonal solvers for a heterogeneous computing environment. The ANZIAM Journal 56, C446-C462 DOI
DJ Warne, NA Kelson, RF Hayward. (2014) Comparison of high level FPGA hardware design for solving tri-diagonal linear systems. Procedia Computer Science, 29:95–101 DOI
DJ Warne, RF Hayward, NA Kelson, JE Banks, L. Mejias. (2014) Pulse-coupled neural network performance for real-time identification of vegetation during forced landing. The ANZIAM Journal, 55:C1–C16 DOI
DJ Warne, NA Kelson, RF Hayward. (2012) Solving tri-diagonal linear systems using field programmable gate arrays. In Proceedings of the 4th International Conference on Computational Methods (ICCM2012). URL

Page updated

Google Sites

Report abuse