Introduction

State of GPU Computing in Engineering Simulations

In engineering simulation, there is a strong quest for more computational power in order to run larger models and get shorter turn-around time. The massively parallel architecture of GPU has been proven to be suitable for many of the key algorithmic patterns used in engineering simulations such as structured grids, unstructured grids, dense and sparse matrix operations, spectral methods, particle methods, and so on.

In the past few years, GPU computing techniques are beginning to be widely adopted in production engineering simulation codes. In the near future, research will be needed on porting more complex numerical algorithms and codes to GPU, some of which are not very suitable ...

Get GPU Computing Gems Jade Edition now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.