I have collected here some numerical tools and examples of the use of GPU parallel computing in the analysis of astronomical data. The source codes will eventually be collected in a public repository but some examples are already included on these pages. Please see specific topics under the “Numerical tools” menu.
The code quoted on these pages is not necessarily optimal (or even always correct!). They are here to show how easy it is to write programs – or to convert existing ones – to take advantage of parallel processing and, in the case of OpenCL, to write a program that performs well both when run on a multi-core CPU or a GPU.