Short info: GPU version of the 2Dxy scheme, 9th UnTRIM Workshop, Trento, 7-9 May 2012.
The ongoing developments of the GPU-implementation of the 2Dxy semi-implicit scheme for general free surface flows. Moving towards a generic application by providing a user interface and embedding the CUDA computational core in C++ structures, as well as code verification and validation. Continue reading
Jacek A. Jankowski. Bringing the principal 2Dxy scheme to a GPU, 8th UnTRIM Workshop, Trento, 9-11 May 2011.
The first part of the presentation concerns the process of finding an optimal GPU porting method for the principal two-dimensional, vertically integrated semi-implicit scheme for general free surface flows. Continue reading
Jacek A. Jankowski. General-purpose GPUs – why interesting?, 7th UnTRIM Workshop, Trento, 3-5 May 2010.
The philosophy behind the general purpose GPU computing with arguments pro and against is discussed with the background of initiated projects to adapt specific numeric schemes to apply this technology. Continue reading
Jacek A. Jankowski. Improving and optimising UnTRIM MPI library, 6th UnTRIM Workshop, Trento, 11-13 May 2009.
A number of improvements to the present MPI version of UnTRIM containing the iterative treatment of the wetting and drying are presented: Continue reading
Jacek A. Jankowski. Verification and validation of the parallel UnTRIM, 5th UnTRIM Workshop, Trento, 19-21 May 2008.
A new UnTRIM version has been released (see Casulli 2008), featuring better data locality, improved interpolations in the advection scheme and, the most important, a radically new mass-conservative iterative approach to drying and wetting treatment based on Newton iterations with a proven convergence. In the presentation, detailed issues detected during the verification and validation of the refreshed MPI-parallel “2009” version of UnTRIM are presented, treated and resolved. Continue reading