Programming for GPUs: Introduction to OpenACC 2.0 and CUDA 5.5 - videos online
On December 4-6, 2013 CSCS organized a 3-day training course on GPU programming. Experts from Cray and NVIDIA, together with CSCS staff introduced the participants to the basics of programming in the latest versions of OpenACC and CUDA, together with tools for code debugging and parallel performance analysis. The course involved also practical training and tutorials.
Themis Athanassiadou from CSCS welcomed the 22 participants and gave a short introduction of CSCS and the 3 day course. One and half day was dedicated to Alistair Hart from Cray who introduced the audience to OpenACC, followed by Peter Messmer from NVIDIA who gave an overview on CUDA, CUDA toolkit and GPU optimization. The last afternoon was reserved to hybrid performance and debugging tools introduced by Jean-Guillaume Piccinali from CSCS.
The videos of the course are now available online »