GPGPU — General-purpose programming on GPU

48h — 6 CFU

Lecture notes

• Course introduction
• C and C++ programming basics (C part 1, C part 2, C++)
• History of graphic cards and GPGPU (part 1, part 2)
• GPGPU programming basics; introduction to CUDA (first step, launching kernels)
• High-level CUDA programming: the CUDA runtime (working with vectors, textures and multi-dimensional data)
• Benchmarking, optimization and debugging (benchmarking and optimization, GDB and CUDA-GDB, asynchronous CUDA commands )
• Introduction to multi-GPU programming
• Low-level CUDA programming: the CUDA driver interface
• OpenCL basics and heterogeneous GPGPU programming: an introduction, OpenCL images and samplers