https://ww8central.ww.uni-erlangen.de/ 2026-05-18T11:09:58+00:00 https://ww8central.ww.uni-erlangen.de/ https://ww8central.ww.uni-erlangen.de/_media/wiki/logo.png text/html 2020-07-08T09:33:57+00:00 Anonymous (anonymous@undisclosed.example.com) https://ww8central.ww.uni-erlangen.de/programming/gpgpu/cuda?rev=1594200837&do=diff CUDA Website <http://numpy.scipy.org> CUDA 8.0 * C++98, no later C++ version Docker/Podman CUDA can be used in Docker/Podman on all Fedora32-systems with NVIDIA-driver installed (even legacy 390) even on legacy systems without CUDA. The difficulty is running it root-less. text/html 2016-02-02T09:29:35+00:00 Anonymous (anonymous@undisclosed.example.com) https://ww8central.ww.uni-erlangen.de/programming/gpgpu/hsa?rev=1454405375&do=diff HSA With HIP and Boltzman AMD supports CUDA on their GPUs and with a complete open source software stack to get this running it is very interesting. Installation <http://gpuopen.com/getting-started-with-boltzmann-components-platforms-installation/> text/html 2016-02-03T09:09:04+00:00 Anonymous (anonymous@undisclosed.example.com) https://ww8central.ww.uni-erlangen.de/programming/gpgpu/opencl?rev=1454490544&do=diff openCL Fedora 23 OpenCL on all systems. Example Intel Intel Integrated Graphics Beignet 1.2 (OpenCL 1.2, OpenCL C 1.2) Results on IvyBridge Platform Name Intel Gen OCL Driver Platform Vendor Intel Platform Version OpenCL 1.2 beignet 1.1.1 (git-2eea2c9) Platform Profile FULL_PROFILE Platform Extensions cl_khr_global_int32_bas… text/html 2017-07-27T14:50:03+00:00 Anonymous (anonymous@undisclosed.example.com) https://ww8central.ww.uni-erlangen.de/programming/gpgpu/start?rev=1501167003&do=diff GPGPU hardware accelerated computing CUDA NVIDIA only. Many libraries, wide support in scientific software. CUDA HSA HSA OpenCL Vendor neutral. Intel GPUs and even possible on the CPU for testing OpenCL Vulkan compute The new graphics language also contains support for computing via OpenCL text/html 2017-06-28T09:40:34+00:00 Anonymous (anonymous@undisclosed.example.com) https://ww8central.ww.uni-erlangen.de/programming/gpgpu/vulkan?rev=1498642834&do=diff Vulkan compute Well documented Example <https://github.com/Erkaman/vulkan_minimal_compute> LunarG SDK The LunarG SDK, Examples KhronosGroup ICD loader and validation layers are already provided with the vulkan vulkan-devel packages. Supported Hardware Intel i-3xxx and later (Haswell i-4xxx , Sky-Lake i-6xxx tested)