I’m going to walk you through a basic install and configuration for a development system to do CUDA and OpenACC GPU programming. This is not a detailed howto but if you have some linux admin skills it will be a reasonable guide to get you started. We’ll do a basic NVIDIA GPU programming setup including CentOS 6.5, CUDA development environment and a PGI compiler setup with OpenACC. The most interesting part may be the OpenACC setup. OpenACC is a relatively new option for GPU programming and allows for a directive (pragma) based coding model.
NVIDIA Tesla K20 plus PGI Accelerator compilers with OpenACC in a package deal with a system. Yes, it’s official. If you’ve wanted to do some development work with OpenACC on Tesla, this is a nice way to get started with a heavily discounted K20 and PGI compiler package pre loaded on a Peak Mini.
Where is NVIDIA heading with High Performance Computing hardware? Ever since Intel announced Xeon Phi Knights Landing as a stand-alone processor integrated at the board level as a full compute unit, I’ve been wondering what NVIDIA would do along these lines. It just makes sense that they would do something similar since getting the GPU off of the PCIe bus and tightly integrated with plentiful system memory would be a huge step forward for usability and performance. Here’s my guess about where NVIDIA is heading.
I had the pleasure of attending the NVIDIA Graphics Technology Conference ( GTC ) last week. Wonderful conference! If you have any doubts about the quality of the conference you are in luck. They have most of the content on-line, you can check it out yourself …
The NVIDIA Tesla accelerator is a well established work-horse for many useful and important High Performance Computing applications and we are happy to be able to provide Tesla acceleration for our “Peak” systems. The developer ecosystem around CUDA is well established, however, at Puget Systems we believe there is new round of developer interest on the horizon that will be catalyzed by the soon to be released 6.x series of the CUDA platform, advances with openACC, new libraries, new hardware, and perhaps significantly, NVIDIA’s acquisition of The Portland Group and their excellent compilers and tools for working with Tesla. So, I’ve loaded up a Peak mini with a Tesla K40 and I’m ready to give Tesla programming a fresh look.