GPU Series: Why Use GPU Accelerators?
Second in a series on GPU computing
10:00 am MST
The second in a series of workshops and tutorials on GPU computing is “Why Use GPU Accelerators” on Thursday, February 17, at 10 a.m. MST. The series of one-hour sessions is for scientists, software engineers, and students in the Earth system sciences to help them prepare to use the extensive GPU-computing capabilities of the new Derecho system. The training is intended for both new and more experienced GPU programmers and requires familiarity with the UNIX shell; Fortran, C/C++, and/or Python; and typical program compilation workflows.
Topics to be covered in this introduction to the pros and cons of GPU computing include:
- Brief history of GPU/accelerator architecture development
- Trends in CPU and GPU performance
- Power consumption, acceleration, availability
- Flynn’s Taxonomy - SIMD vs. SMT vs. SIMT
- Comparing ease of programmability and evaluating GPU software paradigms
- Steps to begin a GPU project or refactoring initiative
As with other sessions in the series, this one will include dedicated Q&A time and some will involve interactive coding with participants.
If you have not already registered for the series and plan to attend this event, please use the link provided. You only need to register once, and can then use the same credentials to attend subsequent sessions on the Zoom meeting platform. By registering and attending, you agree to abide by the UCAR Code of Conduct. These training sessions will be recorded and published publicly for future reference.