GPU Series: Introduction to GPU and Accelerator Architectures
Third in a series on GPU computing
10:00 am MST
The third session in a series of workshops and tutorials on GPU computing – “Introduction to GPU and Accelerator Architectures” – will be on Thursday, March 3, at 10 a.m. MST. The presenter will be Brent Leback, a member of the NVIDIA HPC SDK team and a co-creator of CUDA Fortran. He previously served as Compiler Engineering Manager of the Portland Group and continues to participate in compiler development at NVIDIA. He also regularly serves as a GPU Hackathon mentor.
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 session include:
- Comparison between CPU and GPU hardware (Casper/Derecho specific)
- Memory architectures on GPUs and heterogeneous compute platforms
- How to port CPU code to a GPU and common approaches
- Why running a model on GPUs is a better (or worse) choice
This week we welcome Brent Leback from NVIDIA who will be presenting "Introduction to GPU and Accelerator Architectures." Brent is a part of the NVIDIA HPC SDK team and is a co-creator of CUDA Fortran. He has in the past served as Compiler Engineer Manager of the Portland Group (PGI) and continues to actively participate in compiler development at NVIDIA. Brent also regularly serves as a mentor at GPU Hackathon events.
As with other sessions in the series, this one will include dedicated Q&A time and some will involve interactive coding with participants.
Please use this form to register if you plan to attend and have not already registered for the series. 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.