OpenMP Target Offload Training

Slides and Recordings:

To find the slides and recordings from this training please login to Summit or Andes and open /gpfs/alpine/stf007/world-shared/OpenMP/openmp.txt and follow the instructions.


NVIDIA HPC SDK – OpenMP Target Offload Training

NVIDIA will presented  a 2-part training series for NERSC and OLCF users about using OpenMP target offload with NVIDIA’s HPC SDK compilers. The training introduced OpenMP target offload, the NVIDIA compilers, and best practices for achieving high performance with OpenMP target offload on NVIDIA GPUs.  Access to Cori GPU nodes was provided.

If you have any questions, please contact Suzanne Parete-Koon (


ate & Time    Topics
Day 1, Tuesday, December 8

12:00 pm – 1:30 pm EST

  • Overview of OpenMP Target Offload project
  • Compiler Usage, system requirements, runtime environment variables, etc…
  • Best practices for NVIDIA GPU’s using OpenMP Target Offload directives
  • Live Demo, homework assignment
Day 2, Thursday, December 10

12:30 pm – 2:30 am EST

  • Review homework assignment
  • Code Example walk-throughs
  • Interoperability with CUDA, with CUDA Fortran, with OpenACC
  • NVLAmath Introduction
  • Q and A


Dec 08 2020 - Dec 10 2020


Suzanne Parete-Koon
QR Code

Comments are closed.