Introduction to OpenMP Offload Part 2 : Optimization and Data Movement
Introduction to OpenMP Offload Part 2 : Optimization and Data Movement
September 1, 2022
1:00 – 3:30 PM (EDT)
Virtual via Zoom
Contact: Suzanne Parete-Koon ([email protected])
This session, offered by OLCF and NERSC, is part of the OLCF’s Preparing for Frontier training series. Please click this link to visit the main page to see other sessions in the series.
Overview
The OpenMP API is a scalable model that gives parallel programmers a simple and flexible interface for developing portable parallel applications in C/C++ and Fortran. In its inception it was targeted for shared memory parallelism, however current versions of OpenMP also enable data and program execution to offload from the host CPU to target GPUS and their associated memory on the same node.
This is part 2 of a two-part OpenMP Offload training that will enable application teams and developers to accelerate their code with the use of GPUs, as well as exploiting the latest OpenMP functionality to program multi-core platforms like Summit and Perlmutter.
Both Part 1 and Part 2 of this training will have focused discussion about what you need to to prepare for using OpenMP on Frontier.
For part 2 OLCF/NERSC staff will cover optimization strategies and show how efficient data movement and a better understanding of the hierarchy of parallelism available can lead to improved performance. Hands-on exercises will follow the lectures. NERSC/OLCF staff will also cover best practices for OpenMP Offload.
Agenda
Time in EDT | ||
1:00 PM | Introduction to OpemMP Offload: Data Movement | Swaroop Pophale, Computer Scientist, CSMD, ORNL |
1:45 PM | Break | |
1:50 PM | Best Practices for OpenMP |
Christopher Daley, HPC Performance Engineer, NERSC
|
2:20 PM | Hands on Introduction |
Wael Elwasif, Computer Scientist, ORNL
|
2:40 PM | Hands on Time | |
3:30 PM | Close zoom |
For details about Part 1, which covers offloading basics, (August 11, 2022), see https://www.olcf.ornl.gov/calendar/introduction-to-openmp-offload-part-1/.
Target Audience
The target audience for this event are NERSC Perlmutter users and current OLCF users who do not yet have access to OLCF’s pre-Frontier (HPE+AMD) systems, but would like to start learning how to program GPUs with OpenMP. The hands-on sessions will be performed on OLCF Summit and NERSC Perlmutter. Participants without existing accounts will be provided training accounts on Perlmutter.
Registration
All participants must register to attend this event. Please do so by clicking the dropdown below and submit the registration form.
Registration is Closed
[tw-toggle title=”Slides”]
Slides:
Introduction to OpenMp: Data Movement
Recordings:
Introduction to OpenMp: Data Movement
[/tw-toggle]