IWOMP 2024 Program
The following OpenMP events were hosted at Pawsey Supercomputing Research Centre in Perth, Australia.
- MPI Forum Meeting – 23-24 Sept. 2024
- IWOMP – 23-25 Sept. 2024
- EuroMPI – 25-27 Sept. 2024
IWOMP 2024 Proceedings
See our Proceedings page.
Monday, 23 September
09:00 AM: IWOMP Keynote I
- Keeping Cool Under Pressure: The Art, Science and Future of Supercomputing in Australia
Speaker: Mark Gray (Pawsey Supercomputing Research Centre)
10:00: Break
10:30: Papers Session I: Current and Future OpenMP Optimization
- Towards Locality-Aware Host-to-Device Offloading in OpenMP
Jannis Klinkenberg, Jan Kraus, Christian Terboven and Matthias Müller - Performance Porting the ExaStar Multi-physics App Thornado On Heterogeneous Systems – A Fortran-OpenMP Code-base Evaluation
Mathialakan Thavappiragasam, Austin Harris, Shaoping Quan, Eirik Endeve, Dahai Guo, William Dieter and Brice Videau - Event-Based OpenMP Tasks for Time-Sensitive GPU-Accelerated Systems
Cyril Cetre, Chenle Yu, Sara Royuela, Rémi Barrere, Eduardo Quiñones and Damien Gratadour
12:00: Lunch
13:30: Papers Session II: Targeting More Devices
- Integrating Multi-FPGA Acceleration to OpenMP Distributed Computing
Pedro Henrique Di Francia Rosso, Lucian Petrica, Nusrat Jahan Lisa, Marcio Machado Pereira, Sandro Rigo, Hervé Yviquel, Vanderlei Bonato, Emilio Francesquini and Guido Araujo - Towards a Scalable and Efficient PGAS-based Distributed OpenMP
Baodi Shan, Mauricio Araya-Polo and Barbara Chapman - Multilayer Multipurpose Caches for OpenMP Target Regions on FPGAs
Julian Brandner, Florian Mayer and Michael Philippsen
15:00: Break
15:30: ARB and Implementer Updates
- Michael Klemm: State of the ARB
- Johannes Doerfert: LLVM and OpenMP
- Michael Klemm: AMD Implementation Update
Tuesday, 24 September
09:00: IWOMP Keynote II
- From ‘MPI+OpenMP’ to ‘MPI in harmony with OpenMP’: Developments in the MPI Standard and how they could interact with OpenMP
Speaker: Martin Schulz (TU Munich)
10:00: Break
10:30: Papers Session III: Best Practices
- Survey of OpenMP Practice in General Open Source Software
Tim Jammer, Christian Iwainsky and Christian Bischof - CI/CD Efforts for Validation, Verification and Benchmarking OpenMP Implementations
Aaron Jarmusch, Felipe Cabarcas, Swaroop Pophale, Andrew Kallai, Johannes Doerfert, Luke Peyralans, Seyong Lee, Joel Denny and Sunita Chandrasekaran - Evaluation of Directive-based Programming Models for Stencil Computation on Current GPGPU Architectures
Baodi Shan, Mauricio Araya-Polo and Barbara Chapman
12:00: Lunch
13:30: Papers Session IV: Tools
- Finding Equivalent OpenMP Fortran and C/C++ Code Snippets Using Large Language Models
Naveed Sekender, Pei-Hung Lin and Chunhua Liao - Visualizing Correctness Issues (Data Races) in OpenMP Programs
Feiyang Jin, Alan Tao, Lechen Yu and Vivek Sarkar - Developing an Interactive OpenMP Programming Book with Large Language Models
Xinyao Yi, Anjia Wang, Yonghong Yan and Chunhua Liao
15:00: Break
15:30: Panel
OpenMP 6.0 is Almost Done; What′s Next?
- Moderator: Bronis R. de Supinski
- Panelists: Michael Klemm, Jannis Klinkenberg, Adrien Roussel and Tom Scogland
- Abstract: TR13, the final comment draft for OpenMP 6.0, was released in August and most of the remaining efforts towards finalizing the content were completed last week (the week preceding this workshop). While the paint is not yet dry, we are nonetheless at a good point to examine the brushstrokes. This panel will discuss not only what is in 6.0 but also what is not and whether we should be looking at it for 6.1 or 7.0.
Wednesday, 25 September
09:00: Papers Session V: Simplifying Parallelization
- Automatic Parallelization and OpenMP Offloading of Fortran Array Notation
Ivan Radanov Ivanov, Jens Domke, Toshio Endo and Johannes Doerfert - Detrimental Task Execution Patterns in Mainstream OpenMP Runtimes
Adam Tuft, Tobias Weinzierl and Michael Klemm
10:00: Break
11:00: Joint IWOMP/EuroMPI Keynote
Building a time machine. How supercomputing is giving astronomy a new window into the past
Speaker: Sarah Pearce (SKA)