Plasma Exascale-Performance Simulations CoE - Pushing flagship plasma simulations codes to tackle exascale-enabled Grand Challenges via performance optimisation and codesign | Plasma-PEPSC | Project | Results | HORIZON | CORDIS

Livrables

Initial assessment of developments and revision of the scientific challenges - Enabling Next-Generation Plasma Accelerators for Real World Applications with PIConGPU

Report on performance and update of scientific challenges and performance evaluation plan for PIConGPU.

Initial report on MPI, load balancing and resilience in plasma codes

Initial report on adapting and optimising code using MPI, load balance and resilience in the plasma codes

Initial report on improving parallel I/O, enabling in-situ and compression

Report on strategies for parallel I/O, load balance, and resilience in the plasma codes. First I/O measurement data by at least two codes available and reported in a technical note.

Collaboration plan with definition of common objectives and activities including milestone

This report includes the collaboration plan with associated milestones, the initial communication, and exploitation plans for the project.

Initial assessment of developments and revision of the scientific challenges - Controlling Plasma-Material Interfaces with BIT

Report on performance and update of scientific challenges and performance evaluation plan for the BIT codes

Initial definition of the scientific challenges and initial evaluation plan - Predicting Near-Earth Space Dynamics with Vlasiator

Report on the identified scientific challenges and performance evaluation of the Vlasiator code.

Initial definition of the scientific challenges and initial evaluation plan - Controlling Plasma-Material Interfaces with BIT

Report on the identified scientific challenge and initial performance evaluation for the BIT codes showcase.

Initial report on algorithmic changes and memory optimisation for heterogeneous systems in Plasma-PEPSC codes

Initial report on algorithmic changes and memory optimisation for heterogeneous systems in Plasma-PEPSC codes.

Report on dissemination, community building, collaborations, training, and exploitation

Report on Plasma-PEPSC achievements in dissemination, community building, collaborations, training, and exploitation. This report also includes an updated collaboration plan.

Initial definition of the scientific challenges and initial evaluation plan - Enabling Next-Generation Plasma Accelerators for Real World Applications with PIConGPU

Report on the identified scientific challenges and performance evaluation of the PIConGPU code.

Initial assessment of developments and revision of the scientific challenges - Optimising Magnetically Confined Plasmas with GENE

Report on performance and update of scientific challenges and performance evaluation plan for GENE.

Initial Assessment of Future Emerging Technologies and Algorithms for Plasma Simulations

Initial review and analysis of post-exascale technologies for plasma simulations

Initial assessment of developments and revision of the scientific challenges - Predicting Near-Earth Space Dynamics with Vlasiator

Report on performance and update of scientific challenges and performance evaluation plan for Vlasiator

Initial definition of the scientific challenges and initial evaluation plan - Optimising Magnetically Confined Plasmas with GENE

Report on the identified scientific challenges and performance evaluation of the GENE code.

Data Management Plan

DMP Report for data and codes in Plasma-PEPSC

Publications

MPI performance analysis in vlasiator: Unraveling communication bottlenecks

Auteurs: Jennifer Faj, Jeremy J. Williams, Ivy B. Peng, Urs Ganse, Markus Battarbee, Yann Pfau-Kempf, Leo Kotipalo, Minna Palmroth and Stefano Markidis
Publié dans: SC23: The International Conference for High Performance Computing, Networking, Storage, and Analysis, Denver, CO, USA, November 12-17 2023, 2023
Éditeur: Supercomputing 2023

Enabling Quantum Computer Simulations on AMD GPUs: a HIP Backend for Google's qsim

Auteurs: Stefano Markidis
Publié dans: SC23 Workshop on Quantum Software, 2023, ISSN 2168-135X
Éditeur: ACM
DOI: 10.1145/3624062.3624223

Leveraging HPC Profiling & Tracing Tools to Understand the Performance of Particle-in-Cell Monte Carlo Simulations

Auteurs: Jeremy J. Williams, David Tskhakaya, Stefan Costea, Ivy B. Peng, Marta Garcia-Gasulla and Stefano Markidis
Publié dans: Euro-Par 2023: Parallel Processing Workshops: Euro-Par 2023 International Workshops, Limassol, Cyprus, August 28–September 1, 2023, 2023
Éditeur: Springer Nature
DOI: 10.48550/arXiv.2306.16512

Non-Blocking GPU-CPU Notifications to Enable More GPU-CPU Parallelism

Auteurs: Bengisu Elis, Olga Pearce, David Boehme, Jason Burmark and Martin Schulz
Publié dans: HPCAsia '24: Proceedings of the International Conference on High Performance Computing in Asia-Pacific Region, 2023
Éditeur: Association for Computing Machinery
DOI: 10.1145/3635035.3635036

Quantum Computer Simulations at Warp Speed: Assessing the Impact of GPU Acceleration

Auteurs: Jennifer Faj, Ivy Peng, Jacob Wahlgren and Stefano Markidis
Publié dans: IEEE 19th International Conference on e-Science (e-Science) 2023, 2023
Éditeur: IEEE 19th International Conference on e-Science (e-Science) 2023
DOI: 10.48550/arXiv.2307.14860

Programming Quantum Neural Networks on NISQ Systems: An Overview of Technologies and Methodologies

Auteurs: Stefano Markidis
Publié dans: Entropy, 2023, ISSN 1099-4300
Éditeur: Multidisciplinary Digital Publishing Institute (MDPI)
DOI: 10.3390/e25040694

Plasma Exascale-Performance Simulations CoE - Pushing flagship plasma simulations codes to tackle exascale-enabled Grand Challenges via performance optimisation and codesign

Livrables

Publications

Partager cette page

Télécharger