Plasma-PEPSC D4.2 Report on Improving Parallel I/O, and Enabling In-Situ and Compression Algorithms in Plasma-PEPSC Codes
This report provides updates on the implementation of parallel I/O and its performance for selected Plasma-PEPSC codes. It dives deeply into several asynchronous and hybrid in-situ techniques for real-time data analysis and discusses lossless and lossy compression and reduced precision arithmetics for data storage, communication, and computation. These innovations aim to optimize data workflows, support large-scale simulations, and mitigate storage bottlenecks inherent to modern HPC systems. The report demonstrates substantial improvements in scalability, resilience, and computational throughput, essential for advancing plasma physics research in an exascale computing era.
Institutions
- University of Helsinki (FI)
- Helmholtz-Zentrum Dresden-Rossendorf (DE)
- Max Planck Computing and Data Facility (DE)
- Center for Advanced Systems Understanding (DE)
- Max Planck Institute for Plasma Physics (DE)
- KTH Royal Institute of Technology (SE)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2025-12-18
- DOI
- https://doi.org/10.5281/zenodo.17978491
- Primary Topic
- Plasma Diagnostics and Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00