Two Particle-in-Grid Realisations on Spacetrees
- Submitting institution
-
University of Durham
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 100376
- Type
- D - Journal article
- DOI
-
10.1016/j.parco.2015.12.007
- Title of journal
- Parallel Computing
- Article number
- -
- First page
- 42
- Volume
- 52
- Issue
- -
- ISSN
- 01678191
- Open access status
- Out of scope for open access requirements
- Month of publication
- -
- Year of publication
- 2016
- URL
-
https://doi.org/10.1016/j.parco.2015.12.007
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
3
- Research group(s)
-
A - Innovative Computing
- Citation count
- 5
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Particle-grid simulations usually require mesh sizes to match the particle speeds while they suffer from latency. Our ideas remove the mesh constraint and predict where communication becomes obsolete and, hence, can be skipped. Latency and extremely small time steps induced by fast particles and small meshes are showstoppers in supercomputing. This international, interdisciplinary collaboration removes them and thus lays one foundation stone for an ExCALIBUR bid around SPH. As the techniques facilitated upscaling over the latency-sensitive then-N8 machine, it also led to an invitation to CIUK by the machine vendor OCF plus subsequent research partnerships on hardware-led software co-development.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -