A comparative study of multi-material data structures for computational physics applications
- Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
- Bristol Univ. (United Kingdom); Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
The data structures used to represent the multi-material state of a computational physics application can have a significant impact on its performance. Here, we look at efficient data structures for applications where there may be many materials overall, but where most computational cells contain only one or a few of these materials. We develop simple performance models for selecting possible data structures and programming patterns. We verify the analytic performance models with a small test program of the representative cases. We discuss the impact of these techniques and analysis for multi-material physics applications, which are also applicable to a wide range of sparse computational data structures.
- Research Organization:
- Los Alamos National Laboratory (LANL), Los Alamos, NM (United States); Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
- Sponsoring Organization:
- USDOE National Nuclear Security Administration (NNSA), Office of Defense Programs (DP) (NA-10); Engineering and Physical Sciences Research Council (EPSRC)
- Grant/Contract Number:
- AC52-06NA25396
- OSTI ID:
- 1460652
- Report Number(s):
- LA-UR--17-30766
- Journal Information:
- Computers and Mathematics with Applications (Oxford), Journal Name: Computers and Mathematics with Applications (Oxford) Journal Issue: 2 Vol. 78; ISSN 0898-1221
- Publisher:
- ElsevierCopyright Statement
- Country of Publication:
- United States
- Language:
- English
Similar Records
An electrostatic Particle-In-Cell code on multi-block structured meshes
Physics-Based Multi-State Models of Passive Component Degradation for the R7 Reactor Simulation Environment