1. Dynamic resource allocation for efficient parallel CFD simulations
- Author
-
Houzeaux, G., Badia, R.M., Borrell, R., Dosimont, D., Ejarque, J., Garcia-Gasulla, M., López, V., Universitat Politècnica de Catalunya. Departament d'Arquitectura de Computadors, Barcelona Supercomputing Center, and Universitat Politècnica de Catalunya. CAP - Grup de Computació d'Altes Prestacions
- Subjects
FOS: Computer and information sciences ,J.2 ,General Computer Science ,D.1 ,D.2 ,J.6 ,Computational fluid dynamics ,Computational Engineering, Finance, and Science (cs.CE) ,Supercomputadors ,FOS: Mathematics ,Mathematics - Numerical Analysis ,Computer Science - Computational Engineering, Finance, and Science ,Resource allocation ,Informàtica::Arquitectura de computadors [Àrees temàtiques de la UPC] ,General Engineering ,Elastic computing ,Numerical Analysis (math.NA) ,Dinàmica de fluids computacional ,Supercomputers ,Parallel efficiency ,Computer Science - Distributed, Parallel, and Cluster Computing ,35-04 ,Assignació de recursos ,MPI ,Distributed, Parallel, and Cluster Computing (cs.DC) ,High performance computing ,CFD - Abstract
CFD users of supercomputers usually resort to rule-of-thumb methods to select the number of subdomains (partitions) when relying on MPI-based parallelization. One common approach is to set a minimum number of elements or cells per subdomain, under which the parallel efficiency of the code is "known" to fall below a subjective level, say 80%. The situation is even worse when the user is not aware of the "good" practices for the given code and a huge amount of resources can thus be wasted. This work presents an elastic computing methodology to adapt at runtime the resources allocated to a simulation automatically. The criterion to control the required resources is based on a runtime measure of the communication efficiency of the execution. According to some analytical estimates, the resources are then expanded or reduced to fulfil this criterion and eventually execute an efficient simulation., Comment: 27 pages, 15 figures
- Published
- 2022
- Full Text
- View/download PDF