A parallelized particle tracing code for CFD simulations in Earth sciences

The problem of convective flows in a highly viscous fluid represents a common research direction in Earth Sciences. In order to trace the convective motion of the fluid material, a source of passive particles (or tracers) that flow at a local convection velocity and do not affect the pattern of flow...

Descripción completa

Detalles Bibliográficos
Autores: VLAD CONSTANTIN MANEA, Marina Manea, Mihai Pomeran, Lucian BESUTIU, Luminita Zlagnean
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2012
País:México
Institución:Universidad de Guanajuato
Repositorio:Repositorio Institucional de la Universidad de Guanajuato
Idioma:inglés
OAI Identifier:oai:repositorio.ugto.mx:20.500.12059/1937
Acceso en línea:http://repositorio.ugto.mx/handle/20.500.12059/1937
Access Level:acceso abierto
Palabra clave:info:eu-repo/classification/cti/1
Fluid simulation
Tracers.
Parallelization
HPC (High Performance Computing)
Earth Sciences
Simulation CFD (Computational Fluid Dynamics)
Simulación de fluidos
Paralelización
HPC (Computación de Alto Rendimiento)
Ciencias de la Tierra
Trazadores
Simulación CFD (Dinámica de Fluidos Computacional)
Descripción
Sumario:The problem of convective flows in a highly viscous fluid represents a common research direction in Earth Sciences. In order to trace the convective motion of the fluid material, a source of passive particles (or tracers) that flow at a local convection velocity and do not affect the pattern of flow is commonly used. It is presented a parallelized tracer code that uses passive and weightless particles with their position computed from their displacement during a small time interval at the velocity of flow -previously calculated for a given point in space and time. The tracer code is integrated in the open source package CitcomS, which is widely used in the solid Earth community (www.geodynamics.org). It is benchmarked the tracer code on the state-of-the-art CyberDyn parallel machine, a High Performance Com-puting (HPC) Cluster with 1 344 computing cores available at the Institute of Geodynamics of the Romanian Academy.