A massively parallel code for polarization calculations

Fuente: Redalyc
Guardado en:
Detalles Bibliográficos
Autor principal: Shizuka Akiyama
Formato: Artículo científico
Lenguaje:en
Publicado: Universidad Nacional Autónoma de México 2001
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1876437215167381504
author Shizuka Akiyama
author_facet Shizuka Akiyama
contents A massively parallel code for polarization calculations Shizuka Akiyama Peter Höflich Física, Astronomía y Matemáticas SU PERNOVAE RADIATIVE TRANSFER STARS: ATMOSPHERES We present an implementation of our Monte-Carlo radiation transport methodfor rapidly expanding, NLTE atmospheres for massively parallel computers whichutilizes both the distributed and shared memory models. This allows us to takefull advantage of the fast communication and low latency inherent to nodes withmultiple CPUs, and to stretch the limits of scalability with the number of nodescompared to a version which is based on the shared memory model. Test calculationson a local 20-node Beowulf cluster with dual CPUs showed an improvedscalability by about 40%. 2001 artículo científico 0185-1101 https://www.redalyc.org/articulo.oa?id=57101041 en http://www.redalyc.org/revista.oa?id=571 Revista Mexicana de Astronomía y Astrofísica application/pdf Universidad Nacional Autónoma de México Revista Mexicana de Astronomía y Astrofísica (México) Vol.10
format Artículo científico
id redalyc_57101041
institution Redalyc
language en
publishDate 2001
publisher Universidad Nacional Autónoma de México
spellingShingle A massively parallel code for polarization calculations
Shizuka Akiyama
Física, Astronomía y Matemáticas
SU
PERNOVAE
RADIATIVE TRANSFER
STARS: ATMOSPHERES
A massively parallel code for polarization calculations Shizuka Akiyama Peter Höflich Física, Astronomía y Matemáticas SU PERNOVAE RADIATIVE TRANSFER STARS: ATMOSPHERES We present an implementation of our Monte-Carlo radiation transport methodfor rapidly expanding, NLTE atmospheres for massively parallel computers whichutilizes both the distributed and shared memory models. This allows us to takefull advantage of the fast communication and low latency inherent to nodes withmultiple CPUs, and to stretch the limits of scalability with the number of nodescompared to a version which is based on the shared memory model. Test calculationson a local 20-node Beowulf cluster with dual CPUs showed an improvedscalability by about 40%. 2001 artículo científico 0185-1101 https://www.redalyc.org/articulo.oa?id=57101041 en http://www.redalyc.org/revista.oa?id=571 Revista Mexicana de Astronomía y Astrofísica application/pdf Universidad Nacional Autónoma de México Revista Mexicana de Astronomía y Astrofísica (México) Vol.10
title A massively parallel code for polarization calculations
topic Física, Astronomía y Matemáticas
SU
PERNOVAE
RADIATIVE TRANSFER
STARS: ATMOSPHERES
url https://www.redalyc.org/articulo.oa?id=57101041