Leveraging Task-Parallelism in Energy-Efficient ILU Preconditioners
Impacto
Scholar |
Otros documentos de la autoría: Aliaga Estellés, José Ignacio; Dolz, Manuel F.; Martín Huertas, Alberto F.; Mayo, Rafael; Quintana-Orti, Enrique S.
Metadatos
Mostrar el registro completo del ítemcomunitat-uji-handle:10234/9
comunitat-uji-handle2:10234/7036
comunitat-uji-handle3:10234/61544
comunitat-uji-handle4:
INVESTIGACIONEste recurso está restringido
http://dx.doi.org/10.1007/978-3-642-32606-6_5 |
Metadatos
Título
Leveraging Task-Parallelism in Energy-Efficient ILU PreconditionersAutoría
Fecha de publicación
2012Editor
Springer Berlin HeidelbergISBN
978-3-642-32605-9ISSN
0302-9743; 1611-3349Cita bibliográfica
Aliaga, José I. ; Dolz, Manuel F. ; Martín, Alberto F. ; Mayo, Rafael; Quintana Ortí, Enrique S. "Leveraging Task-Parallelism in Energy-Efficient ILU Preconditioners ". En: ICT as Key Technology against Global Warming – Second International Conference, ICT-GLOW 2012, Vienna, Austria, September 6, 2012. Proceedings / Auweter, Axel [et al.] (Eds.). Berlin : Springer, 2012. (Lecture Notes in Computer Science; 7453) . ISBN: 978-3-642-32605-9, pp. 55-63Tipo de documento
info:eu-repo/semantics/bookPartVersión de la editorial
http://link.springer.com/chapter/10.1007%2F978-3-642-32606-6_5Palabras clave / Materias
Resumen
We analyze the energy-performance balance of a task-parallel computation of an ILU-based preconditioner for the solution of sparse linear systems on multi-core processors. In particular, we elaborate a theoretical ... [+]
We analyze the energy-performance balance of a task-parallel computation of an ILU-based preconditioner for the solution of sparse linear systems on multi-core processors. In particular, we elaborate a theoretical model for the power dissipation, and employ it to explore the effect of the processor power states on the time-power-energy interaction for this calculation. Armed with the insights gained from this study, we then introduce two energy-saving mechanisms which, incorporated into the runtime in charge of the parallel execution of the algorithm, improve energy efficiency by 6.9%, with a negligible impact on performance. [-]
Derechos de acceso
© Springer, Part of Springer Science+Business Media
http://rightsstatements.org/vocab/InC/1.0/
info:eu-repo/semantics/restrictedAccess
http://rightsstatements.org/vocab/InC/1.0/
info:eu-repo/semantics/restrictedAccess