Leveraging task-parallelism in message-passing dense matrix factorizations using SMPSs

In this paper, we investigate how to exploit task-parallelism during the execution of the Cholesky factorization on clusters of multicore processors with the SMPSs programming model. Our analysis reveals that the major difficulties in adapting the code for this operation in ScaLAPACK to SMPSs lie in...

Descripción completa

Detalles Bibliográficos
Autores: Martín Huertas, Alberto Francisco|||0000-0001-5751-4561, Reyes, Ruyman, Badia Sala, Rosa Maria|||0000-0003-2941-5499, Quintana Ortí, Enrique Salvador
Tipo de recurso: artículo
Fecha de publicación:2014
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/23465
Acceso en línea:https://hdl.handle.net/2117/23465
https://dx.doi.org/10.1016/j.parco.2014.04.001
Access Level:acceso abierto
Palabra clave:Parallel computation
Clusters of multi-core processors
Linear algebra
Message-passing numerical libraries
Task parallelism
Computació paralel.la
Àrees temàtiques de la UPC::Informàtica::Arquitectura de computadors::Arquitectures paral·leles
Descripción
Sumario:In this paper, we investigate how to exploit task-parallelism during the execution of the Cholesky factorization on clusters of multicore processors with the SMPSs programming model. Our analysis reveals that the major difficulties in adapting the code for this operation in ScaLAPACK to SMPSs lie in algorithmic restrictions and the semantics of the SMPSs programming model, but also that they both can be overcome with a limited programming effort. The experimental results report considerable gains in performance and scalability of the routine parallelized with SMPSs when compared with conventional approaches to execute the original ScaLAPACK implementation in parallel as well as two recent message-passing routines for this operation. In summary, our study opens the door to the possibility of reusing message-passing legacy codes/libraries for linear algebra, by introducing up-to-date techniques like dynamic out-of-order scheduling that significantly upgrade their performance, while avoiding a costly rewrite/reimplementation.