We discuss how algorithmic skeletons (and structured parallel program- ming models in general) can be used to efficiently and seamlessly program multi- core as well as many-core systems. We introduce a new version of the muskel skeleton library that can be used to target multi/many-core systems and we present experimental results that demonstrate the feasibility of the approach. The experi- mental results presented also give an idea of the computational grains that can be exploited on current, state-of-the-art multi-core systems.

Skeletons for multi/many-core systems

ALDINUCCI, MARCO;
2010-01-01

Abstract

We discuss how algorithmic skeletons (and structured parallel program- ming models in general) can be used to efficiently and seamlessly program multi- core as well as many-core systems. We introduce a new version of the muskel skeleton library that can be used to target multi/many-core systems and we present experimental results that demonstrate the feasibility of the approach. The experi- mental results presented also give an idea of the computational grains that can be exploited on current, state-of-the-art multi-core systems.
2010
ParCo 2009: Parallel Computing
Lyon, France
September 2009
Parallel Computing: From Multicores and GPU's to Petascale
IOS Press
19
265
272
9781607505297
http://dx.doi.org/10.3233/978-1-60750-530-3-265
http://www.booksonline.iospress.nl/Content/View.aspx?piid=16568
Algorithmic skeletons; multi-core; many-core; multi-threading
Marco Aldinucci; Marco Danelutto; Peter Kilpatrick
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2318/75122
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 7
  • ???jsp.display-item.citation.isi??? 1
social impact