Towards a High-Level Implementation of Execution Primitives for Unrestricted, Independent And-Parallelism

Casas, Amadeo; Carro Liñares, Manuel y Hermenegildo, Manuel V. (2008). Towards a High-Level Implementation of Execution Primitives for Unrestricted, Independent And-Parallelism. En: "10th International Symposium on Practical Aspects of Declarative Languages, PADL'08", 07/01/2008-08/01/2008, San Francisco, USA. ISBN 978-3-540-77441-9.

Descripción

Título: Towards a High-Level Implementation of Execution Primitives for Unrestricted, Independent And-Parallelism
Autor/es:
  • Casas, Amadeo
  • Carro Liñares, Manuel
  • Hermenegildo, Manuel V.
Tipo de Documento: Ponencia en Congreso o Jornada (Artículo)
Título del Evento: 10th International Symposium on Practical Aspects of Declarative Languages, PADL'08
Fechas del Evento: 07/01/2008-08/01/2008
Lugar del Evento: San Francisco, USA
Título del Libro: Proceedings of 10th International Symposium on Practical Aspects of Declarative Languages, Practical aspects of declarative languages, PADL'08
Fecha: 7 Enero 2008
ISBN: 978-3-540-77441-9
Materias:
Palabras Clave Informales: Parallelism, Virtual Machines, High-level Implementation.
Escuela: Facultad de Informática (UPM) [antigua denominación]
Departamento: Lenguajes y Sistemas Informáticos e Ingeniería del Software
Licencias Creative Commons: Reconocimiento - Sin obra derivada - No comercial

Texto completo

[img]
Vista Previa
PDF (Document Portable Format) - Se necesita un visor de ficheros PDF, como GSview, Xpdf o Adobe Acrobat Reader
Descargar (206kB) | Vista Previa

Resumen

Most efficient implementations of parallel logic programming rely on complex low-level machinery which is arguably difficult to implement and modify. We explore an alternative approach aimed at taming that complexity by raising core parts of the implementation to the source language level for the particular case of and-parallellism. We handle a significant portion of the parallel implementation at the Prolog level with the help of a comparatively small number of concurrency.related primitives which take case of lower-level tasks such as locking, thread management, stack set management, etc. The approach does not eliminate altogether modifications to the abstract machine, but it does greatly simplify them and it also facilitates experimenting with different alternatives. We show how this approach allows implementing both restricted and unrestricted (i.e., non fork-join) parallelism. Preliminary esperiments show thay the performance safcrifieced is reasonable, although granularity of unrestricted parallelism contributes to better observed speedups.

Más información

ID de Registro: 4369
Identificador DC: http://oa.upm.es/4369/
Identificador OAI: oai:oa.upm.es:4369
URL Oficial: http://www.ist.unomaha.edu/padl2008/
Depositado por: Memoria Investigacion
Depositado el: 30 Sep 2010 08:55
Ultima Modificación: 20 Abr 2016 13:37
  • Open Access
  • Open Access
  • Sherpa-Romeo
    Compruebe si la revista anglosajona en la que ha publicado un artículo permite también su publicación en abierto.
  • Dulcinea
    Compruebe si la revista española en la que ha publicado un artículo permite también su publicación en abierto.
  • Recolecta
  • e-ciencia
  • Observatorio I+D+i UPM
  • OpenCourseWare UPM