High Performance Computing Applications Using Parallel Data Processing Units - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2015

High Performance Computing Applications Using Parallel Data Processing Units

Keyvan Azadbakht
  • Fonction : Auteur
  • PersonId : 999414
Vlad Serbanescu
  • Fonction : Auteur
  • PersonId : 999415
Frank De Boer
  • Fonction : Auteur
  • PersonId : 986104

Résumé

Multicore processors are growing with respect to the number of cores on a chip. In a parallel computation context, multicore platforms have several important features such as exploiting multiple parallel processes, having access to a shared memory with noticeably lower cost than the distributed alternative and optimizing different levels of parallelism. In this paper, we introduce the Parallel Data Processing Unit (PDPU) which is a group of objects that benefits from the shared memory of the multicore configuration and that consists of two parts: a shared memory for maintaining data consistent, and a set of objects that are processing the data, then producing and aggregating the results concurrently. We then implement two examples in Java that illustrate PDPU behavior, and compare them with their actor based counterparts and show significant performance improvements. We also put forward the idea of integrating PDPU with the actor model which will result in an optimization for a specific spectrum of problems in actor based development.
Fichier principal
Vignette du fichier
978-3-319-24644-4_13_Chapter.pdf (597.98 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01446600 , version 1 (26-01-2017)

Licence

Paternité

Identifiants

Citer

Keyvan Azadbakht, Vlad Serbanescu, Frank De Boer. High Performance Computing Applications Using Parallel Data Processing Units. 6th Fundamentals of Software Engineering (FSEN), Apr 2015, Tehran, Iran. pp.191-206, ⟨10.1007/978-3-319-24644-4_13⟩. ⟨hal-01446600⟩
59 Consultations
80 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More