An Efficient Architecture for the Implementation of Message Passing Programming Model on Massive Multiprocessor

  • Authors:
  • Ferid Gharsalli;Amer Baghdadi;Marius Bonaciu;Giedrius Majauskas;Wander Cesario;Ahmed A. Jerraya

  • Affiliations:
  • TIMA laboratory;ENST Bretagne;TIMA laboratory;TIMA laboratory;TIMA laboratory;TIMA laboratory

  • Venue:
  • RSP '04 Proceedings of the 15th IEEE International Workshop on Rapid System Prototyping
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, we present a generic high performancearchitecture model for massive data transfer and computationapplications. This application-specific architecturebecomes crucial particularly for the design of emergingembedded video applications. Indeed, classic embeddedsystem architectures have limited resources (clock frequencyand memory) and consequently they cannot reachthe required performance needed by emerging application.Moreover, the mapping of high level programmingmodels (e.g. message passing) on fixed architectures introduceslots of software and communication overhead.This paper proposes an efficient architecture for the implementationof Message Passing Programming Modelfor MPSoC. Flexibility, scalability and high performanceof this architecture are illustrated through the implementationof a real time embedded DivX video encoder.