FlexTiles: A globally homogeneous but locally heterogeneous manycore architecture

Autor: Romain Brillu, Sébastien Pillement, Aymen Abdellah, Fabrice Lemonnier, Philippe Millet
Přispěvatelé: Institut d'Électronique et des Technologies du numéRique (IETR), Université de Nantes (UN)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), École Nationale Supérieure de Techniques Avancées (ENSTA Paris), Embedded System Lab (ESL), Thales Research and Technology, Charlier, Sandrine, Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)
Jazyk: angličtina
Rok vydání: 2014
Předmět:
Zdroj: Workshop on Rapid Simulation & Performance Evaluation: Methods and Tools (RAPIDO 2014)
Workshop on Rapid Simulation & Performance Evaluation: Methods and Tools
Workshop on Rapid Simulation & Performance Evaluation: Methods and Tools, Jan 2014, Vienne, Austria. pp.RAPIDO 2014
RAPIDO
Popis: International audience; This paper introduces the FlexTiles platform, which consist of a manycore architecture associated with a complete tool ow. The di erent components of the manycore architecture are based on general purpose processors (GPP), low power DSP cores and an eFPGA on which dedicated IPs can be dynamically con gured at run-time. Thus, in order to mask the underlying heterogeneity of such an architecture, innovative software mechanism and hardware interface were de ned. These features enable a breakthrough in term of computing performance while improving the on-line adaptive capabilities. Given the large variety of possible use cases that these platforms must support and the resulting workload variability, o ine approaches are no longer su cient because they do not allow coping with time changing workloads. In order to facilitate the architecture adaptation under di erent scenarios, a programming model that considers both static and dynamic behaviors is proposed. The proposed architecture has been implemented on a FPGA platform and has shown the validity of the proposed solution.
Databáze: OpenAIRE