Dynamic and On-Line Design Space Exploration for Reconfigurable Architectures

  • Authors:

  • Fakhreddine Ghaffari;Michel Auguin;Mohamed Abid;Maher Ben Jemaa

  • Affiliations:

  • I3S, University of Nice Sophia Antipolis, CNRS, les Algorithmes bat. Euclide, 2000, route des Lucioles BP 121, 06903 Sophia-Antipolis, and Université de Sfax TUNISIA,Computer, Electronics and ...;I3S, University of Nice Sophia Antipolis, CNRS, les Algorithmes bat. Euclide, 2000, route des Lucioles BP 121, 06903 Sophia-Antipolis,;Université de Sfax TUNISIA,Computer, Electronics and Smart engineering, systems design (CES), ENIS, BPW 3038 Sfax, Tunisia;Université de Sfax TUNISIA,Computer, Electronics and Smart engineering, systems design (CES), ENIS, BPW 3038 Sfax, Tunisia

  • Venue:
  • Transactions on High-Performance Embedded Architectures and Compilers I
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

The implementation of complex embedded applications requires a mix of processor cores and HW accelerators on a single chip. When designing such complex and heterogeneous System on Chip (SoCs), the HW/SW partitioning needs to be made prior to refining the system description. Traditional system partitioning is generally done at the early stage of system architecture, by defining the tasks to be implemented on the embedded processor(s), and the tasks to be implemented on the hardware. We describe here a new approach of On-line Partitioning Algorithm (OPA) which consists of adapting dynamically the architecture to the processing requirements. A scheduling heuristic is associated to this partitioning approach. We consider soft real time data flow graph oriented applications for which the execution time is dependent on the content of input data. The target architecture is composed of a generic processor connected to a dynamically reconfigurable hardware accelerator. The dynamic reconfiguration allows the self adaptation of the architecture which avoids redesigning a new architecture according to variation of characteristics of applications algorithms. We compare our method with an Off-line static HW/SW partitioning approach. We present results of the OPA on an image processing application. Our experiments included simulation results with SystemC for on-line scheduling and partitioning approaches. An ILP solver is used to compare the experiment results with an off-line static HW/SW partitioning approach.