Pareto Optimal Temporal Partition Methodology for Reconfigurable Architectures Based on Multi-objective Genetic Algorithm

A pare to optimal temporal partition methodology was developed for splitting and mapping large data flow graph (DFG) to the coarse-grained reconfigurable architecture (CGRA). A multi-objective genetic algorithm (MOGA) derived from the SPEA-II algorithm was first time introduced to the temporal partition realm for simultaneously optimizing multiple mutually exclusive objectives. Experiments carried out on the ESL (electronic system level) model of the REmus processor show that MOGA based temporal partition algorithms is superior than heuristic algorithm by reducing execution delay 5%-28%, communication overheads 16%-37% without degradation the resource efficiency. Furthermore, comparisons with weight-based multi-objective simulated annealing algorithm show the pare to optimal algorithm can achieve slight better latency objective (3%), while dramatically decrease the communication overheads by at most 21% and the resource efficiency doesn't get worse.

[1]  Marco Laumanns,et al.  SPEA2: Improving the strength pareto evolutionary algorithm , 2001 .

[2]  Marco D. Santambrogio,et al.  Task graph scheduling for reconfigurable architectures driven by reconfigurations hiding and resources reuse , 2009, GLSVLSI '09.

[3]  João M. P. Cardoso.  On Combining Temporal Partitioning and Sharing of Functional Units in Compilation for Reconfigurable Architectures , 2003, IEEE Trans. Computers.

[4]  Camel Tanougast,et al.  Automated RTR temporal partitioning for reconfigurable embedded real-time system design , 2003, Proceedings International Parallel and Distributed Processing Symposium.

[5]  Ann Gordon-Ross,et al.  Runtime Temporal Partitioning Assembly to Reduce FPGA Reconfiguration Time , 2009, 2009 International Conference on Reconfigurable Computing and FPGAs.

[6]  Horácio C. Neto,et al.  An Enhanced Static-List Scheduling Algorithm for Temporal Partitioning onto RPUs , 1999, VLSI.

[7]  D. J. Smith,et al.  A Study of Permutation Crossover Operators on the Traveling Salesman Problem , 1987, ICGA.

[8]  Leibo Liu,et al.  Temporal partitioning algorithm for a coarse-grained reconfigurable computing architecture , 2009, Proceedings of the 2009 12th International Symposium on Integrated Circuits.

[9]  Pedro C. Diniz,et al.  Compiling for reconfigurable computing: A survey , 2010, CSUR.

[10]  Roberto Guerrieri,et al.  A Heterogeneous Digital Signal Processor for Dynamically Reconfigurable Computing , 2010, IEEE Journal of Solid-State Circuits.

[11]  Markus Weinhardt,et al.  PACT XPP—A Self-Reconfigurable Data Processing Architecture , 2003, The Journal of Supercomputing.

[12]  Dinesh Bhatia,et al.  Temporal Partitioning and Scheduling Data Flow Graphs for Reconfigurable Computers , 1999, IEEE Trans. Computers.

[13]  Fadi J. Kurdahi,et al.  MorphoSys: An Integrated Reconfigurable System for Data-Parallel and Computation-Intensive Applications , 2000, IEEE Trans. Computers.

[14]  Abdellatif Mtibaa,et al.  Temporal partitioning of data flow graph for dynamically reconfigurable architecture , 2011, J. Syst. Archit..

[15]  Nikil D. Dutt,et al.  Integrating Physical Constraints in HW-SW Partitioning for Architectures With Partial Dynamic Reconfiguration , 2006, IEEE Transactions on Very Large Scale Integration (VLSI) Systems.

[16]  Ranga Vemuri,et al.  An integrated temporal partitioning and partial reconfiguration technique for design latency improvement , 2000, Proceedings Design, Automation and Test in Europe Conference and Exhibition 2000 (Cat. No. PR00537).