Mapping of cores on to diagonal mesh-based network-on-chip

In this work, a genetic solution to map cores onto the diagonal mesh-based network-on-chip architecture is proposed. We take the communication cost as the optimization function of the mapping algorithm based on the genetic algorithm. We also, modify an existing deadlock free routing algorithm for diagonal mesh. Although, diagonal mesh is similar to the traditional mesh topology, it generally has better performance, power consumption, and saturation points than mesh and reduces the hop count between the directly connected cores in the core graph. The results of the proposed technique are compared with those of the traditional mesh based network on chips. The comparative study is performed for VOPD and MPEG-4 which are two complex benchmarks. The results show that the combination of diagonal mesh topology, genetic algorithm based mapping technique, and deadlock free routing algorithm outperforms the mesh-based NoCs.

[1]  David S. Johnson,et al.  Computers and Intractability: A Guide to the Theory of NP-Completeness , 1978 .

[2]  Radu Marculescu,et al.  Energy- and performance-aware mapping for regular NoC architectures , 2005, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[3]  Luca Benini,et al.  Analysis of power consumption on switch fabrics in network routers , 2002, DAC '02.

[4]  Srinivasan Murali,et al.  Bandwidth-constrained mapping of cores onto NoC architectures , 2004, Proceedings Design, Automation and Test in Europe Conference and Exhibition.

[5]  An-Yeu Wu,et al.  A New Binomial Mapping and Optimization Algorithm for Reduced-Complexity Mesh-Based On-Chip Network , 2007, First International Symposium on Networks-on-Chip (NOCS'07).

[6]  Radu Marculescu,et al.  Energy-aware mapping for tile-based NoC architectures under performance constraints , 2003, ASP-DAC '03.

[7]  Coniferous softwood.  GENERAL TERMS , 2003 .

[8]  Jinwen Li,et al.  An energy-aware heuristic constructive mapping algorithm for Network on Chip , 2009, 2009 IEEE 8th International Conference on ASIC.

[9]  Ahmad Khademzadeh,et al.  CGMAP: a new approach to Network-on-Chip mapping problem , 2009, IEICE Electron. Express.

[10]  K. Wendy Tang,et al.  Diagonal and Toroidal Mesh Networks , 1994, IEEE Trans. Computers.

[11]  Mitsuo Gen,et al.  Genetic algorithms and engineering optimization , 1999 .

[12]  Ahmad Khademzadeh,et al.  Chain-Mapping for mesh based Network-on-Chip architecture , 2009, IEICE Electron. Express.

[13]  Vincenzo Catania,et al.  A methodology for design of application specific deadlock-free routing algorithms for NoC systems , 2006, Proceedings of the 4th International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS '06).

[14]  Federico Silla,et al.  On the development of a communication-aware task mapping technique , 2004, J. Syst. Archit..

[15]  Ahmad Khademzadeh,et al.  Onyx: A new heuristic bandwidth-constrained mapping of cores onto tile-based Network on Chip , 2009, IEICE Electron. Express.

[16]  Majid Janidarmian,et al.  Application-Specific Networks-on-Chips Design , 2011 .

[17]  Eric R. Ziegel,et al.  Genetic Algorithms and Engineering Optimization , 2002, Technometrics.

[19]  Ahmad Khademzadeh,et al.  Sorena: New on Chip Network Topology Featuring Efficient Mapping and Simple Deadlock Free Routing Algorithm , 2010, 2010 10th IEEE International Conference on Computer and Information Technology.