PLA Based Application Mapping in MBC

  • Somnath Paul
  • Swarup Bhunia
Chapter

Abstract

This chapter presents a unique implementation of MBC framework which realizes function not by LUTs but rather by programmable logic arrays (PLA). While LUTs sizes tend to be exponentially large with large number of inputs, PLA sizes increase at a much lower rate, thereby making them attractive for representing functions with large number of inputs. From the implementation perspective, the benefit is exponentially smaller memory size compared to a LUT based approach. This leads to considerable improvement in performance and energy for the MBC framework. The challenge is however, conventional random access memories cannot be used to store and retrieve the PLA representation. Content-addressable memories (CAM) are ideal candidates for storing the PLA representation. This chapter describes the CAM based MBC architecture and the corresponding software flow.

References

  1. 1.
    J. Hutchby, M. Garner, “Assessment of the Potential and Maturity of Selected Emerging Research Memory Technologies”. Technical report, ITRS, 2010Google Scholar
  2. 2.
    K. Pagiamtzis, A. Sheikholeslami, “Content-Addressable Memory (CAM) circuits and architectures”. IEEE J. Solid State Circuits, 41(3), 712–727 (2006)CrossRefGoogle Scholar
  3. 3.
    A. Agarwal et al., “A Dual-Supply 4GHz 13fJ/bit/search 65x128b CAM in 65nm CMOS”, in ESSCIRC, 2006Google Scholar
  4. 4.
    A. Kaviani, S. Brown, “Hybrid FPGA Architecture”, in Intl. Symp. on FPGAs, 1996Google Scholar
  5. 5.
    F. Heile, A. Leaver, “Hybrid Product Term and LUT based Architectures Using Embedded Memory Blocks”, in Intl. Symp. on FPGAs, 1999Google Scholar
  6. 6.
    [Online], “Predictive Technology Model”. http://ptm.asu.edu/

Copyright information

© Springer Science+Business Media New York 2014

Authors and Affiliations

  • Somnath Paul
    • 1
  • Swarup Bhunia
    • 2
  1. 1.Intel LabsHillsboroUSA
  2. 2.Department of EECSCase Western Reserve UniversityClevelandUSA

Personalised recommendations