Energy-efficient specialization of functional units in a coarse-grained reconfigurable array

Functional units provide the backbone of any spatial accelerator by providing the computing resources. The desire for having rich and expensive functional units is in tension with producing a regular and energy-efficient computing fabric. This paper explores the design trade-off between complex, universal functional units and simpler, limited functional units. We show that a modest amount of specialization reduces the area-delay-energy product of an optimized architecture to 0.86× a baseline architecture. Furthermore, we provide a design guideline that allows an architect to customize the contents of the computing fabric just by examining the profile of benchmarks within the application domains.

[1]  Carl Ebeling,et al.  Static versus scheduled interconnect in Coarse-Grained Reconfigurable Arrays , 2009, 2009 International Conference on Field Programmable Logic and Applications.

[2]  Hartmut Schröder,et al.  Evaluation strategies for coarse grained reconfigurable architectures , 2005, International Conference on Field Programmable Logic and Applications, 2005..

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

[4]  M. Flynn,et al.  Fast division algorithm with a small lookup table , 1999, Conference Record of the Thirty-Third Asilomar Conference on Signals, Systems, and Computers (Cat. No.CH37020).

[5]  Seth Copen Goldstein,et al.  PipeRench: A Reconfigurable Architecture and Compiler , 2000, Computer.

[6]  Ray Andraka,et al.  A survey of CORDIC algorithms for FPGA based computers , 1998, FPGA '98.

[7]  Kiyoung Choi,et al.  Resource sharing and pipelining in coarse-grained reconfigurable architecture for domain-specific optimization , 2005, Design, Automation and Test in Europe.

[8]  Carl Ebeling,et al.  Managing Short-Lived and Long-Lived Values in Coarse-Grained Reconfigurable Arrays , 2010, 2010 International Conference on Field Programmable Logic and Applications.

[9]  Carl Ebeling,et al.  Architecture design of reconfigurable pipelined datapaths , 1999, Proceedings 20th Anniversary Conference on Advanced Research in VLSI.

[10]  Steven J. E. Wilton,et al.  Interconnect architectures for modulo-scheduled coarse-grained reconfigurable arrays , 2004, Proceedings. 2004 IEEE International Conference on Field- Programmable Technology (IEEE Cat. No.04EX921).