Word-length selection for power minimization via nonlinear optimization

This article describes the first method for minimizing the dynamic power consumption of a Digital Signal Processing (DSP) algorithm implemented on reconfigurable hardware via word-length optimization. Fast models for estimating the power consumption of the arithmetic components and the routing power of these algorithm implementations are used within a constrained nonlinear optimization formulation that solves a relaxed version of word-length optimization. Tight lower and upper bounds on the cost of the integer word-length problem can be obtained using the proposed solution, with typical upper bounds being 2.9% and 5.1% larger than the lower bounds for area and power consumption, respectively. Heuristics can then use the upper bound as a starting point from which to get even closer to the known lower bound. Results show that power consumption can be improved by up to 40% compared to that achieved when using simple word-length selection techniques, and further comparisons are made between the minimization of different cost functions that give insight into the advantages offered by multiple word-length optimization.

[1]  Donald E. Knuth,et al.  The art of computer programming: sorting and searching (volume 3) , 1973 .

[2]  Alice C. Parker,et al.  Accuracy sensitive word-length selection for algorithm optimization , 1998, Proceedings International Conference on Computer Design. VLSI in Computers and Processors (Cat. No.98CB36273).

[3]  Farid N. Najm,et al.  Power estimation techniques for FPGAs , 2004 .

[4]  Jan M. Rabaey,et al.  Activity-sensitive architectural power analysis , 1996, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[5]  George A. Constantinides,et al.  High-Level Power Optimization for Digital Signal Processing in Reconfigurable Logic , 2006, 2006 International Conference on Field Programmable Logic and Applications.

[6]  George A. Constantinides,et al.  Parameterized logic power consumption models for FPGA-based arithmetic , 2005, International Conference on Field Programmable Logic and Applications, 2005..

[7]  Wonyong Sung,et al.  Simulation-based word-length optimization method for fixed-point digital signal processing systems , 1995, IEEE Trans. Signal Process..

[8]  David Gregg,et al.  Stochastic Bit-Width Approximation Using Extreme Value Theory for Customizable Processors , 2004, CC.

[9]  Prithviraj Banerjee,et al.  Macro-models for high level area and power estimation on FPGAs , 2004, GLSVLSI '04.

[10]  W. Luk,et al.  Truncation noise in fixed-point SFGs [digital filters] , 1999 .

[11]  Shing-Chow Chan,et al.  The wordlength determination problem of linear time invariant systems with multiple outputs - a geometric programming approach , 2006, 2006 IEEE International Symposium on Circuits and Systems.

[12]  Shilpa Bhoj,et al.  Pre-Route Interconnect Capacitance and Power Estimation in FPGAs , 2007, 2007 International Conference on Field Programmable Logic and Applications.

[13]  Peter Y. K. Cheung,et al.  On the Feasibility of Early Routing Capacitance Estimation for FPGAs , 2007, 2007 International Conference on Field Programmable Logic and Applications.

[14]  Paul T. Boggs,et al.  Sequential Quadratic Programming , 1995, Acta Numerica.

[15]  Steven J. E. Wilton,et al.  Activity-based power estimation and characterization of DSP and multiplier blocks in FPGAs , 2006, 2006 IEEE International Conference on Field Programmable Technology.

[16]  Wayne Luk,et al.  Optimum wordlength allocation , 2002, Proceedings. 10th Annual IEEE Symposium on Field-Programmable Custom Computing Machines.

[17]  David Thomas,et al.  The Art in Computer Programming , 2001 .

[18]  Donald E. Knuth,et al.  The Art of Computer Programming, Vol. 3: Sorting and Searching , 1974 .

[19]  Peter Y. K. Cheung,et al.  Fast word-level power models for synthesis of FPGA-based arithmetic , 2006, 2006 IEEE International Symposium on Circuits and Systems.

[20]  Hai Zhou,et al.  Parallel CAD: Algorithm Design and Programming Special Section Call for Papers TODAES: ACM Transactions on Design Automation of Electronic Systems , 2010 .

[21]  Wayne Luk,et al.  Unifying bit-width optimisation for fixed-point and floating-point designs , 2004, 12th Annual IEEE Symposium on Field-Programmable Custom Computing Machines.

[22]  George A. Constantinides.  Perturbation analysis for word-length optimization , 2003, 11th Annual IEEE Symposium on Field-Programmable Custom Computing Machines, 2003. FCCM 2003..

[23]  George A. Constantinides,et al.  Word-length optimization for differentiable nonlinear systems , 2006, TODE.

[24]  Naresh R. Shanbhag,et al.  Analytical expressions for power dissipation of macro-blocks in DSP architectures , 1999, Proceedings Twelfth International Conference on VLSI Design. (Cat. No.PR00013).

[25]  Wayne Luk,et al.  The Multiple Wordlength Paradigm , 2001, The 9th Annual IEEE Symposium on Field-Programmable Custom Computing Machines (FCCM'01).

[26]  I. Xilinx,et al.  Virtex-II Pro and Virtex-II Pro X Platform FPGAs: Complete data sheet , 2004 .

[27]  Wayne Luk,et al.  Ieee Transactions on Computer-aided Design of Integrated Circuits and Systems Accuracy Guaranteed Bit-width Optimization Abstract— We Present Minibit, an Automated Static Approach for Optimizing Bit-widths of Fixed-point Feedforward Designs with Guaranteed Accuracy. Methods to Minimize Both the In- , 2022 .

[28]  Gerhard J. Woeginger,et al.  The complexity of multiple wordlength assignment , 2002, Appl. Math. Lett..

[29]  Peter Y. K. Cheung,et al.  Glitch-aware output switching activity from word-level statistics , 2008, 2008 IEEE International Symposium on Circuits and Systems.

[30]  Takashi Horiyama,et al.  Bit-Length Optimization Method for High-Level Synthesis Based on Non-linear Programming Technique , 2006, IEICE Trans. Fundam. Electron. Commun. Comput. Sci..

[31]  Prithviraj Banerjee,et al.  Macro-models for high-level area and power estimation on FPGAs , 2006, Int. J. Simul. Process. Model..

[32]  Li Shang,et al.  High-level power modeling of CPLDs and FPGAs , 2001, Proceedings 2001 IEEE International Conference on Computer Design: VLSI in Computers and Processors. ICCD 2001.

[33]  Steven J. E. Wilton,et al.  A Flexible Power Model for FPGAs , 2002, FPL.

[34]  P. Perona,et al.  Bit-width optimization for configurable DSP's by multi-interval analysis , 2000, Conference Record of the Thirty-Fourth Asilomar Conference on Signals, Systems and Computers (Cat. No.00CH37154).

[35]  Seehyun Kim,et al.  Fixed-point optimization utility for C and C++ based digital signal processing programs , 1995, VLSI Signal Processing, VIII.

[36]  Jan M. Rabaey,et al.  Architectural power analysis: The dual bit type method , 1995, IEEE Trans. Very Large Scale Integr. Syst..

[37]  Kaushik Roy,et al.  Low-Power CMOS VLSI Circuit Design , 2000 .