NeuPow: artificial neural networks for power and behavioral modeling of arithmetic components in 45nm ASICs technology

In this paper, we present a flexible, simple and accurate power modeling technique that can be used to estimate the power consumption of modern technology devices. We exploit Artificial Neural Networks for power and behavioral estimation in Application Specific Integrated Circuits. Our method, called NeuPow, relies on propagating the predictors between the connected neural models to estimate the dynamic power consumption of the individual components. As a first proof of concept, to study the effectiveness of NeuPow, we run both component level and system level tests on the Open GPDK 45 nm technology from Cadence, achieving errors below 1.5% and 9% respectively for component and system level. In addition, NeuPow demonstrated a speed up factor of 2490X.

[1]  Jean-Christophe Prévotet,et al.  Power modeling on FPGA: a neural model for RT-level power estimation , 2018, CF.

[2]  Chaitali Chakrabarti,et al.  Accurate models for estimating area and power of FPGA implementations , 2008, 2008 IEEE International Conference on Acoustics, Speech and Signal Processing.

[3]  Luca Benini,et al.  Regression-based RTL power modeling , 2000, TODE.

[4]  T. Riesgo,et al.  Statistical Power Estimation for IP-Based Design , 2006, IECON 2006 - 32nd Annual Conference on IEEE Industrial Electronics.

[5]  Muhammad Shafique,et al.  Adaptive Energy Management for Dynamically Reconfigurable Processors , 2014, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[6]  Paolo Meloni,et al.  Power and clock gating modelling in coarse grained reconfigurable systems , 2016, Conf. Computing Frontiers.

[7]  Yong-Bin Kim,et al.  Power Estimation in Digital CMOS VLSI Chips , 2005, 2005 IEEE Instrumentationand Measurement Technology Conference Proceedings.

[8]  Yuanyuan Yan,et al.  Neural Network Macromodel for High-Level Power Estimation of CMOS Circuits , 2005, 2005 International Conference on Neural Networks and Brain.

[9]  Ah Chung Tsoi,et al.  Universal Approximation Using Feedforward Neural Networks: A Survey of Some Existing Methods, and Some New Results , 1998, Neural Networks.

[10]  Paolo Meloni,et al.  Modelling and Automated Implementation of Optimal Power Saving Strategies in Coarse-Grained Reconfigurable Architectures , 2016, J. Electr. Comput. Eng..

[11]  Domenik Helms,et al.  Leakage Models for High-Level Power Estimation , 2018, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[12]  Sergio Bampi,et al.  A power-predictive environment for fast and power-aware ASIC-based FIR filter design , 2017, 2017 30th Symposium on Integrated Circuits and Systems Design (SBCCI).