DiffPower: GPU-Accelerated Differentiable Switching Power Analysis and Optimization

2026-08-04

AI Summary

This work addresses the longstanding trade-off between speed and accuracy in switching power analysis for physical design by introducing a differentiable power analysis framework based on a process-agnostic bytecode representation. By integrating reverse-mode automatic differentiation, GPU acceleration, and hybrid propagation—combining analytical modeling with parallel simulation—the proposed method enables, for the first time, gradient-based optimization for power consumption. This framework supports novel applications such as cell sizing optimization and power virus generation. Experimental results demonstrate up to a 1002× speedup in power analysis (with a switching activity correlation coefficient of 0.96) and a 904× acceleration in gradient computation compared to single-threaded CPU execution. Furthermore, cell sizing optimization reduces power by up to 2.98×, and power virus generation achieves a 2.13× improvement in efficiency.

Abstract

Accurate and scalable switching power analysis remains a critical bottleneck in modern physical design, often forcing a trade-off between computational speed and modeling fidelity. We present DiffPower, a GPU-accelerated framework for differentiable power analysis and optimization. DiffPower translates design netlists into a PDK-agnostic bytecode representation, enabling analytical gradient computation via reverse-mode automatic differentiation, achieving up to a $1{,}002\times$ speedup over single-threaded CPU propagation on the largest evaluated design, with the GPU advantage growing with design scale. A hybrid propagation methodology fusing analytical modeling with parallel simulation achieves a median toggle-rate correlation of $r{=}0.96$ across ten industrial and benchmark designs. The resulting \emph{power gradients}, computed up to $904\times$ faster than CPU finite-difference methods with near-perfect rank agreement, enable two downstream applications: (1) gradient-weighted cell sizing, which achieves up to $2.98\times$ improvement over local-power heuristics on industrial designs, with even stronger advantages at the 117K-cell scale where competing methods plateau; and (2) power virus generation via gradient ascent, which yields up to $2.13\times$ higher transition-weighted power, replacing a search process that traditionally requires hours.

Problem

• switching power analysis
• physical design
• computational scalability
• modeling fidelity
• power optimization

Innovation

• differentiable power analysis
• GPU acceleration
• automatic differentiation
• power optimization
• hybrid propagation