DiffPower: GPU-Accelerated Differentiable Switching Power Analysis and Optimization
The paper presents DiffPower, a GPU-accelerated framework for differentiable power analysis and optimization, achieving up to 1002x speedup over single-threaded CPU propagation and enabling applications like gradient-weighted cell sizing and power virus generation.
The paper presents a GPU-accelerated framework for differentiable power analysis and optimization, achieving significant speedup and enabling new applications.
Keywords
Before reading this…
Applications
- →gradient-weighted cell sizing
- →power virus generation
To understand this paper, make sure you know these concepts first:
- understanding of electronic designfind papers →
- knowledge of power analysis and optimizationfind papers →
Abstract
More Like ThisAccurate 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.