Thermal and IR Drop Analysis Using Convolutional Encoder-Decoder Networks

TL;DR

This paper introduces encoder-decoder neural networks that rapidly predict on-chip temperature and IR drop maps from power data, significantly reducing analysis time with high accuracy for IC design.

Contribution

It presents two novel neural network models, ThermEDGe and IREDGe, for fast, accurate, and design-independent temperature and IR drop analysis in integrated circuits.

Findings

ThermEDGe predicts temperature with 0.6% average error.

IREDGe predicts IR drop with 0.008% average error.

Models run in milliseconds, outperforming traditional tools.

Abstract

Computationally expensive temperature and power grid analyses are required during the design cycle to guide IC design. This paper employs encoder-decoder based generative (EDGe) networks to map these analyses to fast and accurate image-to-image and sequence-to-sequence translation tasks. The network takes a power map as input and outputs the corresponding temperature or IR drop map. We propose two networks: (i) ThermEDGe: a static and dynamic full-chip temperature estimator and (ii) IREDGe: a full-chip static IR drop predictor based on input power, power grid distribution, and power pad distribution patterns. The models are design-independent and must be trained just once for a particular technology and packaging solution. ThermEDGe and IREDGe are demonstrated to rapidly predict the on-chip temperature and IR drop contours in milliseconds (in contrast with commercial tools that require several hours or more) and provide an average error of 0.6% and 0.008% respectively.