Encoder-Decoder Networks for Analyzing Thermal and Power Delivery Networks

Vidya A. Chhabria, Vipul Ahuja, Ashwath Prabhu, Nikhil Patil, Palkesh Jain, Sachin S. Sapatnekar

Research output: Contribution to journalArticlepeer-review

Abstract

Power delivery network (PDN) analysis and thermal analysis are computationally expensive tasks that are essential for successful integrated circuit (IC) design. Algorithmically, both these analyses have similar computational structure and complexity as they involve the solution to a partial differential equation of the same form. This article converts these analyses into image-to-image and sequence-to-sequence translation tasks, which allows leveraging a class of machine learning models with an encoder-decoder-based generative (EDGe) architecture to address the time-intensive nature of these tasks. For PDN analysis, we propose two networks: (i) IREDGe: a full-chip static and dynamic IR drop predictor and (ii) EMEDGe: electromigration (EM) hotspot classifier based on input power, power grid distribution, and power pad distribution patterns. For thermal analysis, we propose ThermEDGe, a full-chip static and dynamic temperature estimator based on input power distribution patterns for thermal analysis. These networks are transferable across designs synthesized within the same technology and packing solution. The networks predict on-chip IR drop, EM hotspot locations, and temperature in milliseconds with negligibly small errors against commercial tools requiring several hours.

Original languageEnglish (US)
Article number3
JournalACM Transactions on Design Automation of Electronic Systems
Volume28
Issue number1
DOIs
StatePublished - Dec 10 2022

Bibliographical note

Funding Information:
This work was supported in part by the DARPA OpenROAD project, the University of Minnesota Doctoral Dissertation Fellowship, and the Louise Dosdall Fellowship.

Publisher Copyright:
© 2022 Association for Computing Machinery. All rights reserved.

Keywords

  • IR drop
  • Machine learning (ML) for electronic design automation (EDA)
  • Power delivery network (PDN) analysis
  • Thermal analysis
  • U-Nets
  • electromigration (EM)
  • encoder-decoder networks

Fingerprint

Dive into the research topics of 'Encoder-Decoder Networks for Analyzing Thermal and Power Delivery Networks'. Together they form a unique fingerprint.

Cite this