Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Late Breaking Results: Fast System Technology Co-Optimization Framework for Emerging Technology Based on Graph Neural Networks

Tianliang Ma, Guangxi Fan, Xuguang Sun, Zhihui Deng, Kainlu Low, Leilai Shao

Abstract

This paper proposes a fast system technology co-optimization (STCO) framework that optimizes power, performance, and area (PPA) for next-generation IC design, addressing the challenges and opportunities presented by novel materials and device architectures. We focus on accelerating the technology level of STCO using AI techniques, by employing graph neural network (GNN)-based approaches for both TCAD simulation and cell library characterization, which are interconnected through a unified compact model, collectively achieving over a 100X speedup over traditional methods. These advancements enable comprehensive STCO iterations with runtime speedups ranging from 1.9X to 14.1X and supports both emerging and traditional technologies.

Late Breaking Results: Fast System Technology Co-Optimization Framework for Emerging Technology Based on Graph Neural Networks

Abstract

This paper proposes a fast system technology co-optimization (STCO) framework that optimizes power, performance, and area (PPA) for next-generation IC design, addressing the challenges and opportunities presented by novel materials and device architectures. We focus on accelerating the technology level of STCO using AI techniques, by employing graph neural network (GNN)-based approaches for both TCAD simulation and cell library characterization, which are interconnected through a unified compact model, collectively achieving over a 100X speedup over traditional methods. These advancements enable comprehensive STCO iterations with runtime speedups ranging from 1.9X to 14.1X and supports both emerging and traditional technologies.
Paper Structure (5 sections, 1 equation, 3 figures, 4 tables)

This paper contains 5 sections, 1 equation, 3 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Fast System Technology Co-optimization Framework
  3. GNN-based Surrogate Model for TCAD Simulation
  4. Unified Compact Model for Emerging Technologies
  5. GNN-based Fast Cell Library Characterization Model

Figures (3)

  • Figure 1: Fast STCO framework based on GNN.
  • Figure 2: Unified device encoding scheme based on finite element mesh details node and edge features with task-specific self-consistent quantities.
  • Figure 3: Validations of the proposed TFT model with measured I-V curves: (a) CNT-TFT with $L=25\mu m$ and $W=125\mu m$; (b) LTPS-TFT with $L=16\mu m$ and $W=40\mu m$; (c) IGZO-TFT with $L=20\mu m$ and $W=30\mu m$.