Accelerating Nonlinear Time-History Analysis with Complex Constitutive Laws via Heterogeneous Memory Management: From 3D Seismic Simulation to Neural Network Training

2026-04-03

AI Summary

High-fidelity nonlinear time-history simulation is hindered by the substantial computational cost of small time steps and the large memory footprint of state variables. This work proposes a CPU-GPU collaborative computing framework based on heterogeneous memory management that overcomes GPU memory capacity limitations by efficiently orchestrating the host’s large-capacity memory with the GPU’s high-throughput processing capability. For the first time, heterogeneous memory technology is successfully applied to large-scale nonlinear time-history simulations. The proposed method supports complex constitutive modeling and massive parallelism, significantly improving both computational efficiency and energy efficiency. It enables high-fidelity three-dimensional seismic response simulations and generates large-scale, high-quality datasets suitable for training neural network surrogate models.

Technology Category

• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Machine Learning: Hardware-aware ML
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

Nonlinear time-history evolution problems employing high-fidelity physical models are essential in numerous scientific domains. However, these problems face a critical dual bottleneck: the immense computational cost of time-stepping and the massive memory requirements for maintaining a vast array of state variables. To address these challenges, we propose a novel framework based on heterogeneous memory management for massive ensemble simulations of general nonlinear time-history problems with complex constitutive laws. Taking advantage of recent advancements in CPU-GPU interconnect bandwidth, our approach actively leverages the large capacity of host CPU memory while simultaneously maximizing the throughput of the GPU. This strategy effectively overcomes the GPU memory wall, enabling memory-intensive simulations. We evaluate the performance of the proposed method through comparisons with conventional implementations, demonstrating significant improvements in time-to-solution and energy-to-solution. Furthermore, we demonstrate the practical utility of this framework by developing a Neural Network-based surrogate model using the generated massive datasets. The results highlight the effectiveness of our approach in enabling high-fidelity 3D evaluations and its potential for broader applications in data-driven scientific discovery.

Problem

• nonlinear time-history analysis
• complex constitutive laws
• memory bottleneck
• computational cost
• high-fidelity simulation

Innovation

• heterogeneous memory management
• nonlinear time-history analysis
• complex constitutive laws
• GPU memory wall
• neural network surrogate