Minority Report: A Graph Network Oracle for In Situ Visualization

Krishna Kumar,Paul Navrátil,Andrew Solis,Joseph Vantassel

DOI: https://doi.org/10.48550/arXiv.2206.12683

2022-06-26

Abstract:In situ visualization techniques are hampered by a lack of foresight: crucial simulation phenomena can be missed due to a poor sampling rate or insufficient detail at critical timesteps. Keeping a human in the loop is impractical, and defining statistical triggers can be difficult. This paper demonstrates the potential for using a machine-learning-based simulation surrogate as an oracle to identify expected critical regions of a large-scale simulation. These critical regions are used to drive the in situ analysis, providing greater data fidelity and analysis resolution with an equivalent I/O budget to a traditional in situ framework. We develop a distributed asynchronous in situ visualization by integrating TACC Galaxy with CB-Geo MPM for material point simulation of granular flows. We employ a PyTorch-based 3D Graph Network Simulator (GNS) trained on granular flow problems as an oracle to predict the dynamics of granular flows. Critical regions of interests are manually tagged in GNS for in situ rendering in MPM.

Graphics,Computational Engineering, Finance, and Science,Distributed, Parallel, and Cluster Computing,Machine Learning,Geophysics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: **the problem of important simulation phenomena being missed due to the lack of predictability in in - situ visualization techniques**. Specifically, traditional in - situ visualization methods may miss important simulation phenomena due to insufficient sampling rates or insufficient details in key time steps. In addition, it is impractical to keep humans in - the - loop for real - time adjustments, and it is also difficult to define statistical triggers.

To solve these problems, the author proposes a machine - learning - based simulation surrogate, namely Graph Neural Network (GNN), as an "oracle" to identify key regions in large - scale simulations. These key regions are used to drive in - situ analysis, thus providing higher data fidelity and analysis resolution under the same data input - output (I/O) budget.

### Main contributions:
1. **Dynamic Graph Neural Network Simulation (GNS)**: Predict the evolution of particle trajectories.
2. **A method of harvesting metadata using simulation surrogates**: Identify parameters and features to optimize full - scale in - situ analysis.
3. **Demonstration of large - scale column collapse problems**: Show how to use GNS - driven in - situ visualization methods to capture complex physical phenomena that are difficult to express with closed - form solutions.

Through this method, the author hopes to achieve powerful functions similar to post - hoc analysis in in - situ visualization while reducing computational and I/O costs.