Adaptive Neural Network-Based Approximation to Accelerate Eulerian Fluid Simulation
This work addresses the problem of inflexible neural network acceleration for Eulerian fluid simulation in HPC, offering an incremental improvement by automating model adaptation.
The paper tackles the lack of flexibility and generalization in neural network-based acceleration of Eulerian fluid simulation by introducing Smartfluidnet, a framework that automates model generation and dynamic switching to meet execution time and simulation quality requirements, achieving speedups of 1.46x and 590x compared to a state-of-the-art neural model and original simulation, respectively, with better simulation quality.
The Eulerian fluid simulation is an important HPC application. The neural network has been applied to accelerate it. The current methods that accelerate the fluid simulation with neural networks lack flexibility and generalization. In this paper, we tackle the above limitation and aim to enhance the applicability of neural networks in the Eulerian fluid simulation. We introduce Smartfluidnet, a framework that automates model generation and application. Given an existing neural network as input, Smartfluidnet generates multiple neural networks before the simulation to meet the execution time and simulation quality requirement. During the simulation, Smartfluidnet dynamically switches the neural networks to make the best efforts to reach the user requirement on simulation quality. Evaluating with 20,480 input problems, we show that Smartfluidnet achieves 1.46x and 590x speedup comparing with a state-of-the-art neural network model and the original fluid simulation respectively on an NVIDIA Titan X Pascal GPU, while providing better simulation quality than the state-of-the-art model.