A High-Fidelity Flow Solver for Unstructured Meshes on Field-Programmable Gate Arrays

TL;DR

This paper presents a custom FPGA accelerator for a full CFD Poisson solver on unstructured meshes, achieving high performance and energy efficiency, and discusses future opportunities for reconfigurable computing in scientific simulations.

Contribution

It introduces a novel FPGA-based accelerator for the entire Poisson solver in CFD, focusing on unstructured meshes and high-fidelity spectral element methods.

Findings

Achieves over 700 Gflop/s in single-precision performance.

Reduces runtime by more than 2x using geometric factor computation on the fly.

Demonstrates energy-efficient acceleration compared to CPUs.

Abstract

The impending termination of Moore's law motivates the search for new forms of computing to continue the performance scaling we have grown accustomed to. Among the many emerging Post-Moore computing candidates, perhaps none is as salient as the Field-Programmable Gate Array (FPGA), which offers the means of specializing and customizing the hardware to the computation at hand. In this work, we design a custom FPGA-based accelerator for a computational fluid dynamics (CFD) code. Unlike prior work -- which often focuses on accelerating small kernels -- we target the entire Poisson solver on unstructured meshes based on the high-fidelity spectral element method (SEM) used in modern state-of-the-art CFD systems. We model our accelerator using an analytical performance model based on the I/O cost of the algorithm. We empirically evaluate our accelerator on a state-of-the-art Intel Stratix 10 FPGA in terms of performance and power consumption and contrast it against existing solutions on general-purpose processors (CPUs). Finally, we propose a data movement-reducing technique where we compute geometric factors on the fly, which yields significant (700+ Gflop/s) single-precision performance and an upwards of 2x reduction in runtime for the local evaluation of the Laplace operator. We end the paper by discussing the challenges and opportunities of using reconfigurable architecture in the future, particularly in the light of emerging (not yet available) technologies.