Accelerating fully resolved simulation of particle-laden flows on heterogeneous computer architectures--颗粒学报PARTICUOLOGY

Ma, K., Jiang, M., & Liu, Z. (2023). Accelerating fully resolved simulation of particle-laden flows on heterogeneous computer architectures. Particuology, 81, 25-37. https://doi.org/10.1016/j.partic.2022.12.010

Accelerating fully resolved simulation of particle-laden flows on heterogeneous computer architectures

Kuang Ma ^a, Maoqiang Jiang^{a b *}, Zhaohui Liu ^a

a State Key Laboratory of Coal Combustion, School of Energy and Power Engineering, Huazhong University of Science and Technology, Wuhan, 430074, China
b Department of Mechanical Engineering, National University of Singapore, 117575, Singapore

10.1016/j.partic.2022.12.010

Volume 81, October 2023, Pages 25-37

Received 2 June 2022, Revised 23 November 2022, Accepted 14 December 2022, Available online 31 December 2022, Version of Record 11 January 2023.

E-mail: jiangmq@hust.edu.cn; jmq.myllife@163.com

Highlights

• An efficient FR-DNS method for particle-laden flows accelerated by multi-GPUs was developed.

• A single GPU V100 achieved 7.1–11.1 times speed up compared to a single CPU (32 cores).

• MPI parallel efficiency achieved 0.68–0.9 for strong scaling on up to 64 DCUs even for dense flow.

• Up to date largest scale gas-solid flow with 1.6 × 10⁹ meshes and 1.6 × 10⁶ particles was simulated on 32 DCUs.

Abstract

An efficient computing framework, namely PFlows, for fully resolved-direct numerical simulations of particle-laden flows was accelerated on NVIDIA General Processing Units (GPUs) and GPU-like accelerator (DCU) cards. The framework is featured as coupling the lattice Boltzmann method for fluid flow with the immersed boundary method for fluid-particle interaction, and the discrete element method for particle collision, using two fixed Eulerian meshes and one moved Lagrangian point mesh, respectively. All the parts are accelerated by a fine-grained parallelism technique using CUDA on GPUs, and further using HIP on DCU cards, i.e., the calculation on each fluid grid, each immersed boundary point, each particle motion, and each pair-particle collision is responsible by one computer thread, respectively. Coalesced memory accesses to LBM distribution functions with the data layout of Structure of Arrays are used to maximize utilization of hardware bandwidth. Parallel reduction with shared memory for data of immersed boundary points is adopted for the sake of reducing access to global memory when integrate particle hydrodynamic force. MPI computing is further used for computing on heterogeneous architectures with multiple CPUs-GPUs/DCUs. The communications between adjacent processors are hidden by overlapping with calculations. Two benchmark cases were conducted for code validation, including a pure fluid flow and a particle-laden flow. The performances on a single accelerator show that a GPU V100 can achieve 7.1–11.1 times speed up, while a single DCU can achieve 5.6–8.8 times speed up compared to a single Xeon CPU chip (32 cores). The performances on multi-accelerators show that parallel efficiency is 0.5–0.8 for weak scaling and 0.68–0.9 for strong scaling on up to 64 DCU cards even for the dense flow (φ = 20%). The peak performance reaches 179 giga lattice updates per second (GLUPS) on 256 DCU cards by using 1 billion grids and 1 million particles. At last, a large-scale simulation of a gas-solid flow with 1.6 billion grids and 1.6 million particles was conducted using only 32 DCU cards. This simulation shows that the present framework is prospective for simulations of large-scale particle-laden flows in the upcoming exascale computing era.

Graphical abstract

Keywords

Lattice Boltzmann method; Immersed boundary method; Particle-laden flows; Heterogeneous acceleration; General Processing Units

文章导读