Name
Abstract | ||
|---|---|---|
The fast multipole method (FMM) is often used to accelerate the calculation of particle interactions in particle-based methods to simulate incompressible flows. To evaluate the most time-consuming kernels -- the Biot-Savart equation and stretching term of the vorticity equation, we mathematically reformulated it so that only two Laplace scalar potentials are used instead of six. This automatically ensuring divergence-free far-field computation. Based on this formulation, we developed a new FMM-based vortex method on heterogeneous architectures, which distributed the work between multicore CPUs and GPUs to best utilize the hardware resources and achieve excellent scalability. The algorithm uses new data structures which can dynamically manage inter-node communication and load balance efficiently, with only a small parallel construction overhead. This algorithm can scale to large-sized clusters showing both strong and weak scalability. Careful error and timing trade-off analysis are also performed for the cutoff functions induced by the vortex particle method. Our implementation can perform one time step of the velocity+stretching calculation for one billion particles on 32 nodes in 55.9 seconds, which yields 49.12 Tflop/s. |
Year | DOI | Venue |
|---|---|---|
2014 | 10.1109/IPDPSW.2014.110 | IPDPS Workshops |
Keywords | Field | DocType |
heterogeneous architectures,divergence-free far-field computation,vortex methods,hardware resources,computational fluid dynamics,velocity stretching calculation,flow simulation,vorticity equation,graphics processing units,multicore gpu,vortices,internode communication,incompressible flow simulation,gpgpu,particle interaction calculation,time 55.9 s,scalable fast multipole accelerated vortex methods,laplace scalar potentials,vortex particle method,biot-savart equation,two-phase flow,laplace equations,fmm,multicore cpu,heterogeneous algorithm,mathematical model,two phase flow,kernel | Data structure,Multipole expansion,Computer science,Parallel computing,Vortex,Vorticity equation,General-purpose computing on graphics processing units,Fast multipole method,Scalability,Computation | Conference |
Citations | PageRank | References |
0 | 0.34 | 5 |
Authors | ||
5 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Qi Hu | 1 | 0 | 0.68 |
| Nail A. Gumerov | 2 | 404 | 40.62 |
| Rio Yokota | 3 | 209 | 25.73 |
| Lorena A. Barba | 4 | 52 | 7.70 |
| Ramani Duraiswami | 5 | 1721 | 161.98 |