Multi-Level-Memory Structures for Adaptive {SPH} Simulations

A hash map-based sparse data structure for highly adaptive SPH on GPUs.

Spatial adaptivity is only useful if the data structures underneath it can keep up. This paper introduces a multi-level memory structure built on sparse hash maps, designed for the access patterns that highly adaptive SPH simulations actually produce on a GPU.

(Winchenbach & Kolb, 2019)

Presented at the Vision, Modeling and Visualization Conference 2019 in Rostock, Germany. Received an Honorable Mention.

References

2019

Multi-Level-Memory Structures for Adaptive SPH Simulations

Rene Winchenbach, and Andreas Kolb

In 24th International Symposium on Vision, Modeling, and Visualization, VMV 2019, Rostock, Germany, September 30 - October 2, 2019, 2019

Abs DOI Bib

In this paper we introduce a novel hash map-based sparse data structure for highly adaptive Smoothed Particle Hydrodynamics (SPH) simulations on GPUs. Our multi-level- memory structure is based on stacking multiple independent data structures, which can be created efficiently from the same particle data by utilizing self- similar particle orderings. Furthermore, we propose three neighbor list algorithms that improve performance, or significantly reduce memory requirements, when compared to Verlet-lists for the overall simulation. Overall, our proposed method significantly improves the performance of spatially adaptive methods, allows for the simulation of unbounded domains and reduces memory requirements without interfering with the simulation.
@inproceedings{DBLP:conf/vmv/Winchenbach019, author = {Winchenbach, Rene and Kolb, Andreas}, title = {Multi-Level-Memory Structures for Adaptive {SPH} Simulations}, booktitle = {24th International Symposium on Vision, Modeling, and Visualization, {VMV} 2019, Rostock, Germany, September 30 - October 2, 2019}, pages = {99--107}, publisher = {Eurographics Association}, year = {2019}, doi = {10.2312/vmv.20191323}, url = {https://doi.org/10.2312/vmv.20191323}, }