Neighbour lists for smoothed particle hydrodynamics on GPUs
| Main Author: | Ballantyne, John |
|---|---|
| Other Authors: | Winkler, Daniel, Rezavand, Massoud, Rauch, Wolfgang |
| Format: | Dataset |
| Terbitan: |
Mendeley
, 2018
|
| Subjects: | |
| Online Access: |
https:/data.mendeley.com/datasets/nsx8hfcycv |
| ctrlnum |
0.17632-nsx8hfcycv.1 |
|---|---|
| fullrecord |
<?xml version="1.0"?>
<dc><creator>Ballantyne, John</creator><title>Neighbour lists for smoothed particle hydrodynamics on GPUs</title><publisher>Mendeley</publisher><description>The efficient iteration of neighbouring particles is a performance critical aspect of any high performance smoothed particle hydrodynamics (SPH) solver. SPH solvers that implement a constant smoothing length generally divide the simulation domain into a uniform grid to reduce the computational complexity of the neighbour search. Based on this method, particle neighbours are either stored per grid cell or for each individual particle, denoted as Verlet list. While the latter approach has significantly higher memory requirements, it has the potential for a significant computational speedup. A theoretical comparison is performed to estimate the potential improvements of the method based on unknown hardware dependent factors. Subsequently, the computational performance of both approaches is empirically evaluated on graphics processing units. It is shown that the speedup differs significantly for different hardware, dimensionality and floating point precision. The Verlet list algorithm is implemented as an alternative to the cell linked list approach in the open-source SPH solver DualSPHysics and provided as a standalone software package.</description><subject>Computational Physics</subject><contributor>Winkler, Daniel</contributor><contributor>Rezavand, Massoud</contributor><contributor>Rauch, Wolfgang</contributor><type>Other:Dataset</type><identifier>10.17632/nsx8hfcycv.1</identifier><rights>GNU Public License Version 3</rights><rights>http://www.gnu.org/licenses/gpl-3.0.en.html</rights><relation>https:/data.mendeley.com/datasets/nsx8hfcycv</relation><date>2018-02-06T14:36:33Z</date><recordID>0.17632-nsx8hfcycv.1</recordID></dc>
|
| format |
Other:Dataset Other |
| author |
Ballantyne, John |
| author2 |
Winkler, Daniel Rezavand, Massoud Rauch, Wolfgang |
| title |
Neighbour lists for smoothed particle hydrodynamics on GPUs |
| publisher |
Mendeley |
| publishDate |
2018 |
| topic |
Computational Physics |
| url |
https:/data.mendeley.com/datasets/nsx8hfcycv |
| contents |
The efficient iteration of neighbouring particles is a performance critical aspect of any high performance smoothed particle hydrodynamics (SPH) solver. SPH solvers that implement a constant smoothing length generally divide the simulation domain into a uniform grid to reduce the computational complexity of the neighbour search. Based on this method, particle neighbours are either stored per grid cell or for each individual particle, denoted as Verlet list. While the latter approach has significantly higher memory requirements, it has the potential for a significant computational speedup. A theoretical comparison is performed to estimate the potential improvements of the method based on unknown hardware dependent factors. Subsequently, the computational performance of both approaches is empirically evaluated on graphics processing units. It is shown that the speedup differs significantly for different hardware, dimensionality and floating point precision. The Verlet list algorithm is implemented as an alternative to the cell linked list approach in the open-source SPH solver DualSPHysics and provided as a standalone software package. |
| id |
IOS7969.0.17632-nsx8hfcycv.1 |
| institution |
Universitas Islam Indragiri |
| affiliation |
onesearch.perpusnas.go.id |
| institution_id |
804 |
| institution_type |
library:university library |
| library |
Teknologi Pangan UNISI |
| library_id |
2816 |
| collection |
Artikel mulono |
| repository_id |
7969 |
| city |
INDRAGIRI HILIR |
| province |
RIAU |
| shared_to_ipusnas_str |
1 |
| repoId |
IOS7969 |
| first_indexed |
2020-04-08T08:12:26Z |
| last_indexed |
2020-04-08T08:12:26Z |
| recordtype |
dc |
| _version_ |
1686587393212153856 |
| score |
17.538404 |