Fast Coherent Particle Advection through Time-Varying Unstructured Flow Datasets

Mingcheng Chen, Shawn C. Shadden, John C Hart

Research output: Contribution to journalArticle

Abstract

Tracing the paths of collections of particles through a flow field is a key step for many flow visualization and analysis methods. When a flow field is interpolated from the nodes of an unstructured mesh, the process of advecting a particle must first find which cell in the unstructured mesh contains the particle. Since the paths of nearby particles often diverge, the parallelization of particle advection quickly leads to incoherent memory accesses of the unstructured mesh. We have developed a new block advection GPU approach that reorganizes particles into spatially coherent bundles as they follow their advection paths, which greatly improves memory coherence and thus shared-memory GPU performance. This approach works best for flows that meet the CFL criterion on unstructured meshes of uniformly sized elements, small enough to fit at least two timesteps in GPU memory.

Original languageEnglish (US)
Article number7243356
Pages (from-to)1959-1972
Number of pages14
JournalIEEE Transactions on Visualization and Computer Graphics
Volume22
Issue number8
DOIs
StatePublished - Aug 1 2016

Fingerprint

Advection
Data storage equipment
Flow fields
Flow visualization
Graphics processing unit

Keywords

  • GPU
  • Unsteady flow
  • parallel algorithms
  • unstructured mesh

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Computer Graphics and Computer-Aided Design

Cite this

Fast Coherent Particle Advection through Time-Varying Unstructured Flow Datasets. / Chen, Mingcheng; Shadden, Shawn C.; Hart, John C.

In: IEEE Transactions on Visualization and Computer Graphics, Vol. 22, No. 8, 7243356, 01.08.2016, p. 1959-1972.

Research output: Contribution to journalArticle

@article{c79efc2ec80a4b9dbf3cd24b4d3f1731,
title = "Fast Coherent Particle Advection through Time-Varying Unstructured Flow Datasets",
abstract = "Tracing the paths of collections of particles through a flow field is a key step for many flow visualization and analysis methods. When a flow field is interpolated from the nodes of an unstructured mesh, the process of advecting a particle must first find which cell in the unstructured mesh contains the particle. Since the paths of nearby particles often diverge, the parallelization of particle advection quickly leads to incoherent memory accesses of the unstructured mesh. We have developed a new block advection GPU approach that reorganizes particles into spatially coherent bundles as they follow their advection paths, which greatly improves memory coherence and thus shared-memory GPU performance. This approach works best for flows that meet the CFL criterion on unstructured meshes of uniformly sized elements, small enough to fit at least two timesteps in GPU memory.",
keywords = "GPU, Unsteady flow, parallel algorithms, unstructured mesh",
author = "Mingcheng Chen and Shadden, {Shawn C.} and Hart, {John C}",
year = "2016",
month = "8",
day = "1",
doi = "10.1109/TVCG.2015.2476795",
language = "English (US)",
volume = "22",
pages = "1959--1972",
journal = "IEEE Transactions on Visualization and Computer Graphics",
issn = "1077-2626",
publisher = "IEEE Computer Society",
number = "8",

}

TY - JOUR

T1 - Fast Coherent Particle Advection through Time-Varying Unstructured Flow Datasets

AU - Chen, Mingcheng

AU - Shadden, Shawn C.

AU - Hart, John C

PY - 2016/8/1

Y1 - 2016/8/1

N2 - Tracing the paths of collections of particles through a flow field is a key step for many flow visualization and analysis methods. When a flow field is interpolated from the nodes of an unstructured mesh, the process of advecting a particle must first find which cell in the unstructured mesh contains the particle. Since the paths of nearby particles often diverge, the parallelization of particle advection quickly leads to incoherent memory accesses of the unstructured mesh. We have developed a new block advection GPU approach that reorganizes particles into spatially coherent bundles as they follow their advection paths, which greatly improves memory coherence and thus shared-memory GPU performance. This approach works best for flows that meet the CFL criterion on unstructured meshes of uniformly sized elements, small enough to fit at least two timesteps in GPU memory.

AB - Tracing the paths of collections of particles through a flow field is a key step for many flow visualization and analysis methods. When a flow field is interpolated from the nodes of an unstructured mesh, the process of advecting a particle must first find which cell in the unstructured mesh contains the particle. Since the paths of nearby particles often diverge, the parallelization of particle advection quickly leads to incoherent memory accesses of the unstructured mesh. We have developed a new block advection GPU approach that reorganizes particles into spatially coherent bundles as they follow their advection paths, which greatly improves memory coherence and thus shared-memory GPU performance. This approach works best for flows that meet the CFL criterion on unstructured meshes of uniformly sized elements, small enough to fit at least two timesteps in GPU memory.

KW - GPU

KW - Unsteady flow

KW - parallel algorithms

KW - unstructured mesh

UR - http://www.scopus.com/inward/record.url?scp=84977110661&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84977110661&partnerID=8YFLogxK

U2 - 10.1109/TVCG.2015.2476795

DO - 10.1109/TVCG.2015.2476795

M3 - Article

VL - 22

SP - 1959

EP - 1972

JO - IEEE Transactions on Visualization and Computer Graphics

JF - IEEE Transactions on Visualization and Computer Graphics

SN - 1077-2626

IS - 8

M1 - 7243356

ER -