Publication result detail

Weighted Decomposition in High-Performance Lattice-Boltzmann Simulations: Are Some Lattice Sites More Equal than Others?

GROEN, D.; CHACRA, D.; NASH, R.; JAROŠ, J.; BERNABEU, M.; COVENEY, P.

Original Title

Weighted Decomposition in High-Performance Lattice-Boltzmann Simulations: Are Some Lattice Sites More Equal than Others?

English Title

Weighted Decomposition in High-Performance Lattice-Boltzmann Simulations: Are Some Lattice Sites More Equal than Others?

Type

Paper in proceedings (conference paper)

Original Abstract

Obtaining a good load balance is a significant challenge in scaling up lattice-Boltzmann simulations of realistic sparse problems to the exascale. Here we analyse the effect of two optimization techniques on the performance of the HemeLB lattice-Boltzmann simulation environment, when applied to sparse domains. Both techniques are applied prior to partitioning our domain using a third party library. One technique is to provide wall and in/outlet sites with increased weights, and the other is to sort the lattice sites according to a space filling curve. We tested these strategies on a sparse bifurcation and very sparse aneurysm geometry, and find that using weights reduces calculation load imbalance by up to 85%, although the overall communication overhead is higher is some of our runs.

English abstract

Obtaining a good load balance is a significant challenge in scaling up lattice-Boltzmann simulations of realistic sparse problems to the exascale. Here we analyse the effect of two optimization techniques on the performance of the HemeLB lattice-Boltzmann simulation environment, when applied to sparse domains. Both techniques are applied prior to partitioning our domain using a third party library. One technique is to provide wall and in/outlet sites with increased weights, and the other is to sort the lattice sites according to a space filling curve. We tested these strategies on a sparse bifurcation and very sparse aneurysm geometry, and find that using weights reduces calculation load imbalance by up to 85%, although the overall communication overhead is higher is some of our runs.

Keywords

high performance computing; lattice-Boltzmann; domain decomposition

Key words in English

high performance computing; lattice-Boltzmann; domain decomposition

Authors

GROEN, D.; CHACRA, D.; NASH, R.; JAROŠ, J.; BERNABEU, M.; COVENEY, P.

RIV year

2020

Released

19.02.2015

Publisher

Springer Verlag

Location

Stockholm

ISBN

978-3-319-15975-1

Book

Solving Software Challengesfor Exascale

Edition

Lecture Notes in Computer Science

Volume

8759

Pages from

28

Pages to

38

Pages count

11

URL

https://link.springer.com/chapter/10.1007/978-3-319-15976-8_2

BibTex

@inproceedings{BUT163345,
  author="GROEN, D. and CHACRA, D. and NASH, R. and JAROŠ, J. and BERNABEU, M. and COVENEY, P.",
  title="Weighted Decomposition in High-Performance Lattice-Boltzmann Simulations: Are Some Lattice Sites More Equal than Others?",
  booktitle="Solving Software Challengesfor Exascale",
  year="2015",
  series="Lecture Notes in Computer Science",
  volume="8759",
  pages="28--38",
  publisher="Springer Verlag",
  address="Stockholm",
  doi="10.1007/978-3-319-15976-8\{_}2",
  isbn="978-3-319-15975-1",
  url="https://link.springer.com/chapter/10.1007/978-3-319-15976-8_2"
}