TY - GEN
T1 - Parallel performance of h-type Adaptive Mesh Refinement for Nek5000
AU - Peplinski, Adam
AU - Fischer, Paul F.
AU - Schlatter, Philipp
N1 - Copyright:
Copyright 2017 Elsevier B.V., All rights reserved.
PY - 2016/4/26
Y1 - 2016/4/26
N2 - We discuss parallel performance of h-type Adaptive Mesh Refinement (AMR) developed for the high-order spectral element solver Nek5000 within CRESTA project. AMR is a desired feature of the future simulation software, as it gives possibility to increase the accuracy of numerical simulations at minimal computational cost by resolving particular region of the domain. At the same time it increases complexity of the communication pattern and introduces load imbalance, that can have negative effect on the code scalability. In this work we concentrate on the parallel performance of different tools required by AMR and the resulting algorithm limitations. Our implementation is based on available libraries for parallel mesh management (p4est) and partitioning (ParMetis) that provide necessary information for grid refinement/coarsening and redistribution performed within nonconforming version of Nek5000. For simplicity we consider advection-diffusion problem instead of the full Navies-Stokes equations and study both strong and weak scalability for the convected-cone problem. It is a synthetic test case allowing to test AMR with frequent dynamic mesh adjustments.
AB - We discuss parallel performance of h-type Adaptive Mesh Refinement (AMR) developed for the high-order spectral element solver Nek5000 within CRESTA project. AMR is a desired feature of the future simulation software, as it gives possibility to increase the accuracy of numerical simulations at minimal computational cost by resolving particular region of the domain. At the same time it increases complexity of the communication pattern and introduces load imbalance, that can have negative effect on the code scalability. In this work we concentrate on the parallel performance of different tools required by AMR and the resulting algorithm limitations. Our implementation is based on available libraries for parallel mesh management (p4est) and partitioning (ParMetis) that provide necessary information for grid refinement/coarsening and redistribution performed within nonconforming version of Nek5000. For simplicity we consider advection-diffusion problem instead of the full Navies-Stokes equations and study both strong and weak scalability for the convected-cone problem. It is a synthetic test case allowing to test AMR with frequent dynamic mesh adjustments.
KW - Large-scale scientific computing
KW - Nonconforming methods
KW - Parallel adaptive mesh refinement
KW - Spectral elements
UR - http://www.scopus.com/inward/record.url?scp=85014765450&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85014765450&partnerID=8YFLogxK
U2 - 10.1145/2938615.2938620
DO - 10.1145/2938615.2938620
M3 - Conference contribution
AN - SCOPUS:85014765450
T3 - ACM International Conference Proceeding Series
BT - Proceedings of the 2016 Exascale Applications and Software Conference, EASC 2016
PB - Association for Computing Machinery
T2 - 2016 Exascale Applications and Software Conference, EASC 2016
Y2 - 25 April 2016 through 29 April 2016
ER -