Abstract
In this paper we introduce a new parallel solver for the weakly singular space–time boundary integral equation for the heat equation. The space–time boundary mesh is decomposed into a given number of submeshes. Pairs of the submeshes represent dense blocks in the system matrices, which are distributed among computational nodes by an algorithm based on a cyclic decomposition of complete graphs ensuring load balance. In addition, we employ vectorization and threading in shared memory to ensure intra-node efficiency. We present scalability experiments on different CPU architectures to evaluate the performance of the proposed parallelization techniques. All levels of parallelism allow us to tackle large problems and lead to an almost optimal speedup.
Original language | English |
---|---|
Journal | Computers & mathematics with applications |
Early online date | 2019 |
DOIs | |
Publication status | E-pub ahead of print - 2019 |
Fingerprint
Keywords
- Heat equation
- Parallelization
- Space–time boundary element method
- Vectorization
ASJC Scopus subject areas
- Numerical Analysis
- Computational Mathematics
- Computational Theory and Mathematics
- Modelling and Simulation
Fields of Expertise
- Information, Communication & Computing
Treatment code (Nähere Zuordnung)
- Basic - Fundamental (Grundlagenforschung)
Cite this
A parallel space–time boundary element method for the heat equation. / Dohr, Stefan; Zapletal, Jan; Of, Günther; Merta, Michal; Kravcenko, Michal.
In: Computers & mathematics with applications, 2019.Research output: Contribution to journal › Article › Research › peer-review
}
TY - JOUR
T1 - A parallel space–time boundary element method for the heat equation
AU - Dohr, Stefan
AU - Zapletal, Jan
AU - Of, Günther
AU - Merta, Michal
AU - Kravcenko, Michal
PY - 2019
Y1 - 2019
N2 - In this paper we introduce a new parallel solver for the weakly singular space–time boundary integral equation for the heat equation. The space–time boundary mesh is decomposed into a given number of submeshes. Pairs of the submeshes represent dense blocks in the system matrices, which are distributed among computational nodes by an algorithm based on a cyclic decomposition of complete graphs ensuring load balance. In addition, we employ vectorization and threading in shared memory to ensure intra-node efficiency. We present scalability experiments on different CPU architectures to evaluate the performance of the proposed parallelization techniques. All levels of parallelism allow us to tackle large problems and lead to an almost optimal speedup.
AB - In this paper we introduce a new parallel solver for the weakly singular space–time boundary integral equation for the heat equation. The space–time boundary mesh is decomposed into a given number of submeshes. Pairs of the submeshes represent dense blocks in the system matrices, which are distributed among computational nodes by an algorithm based on a cyclic decomposition of complete graphs ensuring load balance. In addition, we employ vectorization and threading in shared memory to ensure intra-node efficiency. We present scalability experiments on different CPU architectures to evaluate the performance of the proposed parallelization techniques. All levels of parallelism allow us to tackle large problems and lead to an almost optimal speedup.
KW - Heat equation
KW - Parallelization
KW - Space–time boundary element method
KW - Vectorization
UR - http://www.scopus.com/inward/record.url?scp=85059669009&partnerID=8YFLogxK
U2 - 10.1016/j.camwa.2018.12.031
DO - 10.1016/j.camwa.2018.12.031
M3 - Article
JO - Computers & mathematics with applications
JF - Computers & mathematics with applications
SN - 0898-1221
ER -