Shared memory pipelined parareal

Publikationstyp
Journal Article
Date Issued
2017
Sprache
English
Author(s)
Ruprecht, Daniel
TORE-URI
http://hdl.handle.net/11420/7070
Start Page
669
End Page
681
Citation
Euro-Par 2017: parallel processing : 23rd International Conference on Parallel and Distributed Computing, Santiago de Compostela, Spain, August 28 – September 1, 2017 : proceedings / Francisco F. Rivera, Tomás F. Pena, José C. Cabaleiro (eds.). - Cham : Springer, 2017. - (Lecture notes in computer science ; 10417). - Seite 669-681
Contribution to Conference
23rd International Conference on Parallel and Distributed Computing, Euro-Par 2017
Publisher DOI
10.1007/978-3-319-64203-1_48
Scopus ID
2-s2.0-85028719035
Publisher
Springer
ISBN of container
978-3-319-64203-1
For the parallel-in-time integration method Parareal, pipelining can be used to hide some of the cost of the serial correction step and improve its efficiency. The paper introduces a basic OpenMP implementation of pipelined Parareal and compares it to a standard MPI-based variant. Both versions yield almost identical runtimes, but, depending on the compiler, the OpenMP variant consumes about 7% less energy and has a significantly smaller memory footprint. However, its higher implementation complexity might make it difficult to use in legacy codes and in combination with spatial parallelisation.
Subjects
OpenMP
Parallel-in-time integration
Parareal
Pipelining
DDC Class
510: Mathematik