Hybrid programming model for implicit PDE simulations on multicore architectures

Handle URI:
http://hdl.handle.net/10754/564330
Title:
Hybrid programming model for implicit PDE simulations on multicore architectures
Authors:
Kaushik, Dinesh K.; Keyes, David E. ( 0000-0002-4052-7224 ) ; Balay, Satish; Smith, Barry F.
Abstract:
The complexity of programming modern multicore processor based clusters is rapidly rising, with GPUs adding further demand for fine-grained parallelism. This paper analyzes the performance of the hybrid (MPI+OpenMP) programming model in the context of an implicit unstructured mesh CFD code. At the implementation level, the effects of cache locality, update management, work division, and synchronization frequency are studied. The hybrid model presents interesting algorithmic opportunities as well: the convergence of linear system solver is quicker than the pure MPI case since the parallel preconditioner stays stronger when hybrid model is used. This implies significant savings in the cost of communication and synchronization (explicit and implicit). Even though OpenMP based parallelism is easier to implement (with in a subdomain assigned to one MPI process for simplicity), getting good performance needs attention to data partitioning issues similar to those in the message-passing case. © 2011 Springer-Verlag.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division; Applied Mathematics and Computational Science Program; Extreme Computing Research Center
Publisher:
Springer Science + Business Media
Journal:
Lecture Notes in Computer Science
Conference/Event name:
7th International Workshop on OpenMP, IWOMP 2011
Issue Date:
2011
DOI:
10.1007/978-3-642-21487-5_2
Type:
Conference Paper
ISSN:
03029743
ISBN:
9783642214868
Appears in Collections:
Conference Papers; Applied Mathematics and Computational Science Program; Extreme Computing Research Center; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorKaushik, Dinesh K.en
dc.contributor.authorKeyes, David E.en
dc.contributor.authorBalay, Satishen
dc.contributor.authorSmith, Barry F.en
dc.date.accessioned2015-08-04T06:23:54Zen
dc.date.available2015-08-04T06:23:54Zen
dc.date.issued2011en
dc.identifier.isbn9783642214868en
dc.identifier.issn03029743en
dc.identifier.doi10.1007/978-3-642-21487-5_2en
dc.identifier.urihttp://hdl.handle.net/10754/564330en
dc.description.abstractThe complexity of programming modern multicore processor based clusters is rapidly rising, with GPUs adding further demand for fine-grained parallelism. This paper analyzes the performance of the hybrid (MPI+OpenMP) programming model in the context of an implicit unstructured mesh CFD code. At the implementation level, the effects of cache locality, update management, work division, and synchronization frequency are studied. The hybrid model presents interesting algorithmic opportunities as well: the convergence of linear system solver is quicker than the pure MPI case since the parallel preconditioner stays stronger when hybrid model is used. This implies significant savings in the cost of communication and synchronization (explicit and implicit). Even though OpenMP based parallelism is easier to implement (with in a subdomain assigned to one MPI process for simplicity), getting good performance needs attention to data partitioning issues similar to those in the message-passing case. © 2011 Springer-Verlag.en
dc.publisherSpringer Science + Business Mediaen
dc.titleHybrid programming model for implicit PDE simulations on multicore architecturesen
dc.typeConference Paperen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.contributor.departmentApplied Mathematics and Computational Science Programen
dc.contributor.departmentExtreme Computing Research Centeren
dc.identifier.journalLecture Notes in Computer Scienceen
dc.conference.date13 June 2011 through 15 June 2011en
dc.conference.name7th International Workshop on OpenMP, IWOMP 2011en
dc.conference.locationChicago, ILen
dc.contributor.institutionArgonne National Laboratory, Argonne, IL 60439, United Statesen
kaust.authorKeyes, David E.en
kaust.authorKaushik, Dinesh K.en
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.