PetClaw: A scalable parallel nonlinear wave propagation solver for Python

Handle URI:
http://hdl.handle.net/10754/333645
Title:
PetClaw: A scalable parallel nonlinear wave propagation solver for Python
Authors:
Alghamdi, Amal; Ahmadia, Aron ( 0000-0002-2573-2481 ) ; Ketcheson, David I. ( 0000-0002-1212-126X ) ; Knepley, Matthew ( 0000-0002-2292-0735 ) ; Mandli, Kyle ( 0000-0002-8267-5989 ) ; Dalcin, Lisandro ( 0000-0001-8086-0155 )
Abstract:
We present PetClaw, a scalable distributed-memory solver for time-dependent nonlinear wave propagation. PetClaw unifies two well-known scientific computing packages, Clawpack and PETSc, using Python interfaces into both. We rely on Clawpack to provide the infrastructure and kernels for time-dependent nonlinear wave propagation. Similarly, we rely on PETSc to manage distributed data arrays and the communication between them.We describe both the implementation and performance of PetClaw as well as our challenges and accomplishments in scaling a Python-based code to tens of thousands of cores on the BlueGene/P architecture. The capabilities of PetClaw are demonstrated through application to a novel problem involving elastic waves in a heterogeneous medium. Very finely resolved simulations are used to demonstrate the suppression of shock formation in this system.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division; Numerical Mathematics Group
Publisher:
19th High Performance Computing Symposium (HPC 2011)
Journal:
Proceedings of the High Performance Computing Symposium 2011
Issue Date:
2011
Type:
Conference Paper
Additional Links:
http://numerics.kaust.edu.sa/papers/petclaw2011/petclaw2011.html; http://hosting.cs.vt.edu/hpc2011/
Appears in Collections:
Conference Papers; Numerical Mathematics Group; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorAlghamdi, Amalen
dc.contributor.authorAhmadia, Aronen
dc.contributor.authorKetcheson, David I.en
dc.contributor.authorKnepley, Matthewen
dc.contributor.authorMandli, Kyleen
dc.contributor.authorDalcin, Lisandroen
dc.date.accessioned2014-11-04T08:51:51Z-
dc.date.available2014-11-04T08:51:51Z-
dc.date.issued2011en
dc.identifier.urihttp://hdl.handle.net/10754/333645en
dc.description.abstractWe present PetClaw, a scalable distributed-memory solver for time-dependent nonlinear wave propagation. PetClaw unifies two well-known scientific computing packages, Clawpack and PETSc, using Python interfaces into both. We rely on Clawpack to provide the infrastructure and kernels for time-dependent nonlinear wave propagation. Similarly, we rely on PETSc to manage distributed data arrays and the communication between them.We describe both the implementation and performance of PetClaw as well as our challenges and accomplishments in scaling a Python-based code to tens of thousands of cores on the BlueGene/P architecture. The capabilities of PetClaw are demonstrated through application to a novel problem involving elastic waves in a heterogeneous medium. Very finely resolved simulations are used to demonstrate the suppression of shock formation in this system.en
dc.language.isoenen
dc.publisher19th High Performance Computing Symposium (HPC 2011)en
dc.relation.urlhttp://numerics.kaust.edu.sa/papers/petclaw2011/petclaw2011.htmlen
dc.relation.urlhttp://hosting.cs.vt.edu/hpc2011/en
dc.subjectwave propagationen
dc.subjectPETScen
dc.subjectpetsc4pyen
dc.subjectClawpacken
dc.subjectPyClawen
dc.titlePetClaw: A scalable parallel nonlinear wave propagation solver for Pythonen
dc.typeConference Paperen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.contributor.departmentNumerical Mathematics Groupen
dc.identifier.journalProceedings of the High Performance Computing Symposium 2011en
dc.eprint.versionPost-printen
dc.contributor.institutionUniversity of Chicagoen
dc.contributor.institutionUniversity of Washingtonen
dc.contributor.institutionCIMECen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.