A hybrid, massively parallel implementation of a genetic algorithm for optimization of the impact performance of a metal/polymer composite plate

Handle URI:
http://hdl.handle.net/10754/562242
Title:
A hybrid, massively parallel implementation of a genetic algorithm for optimization of the impact performance of a metal/polymer composite plate
Authors:
Narayanan, Kiran; Mora Cordova, Angel; Allsopp, Nicholas; El Sayed, Tamer S.
Abstract:
A hybrid parallelization method composed of a coarse-grained genetic algorithm (GA) and fine-grained objective function evaluations is implemented on a heterogeneous computational resource consisting of 16 IBM Blue Gene/P racks, a single x86 cluster node and a high-performance file system. The GA iterator is coupled with a finite-element (FE) analysis code developed in house to facilitate computational steering in order to calculate the optimal impact velocities of a projectile colliding with a polyurea/structural steel composite plate. The FE code is capable of capturing adiabatic shear bands and strain localization, which are typically observed in high-velocity impact applications, and it includes several constitutive models of plasticity, viscoelasticity and viscoplasticity for metals and soft materials, which allow simulation of ductile fracture by void growth. A strong scaling study of the FE code was conducted to determine the optimum number of processes run in parallel. The relative efficiency of the hybrid, multi-level parallelization method is studied in order to determine the parameters for the parallelization. Optimal impact velocities of the projectile calculated using the proposed approach, are reported. © The Author(s) 2012.
KAUST Department:
Physical Sciences and Engineering (PSE) Division; Core Labs
Publisher:
SAGE Publications
Journal:
International Journal of High Performance Computing Applications
Issue Date:
17-Jul-2012
DOI:
10.1177/1094342012451474
Type:
Article
ISSN:
10943420
Sponsors:
This work was fully funded by the KAUST baseline fund.
Appears in Collections:
Articles; Physical Sciences and Engineering (PSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorNarayanan, Kiranen
dc.contributor.authorMora Cordova, Angelen
dc.contributor.authorAllsopp, Nicholasen
dc.contributor.authorEl Sayed, Tamer S.en
dc.date.accessioned2015-08-03T09:57:43Zen
dc.date.available2015-08-03T09:57:43Zen
dc.date.issued2012-07-17en
dc.identifier.issn10943420en
dc.identifier.doi10.1177/1094342012451474en
dc.identifier.urihttp://hdl.handle.net/10754/562242en
dc.description.abstractA hybrid parallelization method composed of a coarse-grained genetic algorithm (GA) and fine-grained objective function evaluations is implemented on a heterogeneous computational resource consisting of 16 IBM Blue Gene/P racks, a single x86 cluster node and a high-performance file system. The GA iterator is coupled with a finite-element (FE) analysis code developed in house to facilitate computational steering in order to calculate the optimal impact velocities of a projectile colliding with a polyurea/structural steel composite plate. The FE code is capable of capturing adiabatic shear bands and strain localization, which are typically observed in high-velocity impact applications, and it includes several constitutive models of plasticity, viscoelasticity and viscoplasticity for metals and soft materials, which allow simulation of ductile fracture by void growth. A strong scaling study of the FE code was conducted to determine the optimum number of processes run in parallel. The relative efficiency of the hybrid, multi-level parallelization method is studied in order to determine the parameters for the parallelization. Optimal impact velocities of the projectile calculated using the proposed approach, are reported. © The Author(s) 2012.en
dc.description.sponsorshipThis work was fully funded by the KAUST baseline fund.en
dc.publisherSAGE Publicationsen
dc.subjectcompositeen
dc.subjectcomputational steeringen
dc.subjectductile fractureen
dc.subjectimpacten
dc.subjectoptimizationen
dc.subjectparallel genetic algorithmen
dc.titleA hybrid, massively parallel implementation of a genetic algorithm for optimization of the impact performance of a metal/polymer composite plateen
dc.typeArticleen
dc.contributor.departmentPhysical Sciences and Engineering (PSE) Divisionen
dc.contributor.departmentCore Labsen
dc.identifier.journalInternational Journal of High Performance Computing Applicationsen
dc.contributor.institutionCray Computing, Stuttgart, Germanyen
kaust.authorAllsopp, Nicholasen
kaust.authorNarayanan, Kiranen
kaust.authorMora Cordova, Angelen
kaust.authorEl Sayed, Tamer S.en
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.