Scaling of a Fast Fourier Transform and a pseudo-spectral fluid solver up to 196608 cores

Handle URI:
http://hdl.handle.net/10754/626160
Title:
Scaling of a Fast Fourier Transform and a pseudo-spectral fluid solver up to 196608 cores
Authors:
Chatterjee, Anando G.; Verma, Mahendra K.; Kumar, Abhishek; Samtaney, Ravi ( 0000-0002-4702-6473 ) ; Hadri, Bilel; Khurram, Rooh Ul Amin
Abstract:
In this paper we present scaling results of a FFT library, FFTK, and a pseudospectral code, Tarang, on grid resolutions up to 81923 grid using 65536 cores of Blue Gene/P and 196608 cores of Cray XC40 supercomputers. We observe that communication dominates computation, more so on the Cray XC40. The computation time scales as Tcomp∼p−1, and the communication time as Tcomm∼n−γ2 with γ2 ranging from 0.7 to 0.9 for Blue Gene/P, and from 0.43 to 0.73 for Cray XC40. FFTK, and the fluid and convection solvers of Tarang exhibit weak as well as strong scaling nearly up to 196608 cores of Cray XC40. We perform a comparative study of the performance on the Blue Gene/P and Cray XC40 clusters.
KAUST Department:
Mechanical Engineering Program; Physical Sciences and Engineering (PSE) Division; KAUST Supercomputing Laboratory (KSL)
Citation:
Chatterjee AG, Verma MK, Kumar A, Samtaney R, Hadri B, et al. (2017) Scaling of a Fast Fourier Transform and a pseudo-spectral fluid solver up to 196608 cores. Journal of Parallel and Distributed Computing. Available: http://dx.doi.org/10.1016/j.jpdc.2017.10.014.
Publisher:
Elsevier BV
Journal:
Journal of Parallel and Distributed Computing
KAUST Grant Number:
K1052
Issue Date:
4-Nov-2017
DOI:
10.1016/j.jpdc.2017.10.014
Type:
Article
ISSN:
0743-7315
Sponsors:
We thank Shashwat Bhattacharya for his valuable help on plotting. Our numerical simulations were performed at Cray XC40 Shaheen II at KAUST supercomputing laboratory (KSL), Saudi Arabia. We thank KAUST scientists for the kind support while performing our simulations. This work was supported by research grants SERB/F/3279 from Science and Engineering Research Board India, K1052 from KAUST, and R&D/20130307 from IIT Kanpur; and by KAUST baseline research funds of Ravi Samtaney.
Additional Links:
http://www.sciencedirect.com/science/article/pii/S0743731517302903
Appears in Collections:
Articles; KAUST Supercomputing Laboratory (KSL); Physical Sciences and Engineering (PSE) Division; Mechanical Engineering Program

Full metadata record

DC FieldValue Language
dc.contributor.authorChatterjee, Anando G.en
dc.contributor.authorVerma, Mahendra K.en
dc.contributor.authorKumar, Abhisheken
dc.contributor.authorSamtaney, Ravien
dc.contributor.authorHadri, Bilelen
dc.contributor.authorKhurram, Rooh Ul Aminen
dc.date.accessioned2017-11-14T12:46:06Z-
dc.date.available2017-11-14T12:46:06Z-
dc.date.issued2017-11-04en
dc.identifier.citationChatterjee AG, Verma MK, Kumar A, Samtaney R, Hadri B, et al. (2017) Scaling of a Fast Fourier Transform and a pseudo-spectral fluid solver up to 196608 cores. Journal of Parallel and Distributed Computing. Available: http://dx.doi.org/10.1016/j.jpdc.2017.10.014.en
dc.identifier.issn0743-7315en
dc.identifier.doi10.1016/j.jpdc.2017.10.014en
dc.identifier.urihttp://hdl.handle.net/10754/626160-
dc.description.abstractIn this paper we present scaling results of a FFT library, FFTK, and a pseudospectral code, Tarang, on grid resolutions up to 81923 grid using 65536 cores of Blue Gene/P and 196608 cores of Cray XC40 supercomputers. We observe that communication dominates computation, more so on the Cray XC40. The computation time scales as Tcomp∼p−1, and the communication time as Tcomm∼n−γ2 with γ2 ranging from 0.7 to 0.9 for Blue Gene/P, and from 0.43 to 0.73 for Cray XC40. FFTK, and the fluid and convection solvers of Tarang exhibit weak as well as strong scaling nearly up to 196608 cores of Cray XC40. We perform a comparative study of the performance on the Blue Gene/P and Cray XC40 clusters.en
dc.description.sponsorshipWe thank Shashwat Bhattacharya for his valuable help on plotting. Our numerical simulations were performed at Cray XC40 Shaheen II at KAUST supercomputing laboratory (KSL), Saudi Arabia. We thank KAUST scientists for the kind support while performing our simulations. This work was supported by research grants SERB/F/3279 from Science and Engineering Research Board India, K1052 from KAUST, and R&D/20130307 from IIT Kanpur; and by KAUST baseline research funds of Ravi Samtaney.en
dc.publisherElsevier BVen
dc.relation.urlhttp://www.sciencedirect.com/science/article/pii/S0743731517302903en
dc.rightsNOTICE: this is the author’s version of a work that was accepted for publication in Journal of Parallel and Distributed Computing. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Journal of Parallel and Distributed Computing, 3 November 2017. DOI: 10.1016/j.jpdc.2017.10.014. © 2017. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/en
dc.subjectExtreme-resolution turbulence simulationen
dc.subjectPseudospectral methoden
dc.subjectFast fourier transformen
dc.subjectTurbulence simulationen
dc.titleScaling of a Fast Fourier Transform and a pseudo-spectral fluid solver up to 196608 coresen
dc.typeArticleen
dc.contributor.departmentMechanical Engineering Programen
dc.contributor.departmentPhysical Sciences and Engineering (PSE) Divisionen
dc.contributor.departmentKAUST Supercomputing Laboratory (KSL)en
dc.identifier.journalJournal of Parallel and Distributed Computingen
dc.eprint.versionPost-printen
dc.contributor.institutionDepartment of Physics, Indian Institute of Technology Kanpur, Kanpur 208016, Indiaen
kaust.authorSamtaney, Ravien
kaust.authorHadri, Bilelen
kaust.authorKhurram, Rooh Ul Aminen
kaust.grant.numberK1052en
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.