000890968 001__ 890968 000890968 005__ 20230111074306.0 000890968 0247_ $$2doi$$a10.1109/IGARSS39084.2020.9324237 000890968 0247_ $$2ISSN$$a0022-7722 000890968 0247_ $$2ISSN$$a1447-073X 000890968 0247_ $$2ISSN$$a1447-6959 000890968 0247_ $$2Handle$$a2128/27298 000890968 0247_ $$2WOS$$aWOS:000664335301039 000890968 037__ $$aFZJ-2021-01284 000890968 082__ $$a610 000890968 1001_ $$0P:(DE-Juel1)178695$$aSedona, Rocco$$b0$$eCorresponding author$$ufzj 000890968 1112_ $$a2020 IEEE International Geoscience and Remote Sensing Symposium$$cOnline event$$d2020-09-26 - 2020-10-02$$gIGARSS 2020$$wHawaii 000890968 245__ $$aScaling Up a Multispectral Resnet-50 to 128 GPUs 000890968 260__ $$bIEEE$$c2020 000890968 300__ $$a1058 - 1061 000890968 3367_ $$2ORCID$$aCONFERENCE_PAPER 000890968 3367_ $$033$$2EndNote$$aConference Paper 000890968 3367_ $$2BibTeX$$aINPROCEEDINGS 000890968 3367_ $$2DRIVER$$aconferenceObject 000890968 3367_ $$2DataCite$$aOutput Types/Conference Paper 000890968 3367_ $$0PUB:(DE-HGF)8$$2PUB:(DE-HGF)$$aContribution to a conference proceedings$$bcontrib$$mcontrib$$s1615191437_4749 000890968 520__ $$aSimilarly to other scientific domains, Deep Learning (DL)holds great promises to fulfil the challenging needs of RemoteSensing (RS) applications. However, the increase in volume,variety and complexity of acquisitions that are carried outon a daily basis by Earth Observation (EO) missions generatesnew processing and storage challenges within operationalprocessing pipelines. The aim of this work is to show thatHigh-Performance Computing (HPC) systems can speed upthe training time of Convolutional Neural Networks (CNNs).Particular attention is put on the monitoring of the classificationaccuracy that usually degrades when using large batchsizes. The experimental results of this work show that thetraining of the model scales up to a batch size of 8,000, obtainingclassification performances in terms of accuracy in linewith those using smaller batch sizes. 000890968 536__ $$0G:(DE-HGF)POF3-512$$a512 - Data-Intensive Science and Federated Computing (POF3-512)$$cPOF3-512$$fPOF III$$x0 000890968 536__ $$0G:(DE-Juel1)PHD-NO-GRANT-20170405$$aPhD no Grant - Doktorand ohne besondere Förderung (PHD-NO-GRANT-20170405)$$cPHD-NO-GRANT-20170405$$x1 000890968 588__ $$aDataset connected to CrossRef Conference 000890968 7001_ $$0P:(DE-Juel1)171343$$aCavallaro, Gabriele$$b1$$ufzj 000890968 7001_ $$0P:(DE-Juel1)158080$$aJitsev, Jenia$$b2$$ufzj 000890968 7001_ $$0P:(DE-Juel1)140202$$aStrube, Alexandre$$b3$$ufzj 000890968 7001_ $$0P:(DE-Juel1)132239$$aRiedel, Morris$$b4$$ufzj 000890968 7001_ $$0P:(DE-HGF)0$$aBook, Matthias$$b5 000890968 773__ $$a10.1109/IGARSS39084.2020.9324237 000890968 8564_ $$uhttps://juser.fz-juelich.de/record/890968/files/Sedona_Rocco_IGARSS_2020.pdf$$yOpenAccess 000890968 909CO $$ooai:juser.fz-juelich.de:890968$$pdnbdelivery$$pdriver$$pVDB$$popen_access$$popenaire 000890968 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)178695$$aForschungszentrum Jülich$$b0$$kFZJ 000890968 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)171343$$aForschungszentrum Jülich$$b1$$kFZJ 000890968 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)158080$$aForschungszentrum Jülich$$b2$$kFZJ 000890968 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)140202$$aForschungszentrum Jülich$$b3$$kFZJ 000890968 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)132239$$aForschungszentrum Jülich$$b4$$kFZJ 000890968 9130_ $$0G:(DE-HGF)POF3-512$$1G:(DE-HGF)POF3-510$$2G:(DE-HGF)POF3-500$$3G:(DE-HGF)POF3$$4G:(DE-HGF)POF$$aDE-HGF$$bKey Technologies$$lSupercomputing & Big Data$$vData-Intensive Science and Federated Computing$$x0 000890968 9131_ $$0G:(DE-HGF)POF4-511$$1G:(DE-HGF)POF4-510$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5112$$aDE-HGF$$bKey Technologies$$lEngineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action$$vEnabling Computational- & Data-Intensive Science and Engineering$$x0 000890968 9141_ $$y2021 000890968 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS 000890968 915__ $$0StatID:(DE-HGF)1030$$2StatID$$aDBCoverage$$bCurrent Contents - Life Sciences 000890968 915__ $$0StatID:(DE-HGF)0600$$2StatID$$aDBCoverage$$bEbsco Academic Search 000890968 915__ $$0StatID:(DE-HGF)1040$$2StatID$$aDBCoverage$$bZoological Record 000890968 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bANAT SCI INT : 2015 000890968 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection 000890968 915__ $$0StatID:(DE-HGF)0111$$2StatID$$aWoS$$bScience Citation Index Expanded 000890968 915__ $$0StatID:(DE-HGF)9900$$2StatID$$aIF < 5 000890968 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess 000890968 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bASC 000890968 915__ $$0StatID:(DE-HGF)0310$$2StatID$$aDBCoverage$$bNCBI Molecular Biology Database 000890968 915__ $$0StatID:(DE-HGF)1050$$2StatID$$aDBCoverage$$bBIOSIS Previews 000890968 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline 000890968 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bThomson Reuters Master Journal List 000890968 9201_ $$0I:(DE-Juel1)JSC-20090406$$kJSC$$lJülich Supercomputing Center$$x0 000890968 980__ $$acontrib 000890968 980__ $$aVDB 000890968 980__ $$aUNRESTRICTED 000890968 980__ $$aI:(DE-Juel1)JSC-20090406 000890968 9801_ $$aFullTexts