KIT | KIT-Bibliothek | Impressum | Datenschutz

Remote Sensing Data Analytics with the Udocker Container Tool using Multi-GPU Deep Learning Systems

Cavallaro, Gabriele; Kozlov, Valentin; Götz, Markus; Riedel, Morris

Multi-GPU systems are in continuous development to deal with the challenges of intensive computational big data
problems. On the one hand, parallel architectures provide a tremendous computation capacity and outstanding scalability.
On the other hand, the production path in multi-user environments faces several roadblocks since they do not grant root
privileges to the users. Containers provide flexible strategies for packing, deploying and running isolated application
processes within multi-user systems and enable scientific reproducibility. This paper describes the usage and advantages
that the uDocker container tool offers for the development of deep learning models in the described context. The experimental results show that uDocker is more transparent to deploy for less tech-savvy researchers and allows the application to achieve processing time with negligible overhead compared to an uncontainerized environment.

Open Access Logo

Postprint §
DOI: 10.5445/IR/1000093125
Frei zugänglich ab 01.01.2021
Zugehörige Institution(en) am KIT Steinbuch Centre for Computing (SCC)
Publikationstyp Proceedingsbeitrag
Publikationsjahr 2019
Sprache Englisch
Identifikator ISBN: 978-92-76-00034-1
KITopen-ID: 1000093125
HGF-Programm 46.12.02 (POF III, LK 01)
Data Activities
Erschienen in Proceedings of 2019 Big Data from Space (BiDS'19). Ed.: S. Pierre
Veranstaltung Big Data from Space (BiDS 2019), München, Deutschland, 19.02.2019 – 21.02.2019
Seiten 177-180
Projektinformation DEEP-HybridDataCloud (EU, H2020, 777435)
Schlagwörter Containers, uDocker, multi-GPU, deep learning, classification, remote sensing
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page