Object storage for deep learning frameworks

Or Ozeri, Effi Ofer, Ronen Kat

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

5 Scopus citations

Abstract

The advent of big datasets and high speed GPUs is fueling the growth in machine and deep learning techniques. In this paper we explore storing the training data in object storage and demonstrate how this can be done effectively while providing sufficient throughput to high performance GPUs.

Original languageEnglish
Title of host publicationDIDL 2018 - Proceedings of the 2nd Workshop on Distributed Infrastructures for Deep Learning, Part of Middleware 2018
PublisherAssociation for Computing Machinery, Inc
Pages21-24
Number of pages4
ISBN (Electronic)9781450361194
DOIs
StatePublished - 10 Dec 2018
Externally publishedYes
Event1st Workshop on Distributed Infrastructures for Deep Learning, DIDL 2018, co-located with the International Middleware Conference 2018 - Rennes, France
Duration: 10 Dec 2018 → …

Publication series

NameDIDL 2018 - Proceedings of the 2nd Workshop on Distributed Infrastructures for Deep Learning, Part of Middleware 2018

Conference

Conference1st Workshop on Distributed Infrastructures for Deep Learning, DIDL 2018, co-located with the International Middleware Conference 2018
Country/TerritoryFrance
CityRennes
Period10/12/18 → …

Keywords

  • Deep learning
  • Machine learning
  • Object storage

ASJC Scopus subject areas

  • Information Systems
  • Software

Fingerprint

Dive into the research topics of 'Object storage for deep learning frameworks'. Together they form a unique fingerprint.

Cite this