Please use this identifier to cite or link to this item: https://hdl.handle.net/20.500.14279/19364
DC FieldValueLanguage
dc.contributor.authorHerodotou, Herodotos-
dc.contributor.authorKakoulli, Elena-
dc.date.accessioned2020-11-11T12:45:05Z-
dc.date.available2020-11-11T12:45:05Z-
dc.date.issued2020-
dc.identifier.citationProceedings of the VLDB Endowment, 2020, vol. 13, no. 1, pp. 43-56en_US
dc.identifier.issn21508097-
dc.identifier.urihttps://hdl.handle.net/20.500.14279/19364-
dc.descriptionPresented at 46th International Conference on Very Large Data Bases, 31 August - 4 September 2020, Japanen_US
dc.description.abstractData-intensive platforms such as Hadoop and Spark are routinely used to process massive amounts of data residing on distributed le systems like HDFS. Increasing memory sizes and new hardware technologies (e.g., NVRAM, SSDs) have recently led to the introduction of storage tiering in such settings. However, users are now burdened with the additional complexity of managing the multiple storage tiers and the data residing on them while trying to optimize their workloads. In this paper, we develop a general framework for automatically moving data across the available storage tiers in distributed le systems. Moreover, we employ machine learning for tracking and predicting le access patterns, which we use to decide when and which data to move up or down the storage tiers for increasing system performance. Our approach uses incremental learning to dynamically rene the models with new le accesses, allowing them to naturally adjust and adapt to workload changes over time. Our extensive evaluation using realistic workloads derived from Facebook and CMU traces compares our approach with several other policies and showcases signicant bene ts in terms of both workload performance and cluster effciency.en_US
dc.formatpdfen_US
dc.language.isoenen_US
dc.relation.ispartofProceedings of the VLDB Endowmenten_US
dc.rightsThis work is licensed under the Creative Commons AttributionNonCommercial-NoDerivatives 4.0 International License.en_US
dc.rightsAttribution-NonCommercial-NoDerivatives 4.0 International*
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/*
dc.subjectHadoopen_US
dc.subjectDistributed File Systemen_US
dc.subjectMapreduceen_US
dc.titleAutomating distributed tiered storage management in cluster computingen_US
dc.typeArticleen_US
dc.collaborationCyprus University of Technologyen_US
dc.subject.categoryComputer and Information Sciencesen_US
dc.journalsOpen Accessen_US
dc.countryCyprusen_US
dc.subject.fieldNatural Sciencesen_US
dc.publicationPeer Revieweden_US
dc.identifier.doi10.14778/3357377.3357381en_US
dc.relation.issue1en_US
dc.relation.volume13en_US
cut.common.academicyear2019-2020en_US
dc.identifier.spage43en_US
dc.identifier.epage56en_US
item.grantfulltextopen-
item.openairecristypehttp://purl.org/coar/resource_type/c_6501-
item.fulltextWith Fulltext-
item.languageiso639-1en-
item.cerifentitytypePublications-
item.openairetypearticle-
crisitem.journal.journalissn2150-8097-
crisitem.journal.publisherACM-
crisitem.author.deptDepartment of Electrical Engineering, Computer Engineering and Informatics-
crisitem.author.deptDepartment of Electrical Engineering, Computer Engineering and Informatics-
crisitem.author.facultyFaculty of Engineering and Technology-
crisitem.author.facultyFaculty of Engineering and Technology-
crisitem.author.orcid0000-0002-8717-1691-
crisitem.author.orcid0000-0003-1489-807X-
crisitem.author.parentorgFaculty of Engineering and Technology-
crisitem.author.parentorgFaculty of Engineering and Technology-
Appears in Collections:Άρθρα/Articles
Files in This Item:
File Description SizeFormat
3357377.3357381.pdfFulltext974.4 kBAdobe PDFView/Open
CORE Recommender
Show simple item record

SCOPUSTM   
Citations

14
checked on Nov 6, 2023

WEB OF SCIENCETM
Citations

10
Last Week
1
Last month
1
checked on Oct 29, 2023

Page view(s) 50

318
Last Week
0
Last month
2
checked on Dec 22, 2024

Download(s)

85
checked on Dec 22, 2024

Google ScholarTM

Check

Altmetric


This item is licensed under a Creative Commons License Creative Commons