Show simple item record

dc.contributor.authorFekry, Aen
dc.contributor.authorCarata, Lucianen
dc.contributor.authorPasquier, Ten
dc.contributor.authorRice, Andrewen
dc.date.accessioned2020-11-11T00:30:34Z
dc.date.available2020-11-11T00:30:34Z
dc.date.issued2020-12-10en
dc.identifier.isbn9781728162515en
dc.identifier.issn2639-1589
dc.identifier.urihttps://www.repository.cam.ac.uk/handle/1810/312750
dc.description.abstractOne of the key challenges for data analytics deployment is configuration tuning. The existing approaches for configuration tuning are expensive and overlook the dynamic characteristics of the analytics environment (i.e. frequent changes in workload due to receiving evolving input sizes or change in the underlying cluster environment). Such workload/environment changes can cause significant performance degradation, with retuning the configuration to accommodate those changes can yield up to 85\% potential execution time saving. We propose SimTune, an approach that accommodates such changes through efficient configuration tuning. SimTune combines workload characterization and Multitask Bayesian optimization to identify similarity across workloads and accelerate finding near-optimal configurations. Our experimental results show that SimTune reduces the search time for finding close-to-optimal configurations by 56-73\% (at the median) when compared to existing state-of-the-art techniques. This means that the amortization of the tuning cost happens significantly faster, enabling practical tuning in the rapidly changing environment of distributed analytics.
dc.description.sponsorshipGoogle Cloud, Amazon AWS
dc.rightsAll rights reserved
dc.titleAccelerating the Configuration Tuning of Big Data Analytics with Similarity-aware Multitask Bayesian Optimizationen
dc.typeConference Object
prism.endingPage275
prism.publicationDate2020en
prism.publicationNameProceedings - 2020 IEEE International Conference on Big Data, Big Data 2020en
prism.startingPage266
dc.identifier.doi10.17863/CAM.59851
dcterms.dateAccepted2020-10-20en
rioxxterms.versionofrecord10.1109/BigData50022.2020.9378085en
rioxxterms.versionAM
rioxxterms.licenseref.urihttp://www.rioxx.net/licenses/all-rights-reserveden
rioxxterms.licenseref.startdate2020-12-10en
dc.contributor.orcidRice, Andrew [0000-0002-4677-8032]
rioxxterms.typeConference Paper/Proceeding/Abstracten
cam.orpheus.successMon Apr 26 07:33:16 BST 2021 - Embargo updated*
cam.orpheus.counter24*
rioxxterms.freetoread.startdate2021-12-10


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record