Bohr: Similarity aware geo-distributed data analytics

Hangyu Li; Hong Xu; Sarana Nutanong

doi:10.1145/3281411.3281418

Bohr: Similarity aware geo-distributed data analytics

Department of Computer Science

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

3 Citations (Scopus)

Abstract

We propose Bohr, a similarity aware geo-distributed data analytics system that minimizes query completion time. The key idea is to exploit similarity between data in different data centers (DCs), and transfer similar data from the bottleneck DC to other sites with more WAN bandwidth. Though these sites have more input data to process, these data are more similar and can be more efficiently aggregated by the combiner to reduce the intermediate data that needs to be shuffled across the WAN. Thus our similarity aware approach reduces the shuffle time and in turn the query completion time (QCT).

We design Bohr based on OLAP data cubes to perform efficient similarity checking among datasets in different sites. We implement Bohr on Spark and deploy it across ten sites of AWS EC2. Our extensive evaluation using realistic query workloads shows that Bohr improves the QCT by up to 50% and reduces the intermediate data by up to 6x compared to state-of-the-art solutions that also use OLAP cubes.

Original language	English
Title of host publication	CoNEXT 2018 - Proceedings of the 14th International Conference on Emerging Networking EXperiments and Technologies
Publisher	Association for Computing Machinery
Pages	267-279
ISBN (Print)	978-1-4503-6080-7
DOIs	https://doi.org/10.1145/3281411.3281418
Publication status	Published - Dec 2018
Event	14th International Conference on Emerging Networking EXperiments and Technologies, CoNEXT 2018 - Heraklion, Greece Duration: 4 Dec 2018 → 7 Dec 2018

Publication series

Name	CoNEXT - Proceedings of the International Conference on Emerging Networking EXperiments and Technologies

Conference

Conference	14th International Conference on Emerging Networking EXperiments and Technologies, CoNEXT 2018
Place	Greece
City	Heraklion
Period	4/12/18 → 7/12/18

Bibliographical note

Full text of this publication does not contain sufficient affiliation information. With consent from the author(s) concerned, the Research Unit(s) information for this record is based on the existing academic department affiliation of the author(s).

Research Keywords

Cloud computing
Data analytics
WAN

Access Output

10.1145/3281411.3281418

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

Li, H., Xu, H., & Nutanong, S. (2018). Bohr: Similarity aware geo-distributed data analytics. In CoNEXT 2018 - Proceedings of the 14th International Conference on Emerging Networking EXperiments and Technologies (pp. 267-279). (CoNEXT - Proceedings of the International Conference on Emerging Networking EXperiments and Technologies). Association for Computing Machinery. https://doi.org/10.1145/3281411.3281418

Li, Hangyu ; Xu, Hong ; Nutanong, Sarana. / Bohr : Similarity aware geo-distributed data analytics. CoNEXT 2018 - Proceedings of the 14th International Conference on Emerging Networking EXperiments and Technologies. Association for Computing Machinery, 2018. pp. 267-279 (CoNEXT - Proceedings of the International Conference on Emerging Networking EXperiments and Technologies).

@inproceedings{59588d37647c4f9398d977121eb51103,

title = "Bohr: Similarity aware geo-distributed data analytics",

abstract = "We propose Bohr, a similarity aware geo-distributed data analytics system that minimizes query completion time. The key idea is to exploit similarity between data in different data centers (DCs), and transfer similar data from the bottleneck DC to other sites with more WAN bandwidth. Though these sites have more input data to process, these data are more similar and can be more efficiently aggregated by the combiner to reduce the intermediate data that needs to be shuffled across the WAN. Thus our similarity aware approach reduces the shuffle time and in turn the query completion time (QCT). We design Bohr based on OLAP data cubes to perform efficient similarity checking among datasets in different sites. We implement Bohr on Spark and deploy it across ten sites of AWS EC2. Our extensive evaluation using realistic query workloads shows that Bohr improves the QCT by up to 50% and reduces the intermediate data by up to 6x compared to state-of-the-art solutions that also use OLAP cubes.",

keywords = "Cloud computing, Data analytics, WAN",

author = "Hangyu Li and Hong Xu and Sarana Nutanong",

note = "Full text of this publication does not contain sufficient affiliation information. With consent from the author(s) concerned, the Research Unit(s) information for this record is based on the existing academic department affiliation of the author(s).; 14th International Conference on Emerging Networking EXperiments and Technologies, CoNEXT 2018 ; Conference date: 04-12-2018 Through 07-12-2018",

year = "2018",

month = dec,

doi = "10.1145/3281411.3281418",

language = "English",

isbn = "978-1-4503-6080-7",

series = "CoNEXT - Proceedings of the International Conference on Emerging Networking EXperiments and Technologies",

publisher = "Association for Computing Machinery",

pages = "267--279",

booktitle = "CoNEXT 2018 - Proceedings of the 14th International Conference on Emerging Networking EXperiments and Technologies",

address = "United States",

}

Li, H , Xu, H & Nutanong, S 2018, Bohr: Similarity aware geo-distributed data analytics. in CoNEXT 2018 - Proceedings of the 14th International Conference on Emerging Networking EXperiments and Technologies. CoNEXT - Proceedings of the International Conference on Emerging Networking EXperiments and Technologies, Association for Computing Machinery, pp. 267-279, 14th International Conference on Emerging Networking EXperiments and Technologies, CoNEXT 2018, Heraklion, Greece, 4/12/18. https://doi.org/10.1145/3281411.3281418

Bohr: Similarity aware geo-distributed data analytics. / Li, Hangyu ; Xu, Hong ; Nutanong, Sarana.
CoNEXT 2018 - Proceedings of the 14th International Conference on Emerging Networking EXperiments and Technologies. Association for Computing Machinery, 2018. p. 267-279 (CoNEXT - Proceedings of the International Conference on Emerging Networking EXperiments and Technologies).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Bohr

T2 - 14th International Conference on Emerging Networking EXperiments and Technologies, CoNEXT 2018

AU - Li, Hangyu

AU - Xu, Hong

AU - Nutanong, Sarana

N1 - Full text of this publication does not contain sufficient affiliation information. With consent from the author(s) concerned, the Research Unit(s) information for this record is based on the existing academic department affiliation of the author(s).

PY - 2018/12

Y1 - 2018/12

N2 - We propose Bohr, a similarity aware geo-distributed data analytics system that minimizes query completion time. The key idea is to exploit similarity between data in different data centers (DCs), and transfer similar data from the bottleneck DC to other sites with more WAN bandwidth. Though these sites have more input data to process, these data are more similar and can be more efficiently aggregated by the combiner to reduce the intermediate data that needs to be shuffled across the WAN. Thus our similarity aware approach reduces the shuffle time and in turn the query completion time (QCT). We design Bohr based on OLAP data cubes to perform efficient similarity checking among datasets in different sites. We implement Bohr on Spark and deploy it across ten sites of AWS EC2. Our extensive evaluation using realistic query workloads shows that Bohr improves the QCT by up to 50% and reduces the intermediate data by up to 6x compared to state-of-the-art solutions that also use OLAP cubes.

AB - We propose Bohr, a similarity aware geo-distributed data analytics system that minimizes query completion time. The key idea is to exploit similarity between data in different data centers (DCs), and transfer similar data from the bottleneck DC to other sites with more WAN bandwidth. Though these sites have more input data to process, these data are more similar and can be more efficiently aggregated by the combiner to reduce the intermediate data that needs to be shuffled across the WAN. Thus our similarity aware approach reduces the shuffle time and in turn the query completion time (QCT). We design Bohr based on OLAP data cubes to perform efficient similarity checking among datasets in different sites. We implement Bohr on Spark and deploy it across ten sites of AWS EC2. Our extensive evaluation using realistic query workloads shows that Bohr improves the QCT by up to 50% and reduces the intermediate data by up to 6x compared to state-of-the-art solutions that also use OLAP cubes.

KW - Cloud computing

KW - Data analytics

KW - WAN

UR - http://www.scopus.com/inward/record.url?scp=85060393770&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85060393770&origin=recordpage

U2 - 10.1145/3281411.3281418

DO - 10.1145/3281411.3281418

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 978-1-4503-6080-7

T3 - CoNEXT - Proceedings of the International Conference on Emerging Networking EXperiments and Technologies

SP - 267

EP - 279

BT - CoNEXT 2018 - Proceedings of the 14th International Conference on Emerging Networking EXperiments and Technologies

PB - Association for Computing Machinery

Y2 - 4 December 2018 through 7 December 2018

ER -

Li H , Xu H , Nutanong S. Bohr: Similarity aware geo-distributed data analytics. In CoNEXT 2018 - Proceedings of the 14th International Conference on Emerging Networking EXperiments and Technologies. Association for Computing Machinery. 2018. p. 267-279. (CoNEXT - Proceedings of the International Conference on Emerging Networking EXperiments and Technologies). doi: 10.1145/3281411.3281418

Bohr: Similarity aware geo-distributed data analytics

Abstract

Publication series

Conference

Bibliographical note

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this