Bds+: An inter-datacenter data replication system with dynamic bandwidth separation

Yuchao Zhang; Xiaohui Nie; Junchen Jiang; Wendong Wang; Ke Xu; Youjian Zhao; Martin J. Reed; Kai Chen; Haiyang Wang; Guang Yao

doi:10.1109/TNET.2021.3054924

Bds+: An inter-datacenter data replication system with dynamic bandwidth separation

Yuchao Zhang, Xiaohui Nie, Junchen Jiang, Wendong Wang, Ke Xu, Youjian Zhao, Martin J. Reed, Kai Chen, Haiyang Wang, Guang Yao

Computer Science (Duluth)

Research output: Contribution to journal › Article › peer-review

6 Scopus citations

Abstract

Many important cloud services require replicating massive data from one datacenter (DC) to multiple DCs. While the performance of pair-wise inter-DC data transfers has been much improved, prior solutions are insufficient to optimize bulk-data multicast, as they fail to explore the rich inter-DC overlay paths that exist in geo-distributed DCs, as well as the remaining bandwidth reserved for online traffic under fixed bandwidth separation scheme. To take advantage of these opportunities, we present BDS+, a near-optimal network system for large-scale inter-DC data replication. BDS+ is an application-level multicast overlay network with a fully centralized architecture, allowing a central controller to maintain an up-to-date global view of data delivery status of intermediate servers, in order to fully utilize the available overlay paths. Furthermore, in each overlay path, it leverages dynamic bandwidth separation to make use of the remaining available bandwidth reserved for online traffic. By constantly estimating online traffic demand and rescheduling bulk-data transfers accordingly, BDS+ can further speed up the massive data multicast. Through a pilot deployment in one of the largest online service providers and large-scale real-trace simulations, we show that BDS+ can achieve 3- 5\times speedup over the provider's existing system and several well-known overlay routing baselines of static bandwidth separation. Moreover, dynamic bandwidth separation can further reduce the completion time of bulk data transfers by 1.2 to 1.3 times.

Original language	English (US)
Article number	9352539
Pages (from-to)	918-934
Number of pages	17
Journal	IEEE/ACM Transactions on Networking
Volume	29
Issue number	2
DOIs	https://doi.org/10.1109/TNET.2021.3054924
State	Published - Apr 2021

Bibliographical note

Funding Information:
Manuscript received March 26, 2019; revised January 7, 2020; accepted January 12, 2021; approved by IEEE/ACM TRANSACTIONS ON NETWORK-ING Editor N. Hegde. Date of publication February 10, 2021; date of current version April 16, 2021. The work of Yuchao Zhang was supported in part by the National Key Research and Development Program of China under Grant 2019YFB1802603, in part by the National Natural Science Foundation of China (NSFC) Youth Science Foundation under Grant 61802024, and in part by the Fundamental Research Funds for the Central Universities under Grant 2482020RC36. The work of Ke Xu was supported in part by the Science and Technology Innovation Project under Grant 2020KJ010501, in part by the China National Funds for Distinguished Young Scientists under Grant 61825204, and in part by the Beijing Outstanding Young Scientist Program under Grant BJJWZYJH01201910003011. (Yuchao Zhang and Xiaohui Nie contributed equally to this work.) (Corresponding author: Yuchao Zhang.) Yuchao Zhang is with the School of Computer Science (National Pilot Software Engineering School), Beijing University of Posts and Telecommunications, Beijing 100876, China (e-mail: yczhang@bupt.edu.cn).

Publisher Copyright:
© 1993-2012 IEEE.

Keywords

Centralized control
Data replication
Dynamic bandwidth separation
Overlay network

Publisher link

10.1109/TNET.2021.3054924

Find It

Find It at U of M Libraries

Cite this

@article{bb6ee817f2424875b2dcfde22b644d71,

title = "Bds+: An inter-datacenter data replication system with dynamic bandwidth separation",

abstract = "Many important cloud services require replicating massive data from one datacenter (DC) to multiple DCs. While the performance of pair-wise inter-DC data transfers has been much improved, prior solutions are insufficient to optimize bulk-data multicast, as they fail to explore the rich inter-DC overlay paths that exist in geo-distributed DCs, as well as the remaining bandwidth reserved for online traffic under fixed bandwidth separation scheme. To take advantage of these opportunities, we present BDS+, a near-optimal network system for large-scale inter-DC data replication. BDS+ is an application-level multicast overlay network with a fully centralized architecture, allowing a central controller to maintain an up-to-date global view of data delivery status of intermediate servers, in order to fully utilize the available overlay paths. Furthermore, in each overlay path, it leverages dynamic bandwidth separation to make use of the remaining available bandwidth reserved for online traffic. By constantly estimating online traffic demand and rescheduling bulk-data transfers accordingly, BDS+ can further speed up the massive data multicast. Through a pilot deployment in one of the largest online service providers and large-scale real-trace simulations, we show that BDS+ can achieve 3- 5\times speedup over the provider's existing system and several well-known overlay routing baselines of static bandwidth separation. Moreover, dynamic bandwidth separation can further reduce the completion time of bulk data transfers by 1.2 to 1.3 times.",

keywords = "Centralized control, Data replication, Dynamic bandwidth separation, Overlay network",

author = "Yuchao Zhang and Xiaohui Nie and Junchen Jiang and Wendong Wang and Ke Xu and Youjian Zhao and Reed, {Martin J.} and Kai Chen and Haiyang Wang and Guang Yao",

note = "Funding Information: Manuscript received March 26, 2019; revised January 7, 2020; accepted January 12, 2021; approved by IEEE/ACM TRANSACTIONS ON NETWORK-ING Editor N. Hegde. Date of publication February 10, 2021; date of current version April 16, 2021. The work of Yuchao Zhang was supported in part by the National Key Research and Development Program of China under Grant 2019YFB1802603, in part by the National Natural Science Foundation of China (NSFC) Youth Science Foundation under Grant 61802024, and in part by the Fundamental Research Funds for the Central Universities under Grant 2482020RC36. The work of Ke Xu was supported in part by the Science and Technology Innovation Project under Grant 2020KJ010501, in part by the China National Funds for Distinguished Young Scientists under Grant 61825204, and in part by the Beijing Outstanding Young Scientist Program under Grant BJJWZYJH01201910003011. (Yuchao Zhang and Xiaohui Nie contributed equally to this work.) (Corresponding author: Yuchao Zhang.) Yuchao Zhang is with the School of Computer Science (National Pilot Software Engineering School), Beijing University of Posts and Telecommunications, Beijing 100876, China (e-mail: yczhang@bupt.edu.cn). Publisher Copyright: {\textcopyright} 1993-2012 IEEE.",

year = "2021",

month = apr,

doi = "10.1109/TNET.2021.3054924",

language = "English (US)",

volume = "29",

pages = "918--934",

journal = "IEEE/ACM Transactions on Networking",

issn = "1063-6692",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "2",

}

TY - JOUR

T1 - Bds+

T2 - An inter-datacenter data replication system with dynamic bandwidth separation

AU - Zhang, Yuchao

AU - Nie, Xiaohui

AU - Jiang, Junchen

AU - Wang, Wendong

AU - Xu, Ke

AU - Zhao, Youjian

AU - Reed, Martin J.

AU - Chen, Kai

AU - Wang, Haiyang

AU - Yao, Guang

N1 - Funding Information: Manuscript received March 26, 2019; revised January 7, 2020; accepted January 12, 2021; approved by IEEE/ACM TRANSACTIONS ON NETWORK-ING Editor N. Hegde. Date of publication February 10, 2021; date of current version April 16, 2021. The work of Yuchao Zhang was supported in part by the National Key Research and Development Program of China under Grant 2019YFB1802603, in part by the National Natural Science Foundation of China (NSFC) Youth Science Foundation under Grant 61802024, and in part by the Fundamental Research Funds for the Central Universities under Grant 2482020RC36. The work of Ke Xu was supported in part by the Science and Technology Innovation Project under Grant 2020KJ010501, in part by the China National Funds for Distinguished Young Scientists under Grant 61825204, and in part by the Beijing Outstanding Young Scientist Program under Grant BJJWZYJH01201910003011. (Yuchao Zhang and Xiaohui Nie contributed equally to this work.) (Corresponding author: Yuchao Zhang.) Yuchao Zhang is with the School of Computer Science (National Pilot Software Engineering School), Beijing University of Posts and Telecommunications, Beijing 100876, China (e-mail: yczhang@bupt.edu.cn). Publisher Copyright: © 1993-2012 IEEE.

PY - 2021/4

Y1 - 2021/4

N2 - Many important cloud services require replicating massive data from one datacenter (DC) to multiple DCs. While the performance of pair-wise inter-DC data transfers has been much improved, prior solutions are insufficient to optimize bulk-data multicast, as they fail to explore the rich inter-DC overlay paths that exist in geo-distributed DCs, as well as the remaining bandwidth reserved for online traffic under fixed bandwidth separation scheme. To take advantage of these opportunities, we present BDS+, a near-optimal network system for large-scale inter-DC data replication. BDS+ is an application-level multicast overlay network with a fully centralized architecture, allowing a central controller to maintain an up-to-date global view of data delivery status of intermediate servers, in order to fully utilize the available overlay paths. Furthermore, in each overlay path, it leverages dynamic bandwidth separation to make use of the remaining available bandwidth reserved for online traffic. By constantly estimating online traffic demand and rescheduling bulk-data transfers accordingly, BDS+ can further speed up the massive data multicast. Through a pilot deployment in one of the largest online service providers and large-scale real-trace simulations, we show that BDS+ can achieve 3- 5\times speedup over the provider's existing system and several well-known overlay routing baselines of static bandwidth separation. Moreover, dynamic bandwidth separation can further reduce the completion time of bulk data transfers by 1.2 to 1.3 times.

AB - Many important cloud services require replicating massive data from one datacenter (DC) to multiple DCs. While the performance of pair-wise inter-DC data transfers has been much improved, prior solutions are insufficient to optimize bulk-data multicast, as they fail to explore the rich inter-DC overlay paths that exist in geo-distributed DCs, as well as the remaining bandwidth reserved for online traffic under fixed bandwidth separation scheme. To take advantage of these opportunities, we present BDS+, a near-optimal network system for large-scale inter-DC data replication. BDS+ is an application-level multicast overlay network with a fully centralized architecture, allowing a central controller to maintain an up-to-date global view of data delivery status of intermediate servers, in order to fully utilize the available overlay paths. Furthermore, in each overlay path, it leverages dynamic bandwidth separation to make use of the remaining available bandwidth reserved for online traffic. By constantly estimating online traffic demand and rescheduling bulk-data transfers accordingly, BDS+ can further speed up the massive data multicast. Through a pilot deployment in one of the largest online service providers and large-scale real-trace simulations, we show that BDS+ can achieve 3- 5\times speedup over the provider's existing system and several well-known overlay routing baselines of static bandwidth separation. Moreover, dynamic bandwidth separation can further reduce the completion time of bulk data transfers by 1.2 to 1.3 times.

KW - Centralized control

KW - Data replication

KW - Dynamic bandwidth separation

KW - Overlay network

UR - http://www.scopus.com/inward/record.url?scp=85100862188&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85100862188&partnerID=8YFLogxK

U2 - 10.1109/TNET.2021.3054924

DO - 10.1109/TNET.2021.3054924

M3 - Article

AN - SCOPUS:85100862188

SN - 1063-6692

VL - 29

SP - 918

EP - 934

JO - IEEE/ACM Transactions on Networking

JF - IEEE/ACM Transactions on Networking

IS - 2

M1 - 9352539

ER -

Bds+: An inter-datacenter data replication system with dynamic bandwidth separation

Abstract

Bibliographical note

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this