Low-latency multi-datacenter databases using replicated commit

Hatem Mahmoud, Faisal Nawab, Alexander Pucher, Divyakant Agrawal, Amr El Abbadi

Research output: Chapter in Book/Report/Conference proceedingChapter

35 Citations (Scopus)

Abstract

Web service providers have been using NoSQL datastores to provide scalability and availability for globally distributed data at the cost of sacrificing transactional guarantees. Recently, major web service providers like Google have moved towards building storage systems that provide ACID transactional guarantees for globally distributed data. For example, the newly published system, Spanner, uses Two-Phase Commit and Two-Phase Locking to provide atomicity and isolation for globally distributed data, running on top of Paxos to provide fault-tolerant log replication. We show in this paper that it is possible to provide the same ACID transactional guarantees for multi-datacenter databases with fewer crossdatacenter communication trips, compared to replicated logging. Instead of replicating the transactional log, we replicate the commit operation itself, by running Two-Phase Commit multiple times in different datacenters and using Paxos to reach consensus among datacenters as to whether the transaction should commit. Doing so not only replaces several inter-datacenter communication trips with intra-datacenter communication trips, but also allows us to integrate atomic commitment and isolation protocols with consistent replication protocols to further reduce the number of cross-datacenter communication trips needed for consistent replication; for example, by eliminating the need for an election phase in Paxos. We analyze our approach in terms of communication trips to compare it against the log replication approach, then we conduct an extensive experimental study to compare the performance and scalability of both approaches under various multi-datacenter setups.

Original languageEnglish
Title of host publicationProceedings of the VLDB Endowment
Pages661-672
Number of pages12
Volume6
Edition9
Publication statusPublished - 2013
Externally publishedYes

Fingerprint

Communication
Web services
Scalability
Network protocols
Availability

ASJC Scopus subject areas

  • Computer Science (miscellaneous)
  • Computer Science(all)

Cite this

Mahmoud, H., Nawab, F., Pucher, A., Agrawal, D., & El Abbadi, A. (2013). Low-latency multi-datacenter databases using replicated commit. In Proceedings of the VLDB Endowment (9 ed., Vol. 6, pp. 661-672)

Low-latency multi-datacenter databases using replicated commit. / Mahmoud, Hatem; Nawab, Faisal; Pucher, Alexander; Agrawal, Divyakant; El Abbadi, Amr.

Proceedings of the VLDB Endowment. Vol. 6 9. ed. 2013. p. 661-672.

Research output: Chapter in Book/Report/Conference proceedingChapter

Mahmoud, H, Nawab, F, Pucher, A, Agrawal, D & El Abbadi, A 2013, Low-latency multi-datacenter databases using replicated commit. in Proceedings of the VLDB Endowment. 9 edn, vol. 6, pp. 661-672.
Mahmoud H, Nawab F, Pucher A, Agrawal D, El Abbadi A. Low-latency multi-datacenter databases using replicated commit. In Proceedings of the VLDB Endowment. 9 ed. Vol. 6. 2013. p. 661-672
Mahmoud, Hatem ; Nawab, Faisal ; Pucher, Alexander ; Agrawal, Divyakant ; El Abbadi, Amr. / Low-latency multi-datacenter databases using replicated commit. Proceedings of the VLDB Endowment. Vol. 6 9. ed. 2013. pp. 661-672
@inbook{fefd57d6024942e188c44f4e17c7660e,
title = "Low-latency multi-datacenter databases using replicated commit",
abstract = "Web service providers have been using NoSQL datastores to provide scalability and availability for globally distributed data at the cost of sacrificing transactional guarantees. Recently, major web service providers like Google have moved towards building storage systems that provide ACID transactional guarantees for globally distributed data. For example, the newly published system, Spanner, uses Two-Phase Commit and Two-Phase Locking to provide atomicity and isolation for globally distributed data, running on top of Paxos to provide fault-tolerant log replication. We show in this paper that it is possible to provide the same ACID transactional guarantees for multi-datacenter databases with fewer crossdatacenter communication trips, compared to replicated logging. Instead of replicating the transactional log, we replicate the commit operation itself, by running Two-Phase Commit multiple times in different datacenters and using Paxos to reach consensus among datacenters as to whether the transaction should commit. Doing so not only replaces several inter-datacenter communication trips with intra-datacenter communication trips, but also allows us to integrate atomic commitment and isolation protocols with consistent replication protocols to further reduce the number of cross-datacenter communication trips needed for consistent replication; for example, by eliminating the need for an election phase in Paxos. We analyze our approach in terms of communication trips to compare it against the log replication approach, then we conduct an extensive experimental study to compare the performance and scalability of both approaches under various multi-datacenter setups.",
author = "Hatem Mahmoud and Faisal Nawab and Alexander Pucher and Divyakant Agrawal and {El Abbadi}, Amr",
year = "2013",
language = "English",
volume = "6",
pages = "661--672",
booktitle = "Proceedings of the VLDB Endowment",
edition = "9",

}

TY - CHAP

T1 - Low-latency multi-datacenter databases using replicated commit

AU - Mahmoud, Hatem

AU - Nawab, Faisal

AU - Pucher, Alexander

AU - Agrawal, Divyakant

AU - El Abbadi, Amr

PY - 2013

Y1 - 2013

N2 - Web service providers have been using NoSQL datastores to provide scalability and availability for globally distributed data at the cost of sacrificing transactional guarantees. Recently, major web service providers like Google have moved towards building storage systems that provide ACID transactional guarantees for globally distributed data. For example, the newly published system, Spanner, uses Two-Phase Commit and Two-Phase Locking to provide atomicity and isolation for globally distributed data, running on top of Paxos to provide fault-tolerant log replication. We show in this paper that it is possible to provide the same ACID transactional guarantees for multi-datacenter databases with fewer crossdatacenter communication trips, compared to replicated logging. Instead of replicating the transactional log, we replicate the commit operation itself, by running Two-Phase Commit multiple times in different datacenters and using Paxos to reach consensus among datacenters as to whether the transaction should commit. Doing so not only replaces several inter-datacenter communication trips with intra-datacenter communication trips, but also allows us to integrate atomic commitment and isolation protocols with consistent replication protocols to further reduce the number of cross-datacenter communication trips needed for consistent replication; for example, by eliminating the need for an election phase in Paxos. We analyze our approach in terms of communication trips to compare it against the log replication approach, then we conduct an extensive experimental study to compare the performance and scalability of both approaches under various multi-datacenter setups.

AB - Web service providers have been using NoSQL datastores to provide scalability and availability for globally distributed data at the cost of sacrificing transactional guarantees. Recently, major web service providers like Google have moved towards building storage systems that provide ACID transactional guarantees for globally distributed data. For example, the newly published system, Spanner, uses Two-Phase Commit and Two-Phase Locking to provide atomicity and isolation for globally distributed data, running on top of Paxos to provide fault-tolerant log replication. We show in this paper that it is possible to provide the same ACID transactional guarantees for multi-datacenter databases with fewer crossdatacenter communication trips, compared to replicated logging. Instead of replicating the transactional log, we replicate the commit operation itself, by running Two-Phase Commit multiple times in different datacenters and using Paxos to reach consensus among datacenters as to whether the transaction should commit. Doing so not only replaces several inter-datacenter communication trips with intra-datacenter communication trips, but also allows us to integrate atomic commitment and isolation protocols with consistent replication protocols to further reduce the number of cross-datacenter communication trips needed for consistent replication; for example, by eliminating the need for an election phase in Paxos. We analyze our approach in terms of communication trips to compare it against the log replication approach, then we conduct an extensive experimental study to compare the performance and scalability of both approaches under various multi-datacenter setups.

UR - http://www.scopus.com/inward/record.url?scp=84882672111&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84882672111&partnerID=8YFLogxK

M3 - Chapter

AN - SCOPUS:84882672111

VL - 6

SP - 661

EP - 672

BT - Proceedings of the VLDB Endowment

ER -