Albatross

Lightweight elasticity in shared storage databases for the cloud using live data migration

Sudipto Das, Shoji Nishimura, Divyakant Agrawal, Amr El Abbadi

Research output: Chapter in Book/Report/Conference proceedingChapter

89 Citations (Scopus)

Abstract

Database systems serving cloud platforms must serve large numbers of applications (or tenants). In addition to managing tenants with small data footprints, different schemas, and variable load patterns, such multitenant data platforms must minimize their operating costs by efficient resource sharing. When deployed over a pay-per-use infrastructure, elastic scaling and load balancing, enabled by low cost live migration of tenant databases, is critical to tolerate load variations while minimizing operating cost. However, existing databases-relational databases and Key-Value stores alike-lack low cost live migration techniques, thus resulting in heavy performance impact during elastic scaling. We present Albatross, a technique for live migration in a multitenant database serving OLTP style workloads where the persistent database image is stored in a network attached storage. Albatross migrates the database cache and the state of active transactions to ensure minimal impact on transaction execution while allowing transactions active during migration to continue execution. It also guarantees serializability while ensuring correctness during failures. Our evaluation using two OLTP benchmarks shows that Albatross can migrate a live tenant database with no aborted transactions, negligible impact on transaction latency and throughput both during and after migration, and an unavailability window as low as 300 ms.

Original languageEnglish
Title of host publicationProceedings of the VLDB Endowment
Pages494-505
Number of pages12
Volume4
Edition8
Publication statusPublished - May 2011
Externally publishedYes

Fingerprint

Elasticity
Operating costs
Resource allocation
Costs
Throughput

ASJC Scopus subject areas

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

Cite this

Das, S., Nishimura, S., Agrawal, D., & El Abbadi, A. (2011). Albatross: Lightweight elasticity in shared storage databases for the cloud using live data migration. In Proceedings of the VLDB Endowment (8 ed., Vol. 4, pp. 494-505)

Albatross : Lightweight elasticity in shared storage databases for the cloud using live data migration. / Das, Sudipto; Nishimura, Shoji; Agrawal, Divyakant; El Abbadi, Amr.

Proceedings of the VLDB Endowment. Vol. 4 8. ed. 2011. p. 494-505.

Research output: Chapter in Book/Report/Conference proceedingChapter

Das, S, Nishimura, S, Agrawal, D & El Abbadi, A 2011, Albatross: Lightweight elasticity in shared storage databases for the cloud using live data migration. in Proceedings of the VLDB Endowment. 8 edn, vol. 4, pp. 494-505.
Das S, Nishimura S, Agrawal D, El Abbadi A. Albatross: Lightweight elasticity in shared storage databases for the cloud using live data migration. In Proceedings of the VLDB Endowment. 8 ed. Vol. 4. 2011. p. 494-505
Das, Sudipto ; Nishimura, Shoji ; Agrawal, Divyakant ; El Abbadi, Amr. / Albatross : Lightweight elasticity in shared storage databases for the cloud using live data migration. Proceedings of the VLDB Endowment. Vol. 4 8. ed. 2011. pp. 494-505
@inbook{6ea39f7e6dd747f48e1bca3ad5b59485,
title = "Albatross: Lightweight elasticity in shared storage databases for the cloud using live data migration",
abstract = "Database systems serving cloud platforms must serve large numbers of applications (or tenants). In addition to managing tenants with small data footprints, different schemas, and variable load patterns, such multitenant data platforms must minimize their operating costs by efficient resource sharing. When deployed over a pay-per-use infrastructure, elastic scaling and load balancing, enabled by low cost live migration of tenant databases, is critical to tolerate load variations while minimizing operating cost. However, existing databases-relational databases and Key-Value stores alike-lack low cost live migration techniques, thus resulting in heavy performance impact during elastic scaling. We present Albatross, a technique for live migration in a multitenant database serving OLTP style workloads where the persistent database image is stored in a network attached storage. Albatross migrates the database cache and the state of active transactions to ensure minimal impact on transaction execution while allowing transactions active during migration to continue execution. It also guarantees serializability while ensuring correctness during failures. Our evaluation using two OLTP benchmarks shows that Albatross can migrate a live tenant database with no aborted transactions, negligible impact on transaction latency and throughput both during and after migration, and an unavailability window as low as 300 ms.",
author = "Sudipto Das and Shoji Nishimura and Divyakant Agrawal and {El Abbadi}, Amr",
year = "2011",
month = "5",
language = "English",
volume = "4",
pages = "494--505",
booktitle = "Proceedings of the VLDB Endowment",
edition = "8",

}

TY - CHAP

T1 - Albatross

T2 - Lightweight elasticity in shared storage databases for the cloud using live data migration

AU - Das, Sudipto

AU - Nishimura, Shoji

AU - Agrawal, Divyakant

AU - El Abbadi, Amr

PY - 2011/5

Y1 - 2011/5

N2 - Database systems serving cloud platforms must serve large numbers of applications (or tenants). In addition to managing tenants with small data footprints, different schemas, and variable load patterns, such multitenant data platforms must minimize their operating costs by efficient resource sharing. When deployed over a pay-per-use infrastructure, elastic scaling and load balancing, enabled by low cost live migration of tenant databases, is critical to tolerate load variations while minimizing operating cost. However, existing databases-relational databases and Key-Value stores alike-lack low cost live migration techniques, thus resulting in heavy performance impact during elastic scaling. We present Albatross, a technique for live migration in a multitenant database serving OLTP style workloads where the persistent database image is stored in a network attached storage. Albatross migrates the database cache and the state of active transactions to ensure minimal impact on transaction execution while allowing transactions active during migration to continue execution. It also guarantees serializability while ensuring correctness during failures. Our evaluation using two OLTP benchmarks shows that Albatross can migrate a live tenant database with no aborted transactions, negligible impact on transaction latency and throughput both during and after migration, and an unavailability window as low as 300 ms.

AB - Database systems serving cloud platforms must serve large numbers of applications (or tenants). In addition to managing tenants with small data footprints, different schemas, and variable load patterns, such multitenant data platforms must minimize their operating costs by efficient resource sharing. When deployed over a pay-per-use infrastructure, elastic scaling and load balancing, enabled by low cost live migration of tenant databases, is critical to tolerate load variations while minimizing operating cost. However, existing databases-relational databases and Key-Value stores alike-lack low cost live migration techniques, thus resulting in heavy performance impact during elastic scaling. We present Albatross, a technique for live migration in a multitenant database serving OLTP style workloads where the persistent database image is stored in a network attached storage. Albatross migrates the database cache and the state of active transactions to ensure minimal impact on transaction execution while allowing transactions active during migration to continue execution. It also guarantees serializability while ensuring correctness during failures. Our evaluation using two OLTP benchmarks shows that Albatross can migrate a live tenant database with no aborted transactions, negligible impact on transaction latency and throughput both during and after migration, and an unavailability window as low as 300 ms.

UR - http://www.scopus.com/inward/record.url?scp=84859414671&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84859414671&partnerID=8YFLogxK

M3 - Chapter

VL - 4

SP - 494

EP - 505

BT - Proceedings of the VLDB Endowment

ER -