Write Fast, Read in the Past: Causal Consistency for Client-Side Applications

Marek Zawirski; Nuno Preguiça; Sérgio Duarte; Annette Bieniusa; Valter Balegas; Marc Shapiro

doi:10.1145/2814576.2814733

Write Fast, Read in the Past: Causal Consistency for Client-Side Applications

Marek Zawirski, Nuno Preguiça, Sérgio Duarte, Annette Bieniusa, Valter Balegas, Marc Shapiro

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

53 Citations (Scopus)

Abstract

Client-side apps (e.g., mobile or in-browser) need cloud data to be available in a local cache, for both reads and updates. For optimal user experience and developer support, the cache should be consistent and fault-tolerant. In order to scale to high numbers of unreliable and resource-poor clients, and large database, the system needs to use resources sparingly. The SwiftCloud distributed object database is the first to provide fast reads and writes via a causally-consistent client-side local cache backed by the cloud. It is thrifty in resources and scales well, thanks to consistent versioning provided by the cloud, using small and bounded metadata. It remains available during faults, switching to a different data centre when the current one is not responsive, while maintaining its consistency guarantees. This paper presents the SwiftCloud algorithms, design, and experimental evaluation. It shows that client-side apps enjoy the high performance and availability, under the same guarantees as a remote cloud data store, at a small cost.

Original language	English
Title of host publication	Proceedings of the 16th Annual Middleware Conference
Place of Publication	New York, NY, USA
Publisher	ACM - Association for Computing Machinery
Pages	75-87
Number of pages	13
ISBN (Print)	978-1-4503-3618-5
DOIs	https://doi.org/10.1145/2814576.2814733
Publication status	Published - 2015
Event	16th International Middleware Conference - Vancouver, Canada Duration: 7 Dec 2015 → 11 Dec 2015

Publication series

Name	Middleware '15
Publisher	ACM

Conference

Conference	16th International Middleware Conference
Country/Territory	Canada
City	Vancouver
Period	7/12/15 → 11/12/15

Keywords

Causal Consistency, Client-side Storage, Eventual Consistency, Fault Tolerance, Geo-replication

Access to Document

10.1145/2814576.2814733

http://doi.acm.org/10.1145/2814576.2814733

Cite this

@inproceedings{bfda0edccf724134b4b18c64043f222a,

title = "Write Fast, Read in the Past: Causal Consistency for Client-Side Applications",

abstract = "Client-side apps (e.g., mobile or in-browser) need cloud data to be available in a local cache, for both reads and updates. For optimal user experience and developer support, the cache should be consistent and fault-tolerant. In order to scale to high numbers of unreliable and resource-poor clients, and large database, the system needs to use resources sparingly. The SwiftCloud distributed object database is the first to provide fast reads and writes via a causally-consistent client-side local cache backed by the cloud. It is thrifty in resources and scales well, thanks to consistent versioning provided by the cloud, using small and bounded metadata. It remains available during faults, switching to a different data centre when the current one is not responsive, while maintaining its consistency guarantees. This paper presents the SwiftCloud algorithms, design, and experimental evaluation. It shows that client-side apps enjoy the high performance and availability, under the same guarantees as a remote cloud data store, at a small cost.",

keywords = "Causal Consistency, Client-side Storage, Eventual Consistency, Fault Tolerance, Geo-replication",

author = "Marek Zawirski and Nuno Pregui{\c c}a and S{\'e}rgio Duarte and Annette Bieniusa and Valter Balegas and Marc Shapiro",

year = "2015",

doi = "10.1145/2814576.2814733",

language = "English",

isbn = "978-1-4503-3618-5",

series = "Middleware '15",

publisher = "ACM - Association for Computing Machinery",

pages = "75--87",

booktitle = "Proceedings of the 16th Annual Middleware Conference",

address = "United States",

note = "16th International Middleware Conference ; Conference date: 07-12-2015 Through 11-12-2015",

}

Zawirski, M, Preguiça, N , Duarte, S, Bieniusa, A, Balegas, V & Shapiro, M 2015, Write Fast, Read in the Past: Causal Consistency for Client-Side Applications. in Proceedings of the 16th Annual Middleware Conference. Middleware '15, ACM - Association for Computing Machinery, New York, NY, USA, pp. 75-87, 16th International Middleware Conference, Vancouver, Canada, 7/12/15. https://doi.org/10.1145/2814576.2814733

Write Fast, Read in the Past: Causal Consistency for Client-Side Applications. / Zawirski, Marek; Preguiça, Nuno ; Duarte, Sérgio et al.
Proceedings of the 16th Annual Middleware Conference. New York, NY, USA: ACM - Association for Computing Machinery, 2015. p. 75-87 (Middleware '15).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Write Fast, Read in the Past

T2 - 16th International Middleware Conference

AU - Zawirski, Marek

AU - Preguiça, Nuno

AU - Duarte, Sérgio

AU - Bieniusa, Annette

AU - Balegas, Valter

AU - Shapiro, Marc

PY - 2015

Y1 - 2015

N2 - Client-side apps (e.g., mobile or in-browser) need cloud data to be available in a local cache, for both reads and updates. For optimal user experience and developer support, the cache should be consistent and fault-tolerant. In order to scale to high numbers of unreliable and resource-poor clients, and large database, the system needs to use resources sparingly. The SwiftCloud distributed object database is the first to provide fast reads and writes via a causally-consistent client-side local cache backed by the cloud. It is thrifty in resources and scales well, thanks to consistent versioning provided by the cloud, using small and bounded metadata. It remains available during faults, switching to a different data centre when the current one is not responsive, while maintaining its consistency guarantees. This paper presents the SwiftCloud algorithms, design, and experimental evaluation. It shows that client-side apps enjoy the high performance and availability, under the same guarantees as a remote cloud data store, at a small cost.

AB - Client-side apps (e.g., mobile or in-browser) need cloud data to be available in a local cache, for both reads and updates. For optimal user experience and developer support, the cache should be consistent and fault-tolerant. In order to scale to high numbers of unreliable and resource-poor clients, and large database, the system needs to use resources sparingly. The SwiftCloud distributed object database is the first to provide fast reads and writes via a causally-consistent client-side local cache backed by the cloud. It is thrifty in resources and scales well, thanks to consistent versioning provided by the cloud, using small and bounded metadata. It remains available during faults, switching to a different data centre when the current one is not responsive, while maintaining its consistency guarantees. This paper presents the SwiftCloud algorithms, design, and experimental evaluation. It shows that client-side apps enjoy the high performance and availability, under the same guarantees as a remote cloud data store, at a small cost.

KW - Causal Consistency, Client-side Storage, Eventual Consistency, Fault Tolerance, Geo-replication

U2 - 10.1145/2814576.2814733

DO - 10.1145/2814576.2814733

M3 - Conference contribution

SN - 978-1-4503-3618-5

T3 - Middleware '15

SP - 75

EP - 87

BT - Proceedings of the 16th Annual Middleware Conference

PB - ACM - Association for Computing Machinery

CY - New York, NY, USA

Y2 - 7 December 2015 through 11 December 2015

ER -

Write Fast, Read in the Past: Causal Consistency for Client-Side Applications

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this