The three dimensions of data consistency

Abstract : Replication and consistency are essential features of any distributed system and have been studied extensively, however a systematic comparison is lacking. Therefore, we developed the Action-Constraint Framework, which captures both the semantics of replicated data and the behaviour of a replication algorithm. It enables us to decompose the problem of ensuring consistency into three simpler, easily understandable subproblems. As the sub-problems are largely orthogonal, sub-solutions can be mixed and matched. Our unified framework enables a systematic exploration of pessimistic vs. optimistic protocols, full vs. partial replication, strong vs. weak consistency, etc.
Keywords : formel rep
Complete list of metadatas

https://hal.inria.fr/inria-00444797
Contributor : Marc Shapiro <>
Submitted on : Thursday, January 7, 2010 - 11:43:46 AM
Last modification on : Wednesday, August 14, 2019 - 10:46:03 AM
Long-term archiving on : Thursday, October 18, 2012 - 12:06:37 PM

File

cdur2005.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00444797, version 1

Citation

Marc Shapiro, Nishith Krishna. The three dimensions of data consistency. Journées Francophones sur la Cohérence des Données en Univers Réparti (CDUR), Nov 2005, Paris, France. pp.54--58. ⟨inria-00444797⟩

Share

Metrics

Record views

280

Files downloads

304