PhysiCS-NMSI: efficient consistent snapshots for scalable snapshot isolation

Alejandro Tomsic 1 Tyler Crain 1 Marc Shapiro 1
1 Regal - Large-Scale Distributed Systems and Applications
Inria de Paris, LIP6 - Laboratoire d'Informatique de Paris 6
Abstract : Non-Monotonic Snapshot Isolation (NMSI), a variant of the widely deployed Snapshot Isolation (SI), aims at improving scalability by relaxing snapshots. In contrast to SI, NMSI snapshots are causally consistent, which allows for more par-allelism and a reduced abort rate. This work documents the design of PhysiCS-NMSI, a trans-actional protocol implementing NMSI in a partitioned data store. It is the first protocol to rely on a single scalar taken from a physical clock for tracking causal dependencies and building causally consistent snapshots. Its commit protocol ensures atomicity and the absence of write-write conflicts. We argue that PhysiCS-NMSI approach increases concur-rency and reduces abort rate and metadata overhead as compared to state-of-art systems.
Document type :
Conference papers
Complete list of metadatas

Cited literature [15 references]  Display  Hide  Download

https://hal.inria.fr/hal-01350657
Contributor : Marc Shapiro <>
Submitted on : Monday, August 1, 2016 - 11:23:33 AM
Last modification on : Thursday, March 21, 2019 - 1:20:58 PM
Long-term archiving on : Tuesday, November 8, 2016 - 7:28:12 PM

Files

main.pdf
Files produced by the author(s)

Identifiers

Citation

Alejandro Tomsic, Tyler Crain, Marc Shapiro. PhysiCS-NMSI: efficient consistent snapshots for scalable snapshot isolation. PaPoC 2016 - 2nd Workshop on the Principles and Practice of Consistency for Distributed Data, Apr 2016, London, United Kingdom. pp.4, ⟨10.1145/2911151.2911166⟩. ⟨hal-01350657⟩

Share

Metrics

Record views

387

Files downloads

105