Transactional storage for geo-replicated systems

By Yair Sovran, Russell Power, Marcos Kawazoe Aguilera, Jinyang Li. Symposium on Operating Systems Principles, 2011

426 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Don't settle for eventual: scalable causal consistency for wide-area storage with COPS (2011, background): “Furthermore, in COPS, scalable datacenters are a ﬁrst-order design goal, while Walter’s sites currently consist of single machines (as a single serialization point for transactions).”
TAO: Facebook's Distributed Data Store for the Social Graph (2013, neutral): “Sovran et al. [31] implement geo-replicated transactions.”
Stronger Semantics for Low-Latency Geo-Replicated Storage (2013, neutral): “Traditional databases, as well as the more recent Walter [52], MDCC [35], Megastore [9], and some Cassandra configurations, provide stronger semantics and a rich data model, but cannot guarantee low latency.”
Making Geo-Replicated Systems Fast as Possible, Consistent when Necessary (2012, neutral): “A first step towards this goal is to allow multiple levels of consistency to coexist [19, 34, 35]: some operations can be executed optimistically, without synchronizing with concurrent actions at other sites, while others require a stronger consistency level and thus require cross-site synchronizati”
Spanner: Google's Globally-Distributed Database (2012, neutral): “Walter [36] provides a variant of snapshot isolation that works within, but not across datacenters.”