Don't settle for eventual: scalable causal consistency for wide-area storage with COPS

By Wyatt Lloyd, Michael J. Freedman, Michael Kaminsky, D. Andersen. Symposium on Operating Systems Principles, 2011

701 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.

Stronger Semantics for Low-Latency Geo-Replicated Storage (2013, positive): “Our previous work, COPS [43], bears the closest similarity to Eiger, as it also uses dependencies to provide causal consistency, and targets low-latency and scalable settings.”
Making Geo-Replicated Systems Fast as Possible, Consistent when Necessary (2012, neutral): “Some systems waive the stable history property, either by rolling back operations and reexecuting them in a different order at some of the replicas [34], or by resorting to a last writer wins strategy, which often results in loss of one of the concurrent updates [25].”
Solving the Straggler Problem with Bounded Staleness (2013, background): “Causal consistency and its variants have attracted renewed attention lately as a way to avoid latencies in geographically-distributed systems [15].”
Trends in big data analytics (2014, background): “COPS [74], scalable causal consistency model, enforces causal dependencies between keys across the entire cluster; all operations on the datastore are ordered and hence inconsistencies due to out-of-order execution are avoided.”
Scaling Memcache at Facebook (2013, positive): “[27] examine causal consistency in COPS, a wide-area storage system.”
TAO: Facebook's Distributed Data Store for the Social Graph (2013, background): “The COPS system [17] provides causal consistency in a highly available key-value store by tracking all dependencies for all keys accessed by a client context.”