Tutorial on geo-replication in data center applications

Abstract

Data center applications increasingly require a *geo-replicated* storage system, that is, a storage system replicated across many geographic locations. Geo-replication can reduce access latency, improve availability, and provide disaster tolerance. It turns out there are many techniques for geo-replication with different trade-offs. In this tutorial, we give an overview of these techniques, organized according to two orthogonal dimensions: level of synchrony (synchronous and asynchronous) and type of storage service (read-write, state machine, transaction). We explain the basic idea of these techniques, together with their applicability and trade-offs.