Multi master made easy : complex topologies with Tungsten Replicator
Replication and High Availability Strategies
4 April 1:50PM - 2:40PM @ Ballroom G
50 minutes conference
Database replication, in its simplest definition, is just data flowing from a master to some slaves. Until recently, the most daring topologies adopted by DBAs was a master-to-master deployment, with perhaps a DRBD in between. The times of circular replication are over! You can set up much more than that, much more easily, and with more robust results. This talk will focus on Tungsten Replicator, with its ability of creating complex topologies easily: * Point-to-point multi-master topologies * Fan-in topologies * Star topologies * Hybrid deployments, where we integrate multi-masters with single slaves, and multi-master with fan-in. * More hybrid deployments, where multi-master topologies stream their data to a non-MySQL server (examples shown with MongoDB). For each topology, we will see some use cases. For example: point-to-point multi-master is useful for a distributed support application, a fan-in is the ideal deployment for a headquarter with branches, a star topology applies to a stationary headquarters with mobile agents, and the hybrids have several explanations that will be shown during the talk. Attendees will learn what can be done with Tungsten Replicator and how to set up the various topologies in minutes.
QA Director, Continuent
QA Director at Continuent I am active member of the MySQL community and long timer open source enthusiast. During the past 25 years I have worked in various IT related fields, with focus on databases, object oriented programming, system administration. Fluent in Italian, English, Perl, Python, SQL, Lua, C, Bash, and good speaker of C , French, Spanish, Java. I work in cyberspace, with a virtual team.