Our first whitepaper: Minimizing downtime and maximizing elasticity with Galera Cluster for MySQL
Submitted by hingo on Mon, 12/03/2012 - 10:01
We have released our first whitepaper about Galera Cluster for MySQL, called Minimizing downtime and maximizing elasticity with Galera Cluster for MySQL.
This is a so called "business" white paper. It is not very technical, rather intended to be "manager understandable". We discuss things like what does a minute of downtime cost for some of our customers, and why an active-active, synchronous multi-master cluster is a good choice to minimize downtime. We also look at how Galera fits cloud use cases, scale-in and scale-out, in particular thanks to our automated node provisioning. Again, we mention some statistics on what you can save in cloud infrastructure costs by efficient scale-out/scale-in solution - also known as elasticity.
For the more technical reader, the mid part of the paper should also function as a quick high level comparison of Galera with other popular HA solutions for MySQL. In practice choosing a MySQL HA solution has usually been an exercise in choosing between various trade-offs, the right choice of course depending on your particular situation. While Galera may have some tradeoffs of its own, we at least point out the ones it doesn't have, such as: You can have both synchronous replication and fast failovers!