The goal of this post is twofold: (i) to show the power of GTIDs and MySQL 5.6 replication infrastructure in general; and (ii) show MySQL users a way to hack multi-source replication rather painlessly and rather quickly on top of MySQL 5.6 when using the new MySQL 5.6 replication features. Multi-source Replication Multi-source replication is often regarded as a mean to aggregate, consolidate, different streams of data into one single server instance, a sink. In many cases the following assumptions hold:
- There are no conflicts between transactions from different sources/masters;
- Replication is asynchronous (just as it is in regular MySQL replication).
Such setups are interesting for some use cases, in particular those that are either: driven by the need to gather data …[Read more]