The multimaster slave replication, also known as multisource replication, uses the GTID mechanism to provide this feature:

Here, for example, you have a master database with slaves for a CMS application. On the other hand, you've got a master and multiple slaves for your e-commerce application. Data could be huge and it is dispersed across several databases and tables.

In this kind of situation, when you want to aggregate data, you generally need to perform tricks with triggers. Alternatively, you need to use any other kind of complicated solution.

That's when multisource replication comes into action. From multiple master or slave databases/tables, you can build a database that will contain data from multiple databases/tables. This is very useful to get all data you want as up-to-date as possible without huge amounts of work.

So if you need to aggregate data easily without needing an Extract-Transform-Load (ETL), use multisource replication. You'll then be able...