I have a client which has a number of websites running on a number of different databases and database vendors from Oracle to SQL server to MySQL. The databases do share some information, notably member names and contact information, etc.
Due to time constraints we do not have time to merge all the data into one single database and all all sites run from the one database. So instead the mid-term solution (2-3 years) is to synchronize this data across all the databases. This would include: inserts, updates and deletes.
Is there a particular technology that will provide a solution. I'm not necessarily looking for a vendor's tool, but more of an engineering practice, design pattern that provides the solution from a solutions architect's level. If this does even exist.
First, you need to be aware. Your problem is not "database replication". It is table replication and sharing.
Given a diversity of vendors, you will want to write custom code yourself, using "basic" SQL. In particular, you do not want to get into a situation where different versions of databases are suddenly incompatible, so your complex heterogeneous solution no longer works.
I would strongly suggest that you separate out the shared information into a separate database. All the vendors you mention support multiple databases on a single server. Separating the common information from the rest will make it easier to maintain this data and to propagate changes. It might also open up the ability to use built-in database replication for the shared data. And, you might even find that a cloud based solution can eventually solve the replication problem, by having all requests go to a central database.