Is there any distributed database, file system or storage service that supports the following use-case that I have?
It must be possible to replicate data (files) over 3 servers in different geographic locations. Each server should be able to push data and replicate it to the other nodes (multi-master). On top of that, it must be possible to partition the replicated data from one node to another, e.g. data pushed to one node can optionally be replicated to only one other node or none at all (not necessarily replicated everywhere).
Currently, I have checked out different solutions, e.g. SQL Server's merge replication and it is close to what I am looking for but not entirely so I cannot utilize it. AWS/Azure's geo-replication support is also not applicable.
I have also considered the fact that I might have to implement this manually, however, I am writing to ask if there are some other potential solutions I can look at that can handle not only data (file) replication across regions but also partition the replicated data. Currently I am investigating Apache Hadoop but the more I read, the more it seems like a custom implementation is the way to go.