Per this article, a Publisher can only have one remote Distributor in doing SQL Server Transactional Replication.
Each Publisher can be assigned to only a single Distributor instance, but multiple publishers can share a Distributor.
If this is the case, I am wondering how to scale out the publication of a database to a large number of Subscribers. Supposing a Publisher had 200 Subscribers with a large number of transactions, what would be a workable distribution architecture that could handle that load?
Scenario is 'one source of truth' - up to 200 read-only slaves in different geographical areas that each get a slice of the central source, based on data pertinent to their local area (filtered rows).
All writes are made to the central master, which then need to propagate down to the local slave in as close to instant as possible. I've done a bit of research and transactional replication seems to be the best bet.