We have a two SQL Server instances that use merge replication to keep in sync.
There is a single table that potentially can add millions of rows in between syncronisations (which is a relative small interval).
This table has a compound primary key (which is also the clustered index).
Replication is currently unable to merge changes between the two database servers (the query times out).
Is there any potential pitfalls with merge replication in this scenario? Is there a performance problem related to the database schema?