August 25, 2005 at 6:39 am
Hi .... i Believe transactional replication also supports bi-directional data synchronisation, so does merge replication. I have a situation where adatabase schema has been scaled across two servers using a network load balancer. The need has arised for having the data redundant as well. My ta=ransactional volume is not too high, probably 5000-6000 data rows. The schema is not a relational schema. What do you think i shud use transactional or merge. Personally, i think i shud use merge as it is designed for two-way data sync. Also, transactional stops if a conflict occurs. would appreciate your thoughts on this.
Thanks
August 26, 2005 at 10:47 am
We use merge replication in a similar scenario, though we have many more transactions. It doesn't scale well to our size, about 300 GB of data/indexes. But for smaller situations it shouldn't be a problem.
Since you have data that's separated, i.e. on one server but not replicated to another server, you'll want to scynchronize the two servers somehow. Look up sp_mergedummyupdate. If that's not necessary, all the better.
Good luck!
Dylan Peters
SQL Server DBA
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply