SQL Server Performance

Replication pointers

Discussion in 'SQL Server 2005 Replication' started by jimforjim, Jan 17, 2007.

  1. jimforjim New Member


    My company is in the process of developing a hosted application for marketing departments in medium to large corporations. Our intention is that each client will have its own dedicated SQL 2K5 database, but that there will be a central repository of common information, such as countries, states, languages, etc, as well as sprocs, views, UDFs, etc. We anticipate that there will be 25-50 of these clients, and all clients need to be on the same version of the application at the same time. We've been looking at replication as a possible solution that would allow us to use a central database as the master to push out schema and data changes. We do not need a bi-directional data flow - only from the master to the subscribers.

    I have these questions:
    * When new entities are added to the master database, can they automatically be added to the publication, or do we need to manually modify the publication to include them?
    * Our intention is to have the databases reside on the same server. Do you have recommendations on how to best implement row filtering? Different SQL Users?

  2. MohammedU New Member

    In this case I believe you can use transactional replication but I don't never done with UDFs...
    You can use horizonal filtering based on client, area etc...

    Replicating Code Modules with SQL Server 2005

    Check the BOL topic "Filtering Published Data "

    Mohammed U.

Share This Page