Results 1 to 3 of 3
  1. #1
    Join Date
    Jul 2009
    Posts
    1

    Question Unanswered: Replication Strategy

    I am not a DBA, but rather a developer, so I need a little guidance - we have no DBA

    Here is my scenario,

    We have an office in London and in Boston. The network connection between the 2 offices suk.

    I created a winforms application with a sql server backend. Currently, the database is in Boston, but London is experiencing vast slowness. Thus, we need a solution, and I was thinking we could put a database in London, and set some form of replication. Btw, this is not read-only data, users are creating/editing sales forecasts and users from London and Boston can work on the same forecast (just not at the same time - application restriction - I lock a forecast when someone has it open)

    Now, I am not sure which replication scenario to choose (merge?) or how many servers / databases I need to set this up.

    Also, none of my tables use Guids, they use ints, is this going to be a problem?

    We have no budget for a DBA, so it looks like the burden falls on me.

  2. #2
    Join Date
    Dec 2002
    Posts
    1,245
    Quote Originally Posted by peryan77
    I am not a DBA, but rather a developer, so I need a little guidance - we have no DBA

    Here is my scenario,

    We have an office in London and in Boston. The network connection between the 2 offices suk.

    I created a winforms application with a sql server backend. Currently, the database is in Boston, but London is experiencing vast slowness. Thus, we need a solution, and I was thinking we could put a database in London, and set some form of replication. Btw, this is not read-only data, users are creating/editing sales forecasts and users from London and Boston can work on the same forecast (just not at the same time - application restriction - I lock a forecast when someone has it open)

    Now, I am not sure which replication scenario to choose (merge?) or how many servers / databases I need to set this up.

    Also, none of my tables use Guids, they use ints, is this going to be a problem?

    We have no budget for a DBA, so it looks like the burden falls on me.
    I would generally say that merge would be your best bet, however, merge is a tricky process to manage and tricky to set up. Yes, rowids (ints) are pain to manage in merge replication. Also, how large is your database? Whatever method you choose will require pushing the database out over the network (unless you back it up to CD and mail it).

    Updateable transactional replication is not nice in this scenario, since the network connection is slow.

    Regards,

    hmscott
    Have you hugged your backup today?

  3. #3
    Join Date
    Jan 2003
    Location
    Nottinghamshire, UK
    Posts
    364
    I'd say just have a go Peryan - it's really not that difficult nowadays

    You'll need a couple of test servers and a bit of trial & error using just one Publication/Article/Table

    Merge is definately the way to go, don't forget to allocate your AutID Assignment into Blocks when setting up.

    Just get comfortable with the terminology (Publisher/Subscriber/Distributor/Agent/Snapshots etc...) & you'll be up & running in no time

    GW
    "Everything should be made as simple as possible, but not simpler." - Albert Einstein
    "Everything should be made as complex as possible, so I look Cleverer." - Application Developer

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •