"Hilary Cotter" wrote:

> It is very similar to DTS. DTS will blast data to the subscriber, and in
> some cases (by default) will use BULK INSERT. Snapshot Replication will
> write the data to the file system, and the bcp (ODBCBCP actually) blast data
> to the subscriber. So DTS is faster for a single subscriber, but for
> multiple subscribers use Snapshot replication.
>
> Handling dependencies is tricky. First off SQL Server will order your
> publications according to DRI so if you are lucky and sysdepends is accurate
> you will have not problems. If you are not so lucky, you will run into pk fk
> violations.
>
> The best way to handle this is to script out your tables, edit the tables to
> disable all constraints and then run this script on your subscriber. Send
> then snapshot over, making sure in the name conflicts section you select
> keep existing table, and with a post snapshot command re-enable your
> constraints.
>
> --
> Hilary Cotter
> Looking for a SQL Server replication book?
> http://www.nwsu.com/0974973602.html
>
> Looking for a FAQ on Indexing Services/SQL FTS
> http://www.indexserverfaq.com
>
> "Ed" <Ed@discussions.microsoft.com> wrote in message
> news:5C2AEEDA-6CFB-4AD9-BBB6-205C6806FD15@microsoft.com...
> > hi,
> > Can someone tell me the snapshot replication is actually to copy an
> > article from a publisher to subscriber? If so, is that the same as using
> the
> > DTS to copy data?
> > If tables in Subscriber have foreign key relationshiop, i am not able to
> > "delete the existing table and recreate it", does that mean tables in
> > Subscriber cannot have any relationship???
> >
> > If Merge and Transactional replication are able to update data on both
> side,
> > can someone breifly tell the real different between those two?
> >
> > Thanks so much for answering my question.
> >
> > Ed
>
>