"Hilary Cotter" <hilary.cotter@gmail.com> wrote in message
news:eOUpD1q0GHA.328@TK2MSFTNGP06.phx.gbl...
> If you want to use merge replication for this it will create its own guid
> column which you can use for your key. There are performance problems with
> using a guid as your pk. I would advise you to add an identity property to
> your id column, make that your pk and use transactional replication for
> performance reasons.
>
> --
> Hilary Cotter
> Director of Text Mining and Database Strategy
> RelevantNOISE.Com - Dedicated to mining blogs for business intelligence.
>
> This posting is my own and doesn't necessarily represent RelevantNoise's
> positions, strategies or opinions.
>
> 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
>
>
>
> "M" <mxchen@hotvoice.com> wrote in message
> news:uIl46gq0GHA.4816@TK2MSFTNGP06.phx.gbl...
> > We have a 20GB database. Each table has a ID column with Identity and
> > PrimaryKey.
> > Up to Now we only do snapshot replication.
> > Now we need do merge replication.
> > Because the ID column only access by code, I am thinking change the ID
> > column with uniqueidentity data type.
> > but I do not know how to do it.
> > What I try to do is replication each table to a new table without the ID
> > Column, then add the ID column allow Null.
> > After save the table then run SQL something like:
> >
> > Update Tb1 Set ID=NewID() where ID is null.
> >
> > Then Change the ID Column do not allow Null.
> >
> > Any better way to do it?
> >
> > Thanks
> >
> >
> >
>
>

"M" <mxchen@hotvoice.com> wrote in message
news:uIl46gq0GHA.4816@TK2MSFTNGP06.phx.gbl...
> We have a 20GB database. Each table has a ID column with Identity and
> PrimaryKey.
> Up to Now we only do snapshot replication.
> Now we need do merge replication.
> Because the ID column only access by code, I am thinking change the ID
> column with uniqueidentity data type.
> but I do not know how to do it.
> What I try to do is replication each table to a new table without the ID
> Column, then add the ID column allow Null.
> After save the table then run SQL something like:
>
> Update Tb1 Set ID=NewID() where ID is null.
>
> Then Change the ID Column do not allow Null.
>
> Any better way to do it?
>
> Thanks
>
>
>