Re: SELECT's take a long time compared to other DBMS

You would "get" all rows, but they'd be stored server side until your
client asked for them.

I.e. a cursor would level the field here, since you say that the other
test cases stored the entire result set on the server.  Or did I
misunderstand what you meant there?

On Thu, 4 Sep 2003, Relaxin wrote:

> All rows are required.
>
> ""Shridhar Daithankar"" <shridhar_daithankar@persistent.co.in> wrote in
> message news:3F573E8B.31916.A1063F8@localhost...
> > On 4 Sep 2003 at 0:48, Relaxin wrote:
> > > All of the databases that I tested the query against gave me immediate
> > > access to ANY row of the resultset once the data had been returned.
> > > Ex. If  I'm currently at the first row and then wanted to goto the
> 100,000
> > > row, I would be there immediately, and if I wanted to then goto the 5
> > > row...same thing, I have the record immediately!
> > >
> > > The other databases I tested against stored the entire resultset on the
> > > Server, I'm not sure what PG does...It seems that brings the entire
> > > resultset client side.
> > > If that is the case, how can I have PG store the resultset on the Server
> AND
> > > still allow me immediate access to ANY row in the resultset?
> >
> > You can use a cursor and get only required rows.
> >
> >
> > Bye
> >  Shridhar
> >
> > --
> > Nick the Greek's Law of Life: All things considered, life is 9 to 5
> against.
> >
> >
> > ---------------------------(end of broadcast)---------------------------
> > TIP 1: subscribe and unsubscribe commands go to majordomo@postgresql.org
> >
>
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 5: Have you checked our extensive FAQ?
>
>                http://www.postgresql.org/docs/faqs/FAQ.html
>