Re: 9i streams vs triggers

"Niall Litchfield" <n-litchfield_at_audit-commission.gov.uk> wrote in message news:3e1efa52$0$246$ed9e5944_at_reading.news.pipex.net...
> "servant" <mjohns1_at_uhc.com> wrote in message
> news:avmkgh$6kq$1_at_gabriel.uhc.com...
> > Sybrand,
> >
> > I am having trouble understanding how I can design my database to have
> data
> > that is updated and selected simultaneously without contention. The
> update
> > will lock the row and block the select and vice versa, right? Can you
> > expand . . .?
>
> Let me jump in before you get told to RTFM. In an Oracle database an
update
> will not block the select and vice versa. It just won't happen (unless an
> app developer codes it to).
>

Too late :>)

> Now none of this addresses you complexity issue, but it seems likely to me
> that coding views and stored procedures might well address your reporting
> requirements before you get onto ETL. Of course it may not do if the
> massaging is really complex, but there is an awful lot you can do with
plain
> SQL or PL/SQL. What you may run into is different performance
> characteristics for reporting needs to OLTP needs.
>

My main concern with the complexity is to be able to call my existing Java classes so I don't have to write the business rules twice. It is my understanding that I can write Java in PL/SQL so this may work. I am being told that the massaging is not simple but I am going to assume it is not rocket science either. If it is then I'll probably need another solution. I am trying to get a meeting with the data modeler which I hope will shed some light on this.

>
> --
> Niall Litchfield
> Oracle DBA
> Audit Commission UK
> *****************************************
> Please include version and platform
> and SQL where applicable
> It makes life easier and increases the
> likelihood of a good answer
>
> ******************************************
>
>
Received on Fri Jan 10 2003 - 12:00:22 CST