Re: reading the oracle files

When I say I know there is a way, it means that I know there are programs like "fast uploading" (or something like that. the pupose of the query is to create a usage repot of all celular services (SMS, MMS, PTT...) for a big provider in USA, so it will be much more that 40,000,000. in fact this is the amount of record I am testing it on.

The goal is to see if 2 table are syncronised. meaning that there are no record in one that does not exist in the other and other way around.

I tried, using MINUS , not exists , not in + HASH_AJ .... but I did not get lower than 8 minutes for 10,000,000 records. and this is whay I am tring to find another way.

regards,
Lior

DA Morgan a écrit :
> liorlew_at_gmail.com wrote:
> > Hello,
> >
> > I have recently been asked to extract a very large volume of Data from
> > an oracle DB (something like 40,000,000 records).
> >
> > I know that there is some way to open the oracle files directly and
> > extracting the Data from there (uploading). this will do it must
> > faster.
> >
> > does any one know how to do it? I am using c and c++
> >
> > thanks,
> >
> > Lior
>
> <LACED WITH SARCASM>
> I just get a kick out of people that say "I know that there is some way
> to" when in truth they know nothing of the kind. And have you considered
> that your employer signed a legally binding agreement that does not
> contain an exception for you to end-run the license?
> </LACED WITH SARCASM>
>
> 40M rows is not a large volume. Dump them using UTL_FILE or spool to a
> file. You'll spend more time trying to figure out some C/C++ kludge than
> any competent PL/SQL programmer would spend writing it all out to a file
> even on the slowest disk available. In fact you've already spent more
> time on it than I would have.
> --
> Daniel A. Morgan
> http://www.psoug.org
> damorgan_at_x.washington.edu
> (replace x with u to respond)
Received on Fri Jul 08 2005 - 05:02:23 CDT