This has nothing to do with Perl. It's an issue with your
application and how you are using the database.
A full table scan on in an Oracle database reads all blocks
allocated to that table that have ever had data in them. It
doesn't matter if you've deleted all the rows. They're ways
around that ( truncate table ) but I'm not going into this in
depth here.
Since we don't know what your 80 columns contain ( 80 columns!?
Really?! ), or *anything* about your system, it's impossible
make any useful speculation on your performance problem.
For Oracle related questions, you may join an Oracle list.
You'll find instruction for joining one at www.cybcon.com/~jkstill/util
Jared
On Thursday 28 June 2001 07:12, Michael A. Chase wrote:
> This does not belong on dbi-dev. I am replying to dbi-users where it
> belongs.
>
> Do you have an index on that table with column id as the leading column?
>
> How much time does the exact same query take from SQL*Plus?
> --
> Mac :})
> ** I normally forward private questions to the appropriate mail list. **
> Give a hobbit a fish and he eats fish for a day.
> Give a hobbit a ring and he eats fish for an age.
> ----- Original Message -----
> From: "Konstantin" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Thursday, June 28, 2001 04:27
> Subject: very slow perl Oracle applicaton
>
>
> I have a Perl DBI application connected to Oracle database. In this
> database there is a table called "const_db" which includes about 80 fields.
> After the creation of this table, everything is working fine, but after I
> load 40,000 records using SQL*Loader, this application starts working
> extremely slow. The parts of this application which have to call for a
> query "select * from const_db where id=X" are working so slow, that the
> server deletes the process before it ends. The most unclear thing is that
> even if I delete those 40,000 records, the speed remains very slow.
