Re: intermittent commits

I'll check that out. Thanks.. By the way - there is no way to bypass rollback segments is there? (people cringe) What I mean is, unrecoverble options largely
refer to no redo information, but there is no way to bypass rollback segment activity
is there? (possibly architecturaly abhorant?)

Jonathan Lewis wrote:

> If you want to follow up the idea of breaking up by
> rowid, I published an item about generating a set
> of covering rowid ranges (version 7-ish) for an table
> on my website a couple of days ago.
>
> It is proably the most efficient 'in-situ' strategy for
> a job such as this, although 'create fixed_table from
> previous_table unrecoverable' can sometimes be used
> instead and is likely to be faster.
>
> --
>
> Jonathan Lewis
> Yet another Oracle-related web site: www.jlcomp.demon.co.uk
>
> >Doug Cowles wrote in message <37441EAC.7291CCAF_at_bigfoot.com>...
> >>How do you update a 700,000 row table with intermittent commits?
> >>There is a script on www.orafaq.org that is in blatent violation of the
> >>fetch out of sequence problem. It selects for update and then commits
> >>in between,
> >>which doesn't work. I was thinking of selecting min and max rowid's and
> >>breaking
> >>them into batches. Any ideas?
Received on Thu May 20 1999 - 14:18:54 CDT