Re: switching to 'direct path read' should help prevent thrashing buffer cache ?

bdbafh wrote:
> On Dec 20, 5:21 pm, p.santos..._at_gmail.com wrote:
> > Folks,
> > Our environment is neither an OLTP nor a DSS system. We have
> > transactions that cross boundaries.
> >
> > Particularly we have these one of jobs that many of our customers run
> > at any time during the
> > day in which they chose to clear a table column.
> > Most clients clear anywhere from 1 million rows to 11 million rows
> > per job.
> >
> > The pseudo-code and SQL looks like this:
> >
> > while SQL%ROWCOUNT < 500,000
> > do
> > UPDATE DEMO_<CLIENT_ID>
> > set col1= null,
> > col2= null,
> > col3= null
> > WHERE ( col1 is not null OR
> > col2 is not null OR
> > col3 is not null ) AND ROWNUM <= 500,000;
> > commit;
> > done.
> >

>

But you could be a good bdbafh and note the commit inside the loop.

jg

--
@home.com is bogus.
http://crave.cnet.co.uk/0,39029477,49286226,00.htm