The filesystem which had the greatest length of time to resize (34
minutes) was /u08.....therefore, that is where I did the mkfile.....so
that I could at least compare that on the same filesystem which was
giving our Oracle folks so much grief.....and from the results below,
I'd say that the mkfile's went fairly quickly.....
Sean M <smckeownNO_at_BACKSIESearthlink.net> wrote in message news:<3CF65F06.FCB4A4E2_at_BACKSIESearthlink.net>...
> Richard E Sgrignoli wrote:
> >
> > I did a mkfile as you suggested within the same filesystem with the
> > various sizes; the results were:
> >
> > mkfile 50m test1 (took 2.61 seconds)
> >
> > mkfile 100m test2 (took 4.21 seconds)
> >
> > mkfile 200m test3 (took 7.59 seconds)
>
> But which filesystem did you do these tests on? These results are
> really not very helpful unless you can compare them amongst *all* three
> filesystems in question. The idea is to see if the filesystem that
> takes a lot longer to extend an Oracle datafile *also* takes a lot
> longer to simply create a large empty file, in which case you could
> reasonably rule out Oracle as the problem.
>
> Regards,
> Sean
Received on Fri May 31 2002 - 08:24:56 CDT
