"Howard J. Rogers" <hjr_at_dizwell.com> wrote in message news:<3fd2c366$0$14054$afc38c87_at_news.optusnet.com.au>...
> "JEDIDIAH" <jedi_at_nomad.mishnet> wrote in message
> news:3fd15feb_2_at_athenanews.com...
> > On 2003-12-05, Ryan Gaffuri <rgaffuri_at_cox.net> wrote:
> > > We are using an NAS. Are there any conventions to handling Physical
> > > I/O distribution on NAS or SANs? They are relatively new. Since we
> > > dont control which disk or mount point datafiles go on, we really
> > > can't spread out the I/O. Does it matter if we put tables in different
> >
> > I would expect that there is still some mechanism to divy up logical
> > drives among physical drives. You probably want to isolate sequential
> > access files from random access ones.
> >
> > IOW, isolate your online redo logs from your tablespace files.
> >
> > High latency on redo log writes can really clobber performance.
> >
>
>
> Got it in one. I really don't care these days if people insist in RAID-5 for
> their data files. I could live with that. But if you make the redo logs
> share the same devices as datafiles, you're on your own.
>
> As I put it to my students: anything that slows down LGWR is going to be
> felt by the users.
>
> Regards
> HJR
how do you advise you students manage logfiles on SAN or NAS?
Received on Mon Dec 08 2003 - 13:28:49 CST
