I've read something similar indeed.
Aswell as most IDE disks really do have less quality components. Scsi
disks have also smaller, thicker platters to enable them to run more
stable and on higher speeds, larger magnets etc etc.
For those of you who can read Dutch:
http://www.tweakers.net/reviews/362/3 In the second-last paragraph is a
summary of Maxtor's view on the differences between ATA and SCSI.
Regards,
Arjen
> -----Oorspronkelijk bericht-----
> Van: pgsql-general-owner@postgresql.org
> [mailto:pgsql-general-owner@postgresql.org] Namens Dennis Gearon
> Verzonden: donderdag 27 maart 2003 18:10
> Aan: Daniel R. Anderson
> CC: Pgsql-General
> Onderwerp: Re: [GENERAL] Performance tuning in PostgreSQL
>
>
> I read a recent article on Tom's hardware that said, even
> given the same speed
> of data from the read heads and the same buffer size, a SCSI
> drive will work
> better for a server, and the IDE drive will work better for
> the desktop. The
> caching algorithms are optimised with the assumption that a
> SCSI drive will BE
> on a server and an IDE drive will BE on a desktop.
>
> Daniel R. Anderson wrote:
> > On Wed, 2003-03-26 at 12:56, Dennis Gearon wrote:
> >
> >>In General, the rotational speed is higher on SCSCI disks, and this
> >>increases
> >>the tranfer rate from the disc, which is the limitation for
> anything not in the
> >>disk's cache. Given the same areal dinsity, a 15,000 SCSI
> drive will be 50%
> >>faster in tranfer rate than a 10,000 IDE drive.
> >
> >
> > For anybody interested I got the story off of slashdot:
> >
> >
> http://slashdot.org/article.pl?sid=>
03/02/21/0553249&mode=thread&tid=13
> > 7
> >
> > The claim is that these ATA drives have "SCSI-like specs at
> 30% less
> > of the price". SCSI-LIKE != SCSI though. :-(
> >
>
>
> ---------------------------(end of
> broadcast)---------------------------
> TIP 1: subscribe and unsubscribe commands go to
> majordomo@postgresql.org
>
