[dba-Tech] Performance in disc partitions

With RAID (OnBoard RAID, ASUS) I did have some pretty bad experience with 
some MBs.
After some time operating ok the RAID went bad again and again and required 
rebuild, until I decided to no longer use the RAID.

Lembit

----- Original Message ----- 
From: "jwcolby" <jwcolby at colbyconsulting.com>
To: "'Discussion of Hardware and Software issues'" 
<dba-tech at databaseadvisors.com>
Sent: Monday, January 28, 2008 4:15 PM
Subject: Re: [dba-Tech] Performance in disc partitions


> Since you mentioned the RAID thing I might as well add that as soon as you
> put stuff on RAID that advantage may go away.  Raid takes over and puts
> stuff where it will on the disks under its care.  It might very well 
> create
> the first volume on the edge, who knows.  OTOH it might not.
>
>
> John W. Colby
> Colby Consulting
> www.ColbyConsulting.com
> -----Original Message-----
> From: dba-tech-bounces at databaseadvisors.com
> [mailto:dba-tech-bounces at databaseadvisors.com] On Behalf Of DJK(John)
> Robinson
> Sent: Monday, January 28, 2008 9:24 AM
> To: 'Discussion of Hardware and Software issues'
> Subject: Re: [dba-Tech] Performance in disc partitions
>
> Well, I've trawled the web, and come up with the usual bunch of 
> conflicting,
> unsubstantiated tosh.  My favorite was along the lines of "running RAID 1
> completely eliminates the need for backup".  Yeah, right.  Step away from 
> my
> system, sir.
>
> The consensus, though, is that the first partition goes next to the edge,
> with the last one by the spindle.  People differ in saying what difference
> this makes - a lot or hardly any:
> http://partition.radified.com/partitioning_2.htm
> http://www.xsibase.com/forum/index.php?board=15;action=display;threadid=3452
> 8
>
> I found PassMark Performance Test (30-day free trial) at
> http://www.passmark.com/products/pt.htm;  I've no idea how good it is
> generally
>
> I tried this on first and last partitions on a 500GB disc and got results
> suggesting 25%-45% better performance on the *first* partition, which
> supports the 'edge is first is better' theory.  I'm not publishing the
> results, because they weren't on a scientific basis, but I *am* going to 
> put
> my active stuff first, near the edge, and my dormant stuff last!
>
> John
>
> _______________________________________________
> dba-Tech mailing list
> dba-Tech at databaseadvisors.com
> http://databaseadvisors.com/mailman/listinfo/dba-tech
> Website: http://www.databaseadvisors.com
>
>
> -- 
> No virus found in this incoming message.
> Checked by AVG Free Edition.
> Version: 7.5.516 / Virus Database: 269.19.13/1246 - Release Date: 
> 27.01.2008 18:39
>
>