Re: RAIDframe performance (RAID-5)

[Johan Ihren <johani%johani.org@localhost>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

> On 26 Aug 2008, at 19:10, Johan Ihren wrote:
> > I just configured a 150GB RAID-5 out of 3 75GB components (one half  
> > of a
> > reasonably modern 160GB disk each). Write performance suck badly.
>
> What file system parameters are you using? On at least at RAIDframe  
> RAID 1
> a block size of 16384 bytes and fragment size of 2046 bytes.

Apologies for the slow response, the box is remote and I didn't get  
access to it until today.

I'm using 16384 and 2048 (it's the raid0g partition below that I'm  
testing with).

- -bash-3.2# disklabel raid0 | grep Cyl
 a:   1048576        63     4.2BSD      0     0     0  # (Cyl.       
0*-   1024*)
 b:   4194304   1048639       swap                     # (Cyl.    
1024*-   5120*)
 d: 327679744         0     unused      0     0        # (Cyl.      0  
- - 319999*)
 e:  16777216   5242943     4.2BSD      0     0     0  # (Cyl.    
5120*-  21504*)
 f:   4194304  22020159     4.2BSD      0     0     0  # (Cyl.   
21504*-  25600*)
 g: 301465281  26214463     4.2BSD   2048 16384     0  # (Cyl.   
25600*- 319999*)

Side note: for many years I used to edit my disklabels manually before  
re-writing them with

disklabel -r -R dev protofile

but for convenience I've switched to using "disklabel -i dev" and then  
just hack it up interactively. I now notice that the disklabels  
created by "disklabel -i" have "0" for bsize and fsize (and also cpg/ 
sgs). Haven't thought about that before, but is this intended behaviour?

Regards,

Johan

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Darwin)

iD8DBQFItvHaKJmr+nqSTbYRAmIUAJ9wJZWc7bx/PQspaoMVB3lSHee60wCgoGMX
Hrz5wbHxhinZeVmo3iUnXH4=
=jfgA
-----END PGP SIGNATURE-----