[Top] [All Lists]

Wiki entry for disk write caches and RAID controllers

To: xfs@xxxxxxxxxxx
Subject: Wiki entry for disk write caches and RAID controllers
From: Michael Monnerie <michael.monnerie@xxxxxxxxxxxxxxxxxxx>
Date: Thu, 5 Feb 2009 10:19:24 +0100
Organization: it-management http://it-management.at
User-agent: KMail/1.10.3 (Linux/; KDE/4.1.3; x86_64; ; )
I updated the sections about "disk write cache" and "RAID controllers", 
please review and comment.


I merged my original paragraph about "disk write cache" with the "What 
is the problem with the write cache on journaled filesystems?" text.

On Donnerstag 05 Februar 2009 Dave Chinner wrote:
> What I missed was the "barriers turned on" - I was referring
> (context not quoted) to the fact that RAID5 is not unіque in it's
> ability to trash the filesystem on powerfail. You are right,
> barriers on a single disk should prevent filesystem corruption
> and will prevent loss of synchronously written data. only
> asynchronously written data will get lost (just like all the
> stuff sitting in RAM).

So can we say that 1 disk can let write cache on with barriers enabled?

For the Xyratex case: If all write caches are off (or write through), 
does it matter if you use [no]barrier? I guess no, just want to be sure.

mfg zmi
// Michael Monnerie, Ing.BSc    -----      http://it-management.at
// Tel: 0660 / 415 65 31                      .network.your.ideas.
// PGP Key:         "curl -s http://zmi.at/zmi.asc | gpg --import"
// Fingerprint: AC19 F9D5 36ED CD8A EF38  500E CE14 91F7 1C12 09B4
// Keyserver: wwwkeys.eu.pgp.net                  Key-ID: 1C1209B4

<Prev in Thread] Current Thread [Next in Thread>
  • Wiki entry for disk write caches and RAID controllers, Michael Monnerie <=