xfs
[Top] [All Lists]

Re: XFS performance issue solved

To: Federico Sevilla III <jijo@xxxxxxxxxxxxxxx>
Subject: Re: XFS performance issue solved
From: Juha Saarinen <juha@xxxxxxxxxxxx>
Date: Tue, 15 May 2001 14:12:06 +1200 (NZST)
Cc: Linux XFS Mailing List <linux-xfs@xxxxxxxxxxx>
In-reply-to: <Pine.LNX.4.21.0105150948440.434-100000@kalapati.jijo.local>
Sender: owner-linux-xfs@xxxxxxxxxxx
On Tue, 15 May 2001, Federico Sevilla III wrote:

> hdparm's -W1 flag turns on the hard drive's write-cacheing, which speeds
> things up, but makes unclean power downs more dangerous. Does anyone know
> how XFS will handle such an event? While on the ReiserFS mailing list
> awhile back, I was under the impression that this could do wonderful
> damage to the filesystem, although I don't know how things are now. But
> then XFS seems to be infinitely more stable than ReiserFS, and the on-disk
> format is much more mature, so maybe (I hope) it handles things
> differently.

Write-caching for drives is inherently dangerous, but how dangerous
depends on how the drive itself is designed. This issue was recently
discussed at huge length on the FreeBSD-stable list, with some people
advocating that your data would never be safe with write-caching, whilst
others said it would, depending on the drive, and what precautions you
take.

Some hard drives have large power capacitors which should have enough
juice in them to write out the cache contents to disk when the power goes
out. Doesn't apply to caching SCSI controllers of course.

> Also, would anyone know how to force the drive to flush its write-cache
> onto disk, like sync makes linux flush its buffers?

You can try the hdparm -f paramater, but whether or not it actually forces
the drive to write out the cache is anyone's guess.

-- 
Regards,


Juha

PGP fingerprint:
B7E1 CC52 5FCA 9756 B502  10C8 4CD8 B066 12F3 9544


<Prev in Thread] Current Thread [Next in Thread>