xfs
[Top] [All Lists]

Re: Flash drive suitability for XFS journals

To: Jason Parker-Burlingham <jasonp@xxxxxxxxx>
Subject: Re: Flash drive suitability for XFS journals
From: Daniel Pittman <daniel@xxxxxxxxxxxx>
Date: Wed, 02 Jul 2003 10:28:44 +1000
Cc: linux-xfs@xxxxxxxxxxx
In-reply-to: <87n0fy9g8n.fsf@freezer.burling> (Jason Parker-Burlingham's message of "Tue, 01 Jul 2003 15:52:40 -0400")
References: <87n0fy9g8n.fsf@freezer.burling>
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: Gnus/5.090016 (Oort Gnus v0.16) XEmacs/21.5 (cassava)
On Tue, 01 Jul 2003, Jason Parker-Burlingham wrote:
> While attempting to recover an unjournalled filesystem last week, it
> occurred to me that it should be possible to store the XFS log on a
> simple solid-state device. 
[...]

> A little talk on a local LUG list pointed out that these things often
> have a limited number of write cycles, typically in the millions.  One
> poster suggested that write levelling could reduce the impact of this
> limitation

Any of the off-the-shelf devices that emulate mass storage on flash do
their own internal wear leveling, because otherwise the first few
blocks, where Windows puts the FAT, die very quickly...

[...]

> 0) Will this really be useful?  My hope is that storing the log on a
>    device which doesn't use the IDE bus will save the log from
>    becoming corrupted when the IDE disks start to fail. 

Very, very few IDE disk failures cause bus corruption.

[...]

> 3) Is there some other technology suitable for the small office
>    situation which might fill this need?

You will almost certainly get better performance with a compact flash or
PCMCIA flash card attached to a PCMCIA socket or to the IDE bus
directly.

        Daniel

-- 
We are always getting ready to live, but never living.
        -- Ralph Waldo Emerson, _Journals_, 1834


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