Subject: Re: [linux-audio-user] Journaling filesystems [was: Ardour Crash + can't boot]
From: Mark Knecht (mknecht_AT_controlnet.com)
Date: Mon Apr 26 2004 - 21:42:54 EEST
Jan Depner wrote:
> I would be willing to bet that XFS might be even better than reiserfs
> but I have no data on that. Mark Knecht documented the responses of the
> different filesystems using Benno Senoner's Latency Test program. I
> have the results on my site at:
>
> http://myweb.cableone.net/eviltwin69/Arcana.html
>
>
> My own (totally unscientific) results are also commented on there.
>
>
> Jan
>
Hi,
I think the value of XFS vs. reiserfs will likely be a bit dependent
on *what* audio application you are running.
I was interested in XFS when I did the tests on Jan's site but I
didn't (and still don't) have the skillset to add XFS to a 2.4 series
kernel so I left it out. I think it would be interesting to give XFS a try.
When I was doing all of this stuff I did some reading from a number
of interesting sites. One of the reasons someone *might* be interested
in XFS over reiserfs is that (and this is totally from memory right now
so I could have it backward) XFS is apparently better tuned for large
files. I would think that, based on my work with Pro Tools where I end
up with large 500MB-2GB wave files all the time, folks primarily running
Audour for audio recording might do better with XFS. That needs to be
verified, but I wouldn't be suprised if it worked out that way.
On the other hand, someone running some audio app that makes use of
a large number of small files, for instance a tracker like Skale or
Cheese Tracker, might find reiserfs works better.
If I have the large/small - reiserfs/XFS thing backward then please
just reverse the example.
Once again, this is all just supposition. It needs to be looked at
in a controlled manner.
Cheers,
Mark
This archive was generated by hypermail 2b28 : Mon Apr 26 2004 - 21:40:11 EEST