xfs
[Top] [All Lists]

Re: streaming media and realtime subvolume

To: Seth Mos <knuffie@xxxxxxxxx>
Subject: Re: streaming media and realtime subvolume
From: George Georgalis <georgw@xxxxxxxxx>
Date: Thu, 7 Nov 2002 12:29:52 -0500
Cc: linux-xfs@xxxxxxxxxxx
In-reply-to: <4.3.2.7.2.20021107162618.035ef2e0@xxxxxxxxxxxxx>
References: <20021107151729.GA7085@trot> <4.3.2.7.2.20021107162618.035ef2e0@xxxxxxxxxxxxx>
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: Mutt/1.3.28i
On Thu, Nov 07, 2002 at 04:31:42PM +0100, Seth Mos wrote:
>At 10:17 7-11-2002 -0500, George Georgalis wrote:

>>I expect a single 7200rpm ide drive to be able to handle the stream.
>
>reading from or writing to?

reading only.

>>I'm using raid for high availability; but uninterrupted streams is
>>a must. What mount parameters can I use to assure uninterrupted io?
>
>Is it a single file?

Have some control over that, but it may not always be one file.

>>Are there other adjustments I should be thinking of? I've seen a few
>>mentions of the Linux xfs realtime subvolume but no doc, is it ready for
>>production? From what I can tell, it's a non journeled contiguous data
>>block. Maybe I should just use ext2 for the media files? Would that be
>
>Read speed is about as fast on most filesystems, at least for a IDE disk.
>That is, xfs, ext2/3 and reiserfs are almost as fast reading a single large 
>file. XFS is good with large files in particular.

Good.

>>higher performance? I'm not too worried about fsck, because in the case
>>of corruption I can remake the filesystem (data partition) and renew the
>>data from the node server.
>
>If data corruption is not a issue and you do a lot of writing to the 
>filesystem, it might* be faster to use a non journaling filesystem.
>
>*Baring a number of caveats ofcourse. For most people this is negligble but 
>not always so. It depends on too much factors.

The media files will only be written to occasionally. peak or average
read times I don't think will be an issue, _continuous_ audio/video is the
priority. I don't know specifically what the problem was with the blade
kernel but the box would not respond for several seconds at times,
apparently for journal commitment? 

I imagine much of what I need will be os tuning (tips welcome) eg
minimal remote logging; but re xfs, I'm looking for mount parameters or
anything that would make disk reads as continuous as possible.

Thanks,
// George


-- 
GEORGE GEORGALIS, System Admin/Architect    cell: 347-451-8229 
Security Services, Web, Mail,            mailto:george@xxxxxxxxx 
File, Print, DB and DNS Servers.       http://www.galis.org/george 


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