Xen 
 
Home About Xen.org Xen Xen Summit Wiki Mailing List Bug Tracker Xen Downloads
 
   
 

xen-devel

Re: [Xen-devel] poor domU VBD performance.

To: Keir Fraser <Keir.Fraser@xxxxxxxxxxxx>
Subject: Re: [Xen-devel] poor domU VBD performance.
From: Andrew Theurer <habanero@xxxxxxxxxx>
Date: Thu, 31 Mar 2005 10:02:07 -0600
Cc: Ian Pratt <m+Ian.Pratt@xxxxxxxxxxxx>, Philip R Auld <pauld@xxxxxxxxxxx>, Kurt Garloff <garloff@xxxxxxx>, Xen development list <xen-devel@xxxxxxxxxxxxxxxxxxx>, Vincent Hanquez <tab@xxxxxxxxx>, Jens Axboe <axboe@xxxxxxx>, Christian Limpach <Christian.Limpach@xxxxxxxxxxxx>
Delivery-date: Thu, 31 Mar 2005 16:02:03 +0000
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <a4afc67b756d65d2e797a956f228cdbf@xxxxxxxxxxxx>
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
References: <A95E2296287EAD4EB592B5DEEFCE0E9D1E3930@xxxxxxxxxxxxxxxxxxxxxxxxxxx> <20050331070514.GH9204@xxxxxxx> <20050331071043.GI9204@xxxxxxx> <63537e2b84ddbba6cb3d970f73c6ab35@xxxxxxxxxxxx> <20050331081900.GK9204@xxxxxxx> <20050331143312.GB13179@xxxxxxxxxxxxxxxxxx> <20050331153449.GE12579@xxxxxxxxxxxxxxxxx> <20050331153925.GP9204@xxxxxxx> <a4afc67b756d65d2e797a956f228cdbf@xxxxxxxxxxxx>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Mozilla Thunderbird 0.8 (Windows/20040913)
Keir Fraser wrote:


On 31 Mar 2005, at 16:39, Jens Axboe wrote:

Not necessarily - either your io rate is not fast enough to sustain a
substantial queue depth, in that case you get plugging on basically
every io anyways. If on the other hand the io rate is high enough to
maintain a queue depth of > 1, then the plugging will never take place
because the queue never empties.

So all in all, I don't think the temporary work-around will be such a
bad idea. I would still rather implement the queue tracking though, it
should not be more than a few lines of code.


I've checked in something along the lines of what you described into both the 2.0-testing and the unstable trees. Looks to have identical performance to the original simple patch, at least for a bulk 'dd'.

I'll do a pull of unstable and see what I get with o_direct, thanks.

-Andrew


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel