Re: [Xen-users] loop versus tap?

To:	"Tom Horsley" <tom.horsley@xxxxxxx>
Subject:	Re: [Xen-users] loop versus tap?
From:	"Todd Deshane" <deshantm@xxxxxxxxx>
Date:	Thu, 29 May 2008 21:45:11 -0400
Cc:	"xen-users @ lists. xensource. com" <xen-users@xxxxxxxxxxxxxxxxxxx>
Delivery-date:	Thu, 29 May 2008 18:45:42 -0700
Dkim-signature:	v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from:reply-to:to:subject:cc:in-reply-to:mime-version:content-type:references; bh=Up6fw1zLWHbB3Jwrd/qXiHUj2Wj8kcmsF0PN66RS7cc=; b=rFWvJzlHElll0YDVKYDvrLBDl6UbV1hQRJoyGUtqu6zC6y8Wc9J7wr+SVB1lPKcLkkLLQPLNcdAFmxFv2yOyy5Z/ZN+oAAEz5bD8vyN84lEkrMLoHq9SqbJCJxgZgHww3kTbiqzsYj0v39gvfsUzjJ9bM1nwokEVtrFAtob3JsA=
Domainkey-signature:	a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:reply-to:to:subject:cc:in-reply-to:mime-version:content-type:references; b=HTf06YOb4it91WyYe+Y1stV8unnFO9aJQrv0cCpJnvac6EHlNbvP/uvSQ/QgB8guP/xPaokspIHO7Flh0xCBXszzMda2cwF1nfa6hGiu9ghJufuLn/WHsLq2EbjwjOri82dbpBV3ZDcfzsIN5tRdW5bnm8J0w+xkTs6RVt/9HZs=
Envelope-to:	www-data@xxxxxxxxxxxxxxxxxx
In-reply-to:	<20080529205117.209c6553@zooty>
List-help:	<mailto:xen-users-request@lists.xensource.com?subject=help>
List-id:	Xen user discussion <xen-users.lists.xensource.com>
List-post:	<mailto:xen-users@lists.xensource.com>
List-subscribe:	<http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=subscribe>
List-unsubscribe:	<http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=unsubscribe>
References:	<20080529205117.209c6553@zooty>
Reply-to:	deshantm@xxxxxxxxx
Sender:	xen-users-bounces@xxxxxxxxxxxxxxxxxxx

Hi Tom,

On Thu, May 29, 2008 at 8:51 PM, Tom Horsley <tom.horsley@xxxxxxx> wrote:

I noticed a vast improvement in stability and performance
when I replaced file: disk definitions with tap:aio:
in my virtual machines.

That is the purpose of the block tap driver.

I think file: uses the /dev/loopN mechanism (right?).

That's right. It's the loopback driver

Out of curiosity, could whatever underlying mechanism
(I have no idea what it is) that implements tap:aio
take the place of other uses of /dev/loopN on linux?

I will defer this one to someone else, but give you something
that explains all about the block tap driver:
http://www.usenix.org/event/usenix05/tech/general/full_papers/short_papers/warfield/warfield_html/

Cheers,
Todd

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