From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Jan Beulich" Subject: Re: [Xen-devel] [PATCHv1] xen-netfront: always keep the Rx ring full of requests Date: Thu, 02 Oct 2014 14:46:41 +0100 Message-ID: <542D7361020000780003C014@mail.emea.novell.com> References: <1412256826-18874-1-git-send-email-david.vrabel@citrix.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 8BIT Cc: , "Boris Ostrovsky" , To: "David Vrabel" Return-path: Received: from mail.emea.novell.com ([130.57.118.101]:43791 "EHLO mail.emea.novell.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753707AbaJBNqo convert rfc822-to-8bit (ORCPT ); Thu, 2 Oct 2014 09:46:44 -0400 In-Reply-To: <1412256826-18874-1-git-send-email-david.vrabel@citrix.com> Content-Disposition: inline Sender: netdev-owner@vger.kernel.org List-ID: >>> On 02.10.14 at 15:33, wrote: > A full Rx ring only requires 1 MiB of memory. This is not enough > memory that it is useful to dynamically scale the number of Rx > requests in the ring based on traffic rates. The performance benefits are nice, but does the above statement scale to hundreds of guests with perhaps multiple NICs and/or queues? Jan > Keeping the ring full of Rx requests handles bursty traffic better > than trying to converges on an optimal number of requests to keep > filled. > > On a 4 core host, an iperf -P 64 -t 60 run from dom0 to a 4 VCPU guest > improved from 5.1 Gbit/s to 5.6 Gbit/s. Gains with more bursty > traffic are expected to be higher. > > Signed-off-by: David Vrabel