qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Julian Stecklina <jsteckli@os.inf.tu-dresden.de>
Cc: "snabb-devel@googlegroups.com" <snabb-devel@googlegroups.com>,
	qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] [snabb-devel:276] Re: snabbswitch integration with QEMU for userspace ethernet I/O
Date: Tue, 28 May 2013 17:33:45 +0200	[thread overview]
Message-ID: <51A4CE59.6060307@os.inf.tu-dresden.de> (raw)
In-Reply-To: <CAA2XHbfjBtL9KoO9tc8OLjowk4EUX3oudd+ufE1WNQ3M5WLNig@mail.gmail.com>

On 05/28/2013 04:42 PM, Luke Gorrie wrote:
> On 28 May 2013 13:53, Michael S. Tsirkin <mst@redhat.com
> <mailto:mst@redhat.com>> wrote:
> 
>     Yes, you can maybe trade some of this latency for power/CPU cycles by
>     aggressive polling.  Doing this in a way that does not waste a lot of
>     power would be tricky.
> 
> 
> For what it's worth, here is my mental model right now:
> 
> Administrator budgets one CPU core for network I/O (VM and NIC).
> Switch uses that CPU to deliver sufficient speed (e.g. 10 M packets/sec
> ~ 40Gbps).
> Switch uses micro-sleeps to cut cpu usage to ~ 1% in idle periods.

With virtio the backend can decide whether it wants to be notified by
the client. If you disable all notifications you are in polling mode. If
your backend/switch doesn't find anything to do it can reenable
notifications and block. Thus you would naturally revert to non-polling
mode, when the network load is low.

Julian

  reply	other threads:[~2013-05-28 15:34 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-05-26  9:32 [Qemu-devel] snabbswitch integration with QEMU for userspace ethernet I/O Luke Gorrie
2013-05-27  9:34 ` Stefan Hajnoczi
2013-05-27 15:18   ` Michael S. Tsirkin
2013-05-27 15:43   ` Paolo Bonzini
2013-05-27 16:18     ` Anthony Liguori
2013-05-27 16:18       ` Paolo Bonzini
2013-05-27 17:01         ` Anthony Liguori
2013-05-27 17:13           ` Michael S. Tsirkin
2013-05-27 18:31             ` Anthony Liguori
2013-05-28 10:39       ` Luke Gorrie
2013-05-28 10:10   ` Luke Gorrie
2013-05-28 10:35     ` Stefan Hajnoczi
2013-05-28 11:36     ` Julian Stecklina
2013-05-28 11:53       ` Michael S. Tsirkin
2013-05-28 12:09         ` Julian Stecklina
2013-05-28 13:56           ` Michael S. Tsirkin
2013-05-28 15:35             ` Julian Stecklina
2013-05-28 15:44               ` Michael S. Tsirkin
2013-05-28 12:48         ` [Qemu-devel] [snabb-devel:276] " Luke Gorrie
2013-05-28 13:12           ` Julian Stecklina
2013-05-28 13:42             ` [Qemu-devel] [snabb-devel:280] " Luke Gorrie
2013-05-28 14:42         ` [Qemu-devel] [snabb-devel:276] " Luke Gorrie
2013-05-28 15:33           ` Julian Stecklina [this message]
2013-05-28 17:00       ` [Qemu-devel] " Anthony Liguori
2013-05-28 17:17         ` Michael S. Tsirkin
2013-05-28 18:55           ` Anthony Liguori
2013-05-29 10:31             ` Stefano Stabellini
2013-05-29 12:25               ` Michael S. Tsirkin
2013-05-29 13:04                 ` Stefano Stabellini
2013-06-04 12:19               ` [Qemu-devel] [snabb-devel:300] " Luke Gorrie
2013-06-04 12:49                 ` Julian Stecklina
2013-06-04 20:09                   ` [Qemu-devel] [snabb-devel:326] " Luke Gorrie
2013-06-04 12:56                 ` [Qemu-devel] [snabb-devel:300] " Michael S. Tsirkin
2013-06-05  6:09                   ` [Qemu-devel] [snabb-devel:327] " Luke Gorrie
2013-05-29  7:49           ` [Qemu-devel] " Stefan Hajnoczi
2013-05-29  9:08             ` Michael S. Tsirkin
2013-05-29 14:21               ` Stefan Hajnoczi
2013-05-29 14:48                 ` Michael S. Tsirkin
2013-05-29 16:02                 ` Julian Stecklina
2013-05-30  2:35                   ` ronnie sahlberg
2013-05-30  6:46                   ` Stefan Hajnoczi
2013-05-30  6:55                     ` Michael S. Tsirkin
2013-05-30  7:11                     ` [Qemu-devel] [snabb-devel:308] " Luke Gorrie
2013-05-30  8:08                     ` [Qemu-devel] " Julian Stecklina
2013-05-29 12:32         ` Julian Stecklina
2013-05-29 14:31           ` Stefan Hajnoczi
2013-05-29 15:59             ` Julian Stecklina
2013-05-28 11:58     ` Stefan Hajnoczi
2013-10-21 10:29       ` Luke Gorrie

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=51A4CE59.6060307@os.inf.tu-dresden.de \
    --to=jsteckli@os.inf.tu-dresden.de \
    --cc=qemu-devel@nongnu.org \
    --cc=snabb-devel@googlegroups.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).