From: Jesse Gross <jesse@nicira.com>
To: Simon Horman <horms@verge.net.au>
Cc: Eric Dumazet <eric.dumazet@gmail.com>,
Rusty Russell <rusty@rustcorp.com.au>,
virtualization@lists.linux-foundation.org, dev@openvswitch.org,
virtualization@lists.osdl.org, netdev@vger.kernel.org,
kvm@vger.kernel.org, "Michael S. Tsirkin" <mst@redhat.com>
Subject: Re: Flow Control and Port Mirroring Revisited
Date: Thu, 13 Jan 2011 10:45:38 -0500 [thread overview]
Message-ID: <AANLkTimO=5HmTJO1kmHGAWa-HTac+3d0TbrmJX5W4hVu@mail.gmail.com> (raw)
In-Reply-To: <20110113064718.GA17905@verge.net.au>
On Thu, Jan 13, 2011 at 1:47 AM, Simon Horman <horms@verge.net.au> wrote:
> On Mon, Jan 10, 2011 at 06:31:55PM +0900, Simon Horman wrote:
>> On Fri, Jan 07, 2011 at 10:23:58AM +0900, Simon Horman wrote:
>> > On Thu, Jan 06, 2011 at 05:38:01PM -0500, Jesse Gross wrote:
>> >
>> > [ snip ]
>> > >
>> > > I know that everyone likes a nice netperf result but I agree with
>> > > Michael that this probably isn't the right question to be asking. I
>> > > don't think that socket buffers are a real solution to the flow
>> > > control problem: they happen to provide that functionality but it's
>> > > more of a side effect than anything. It's just that the amount of
>> > > memory consumed by packets in the queue(s) doesn't really have any
>> > > implicit meaning for flow control (think multiple physical adapters,
>> > > all with the same speed instead of a virtual device and a physical
>> > > device with wildly different speeds). The analog in the physical
>> > > world that you're looking for would be Ethernet flow control.
>> > > Obviously, if the question is limiting CPU or memory consumption then
>> > > that's a different story.
>> >
>> > Point taken. I will see if I can control CPU (and thus memory) consumption
>> > using cgroups and/or tc.
>>
>> I have found that I can successfully control the throughput using
>> the following techniques
>>
>> 1) Place a tc egress filter on dummy0
>>
>> 2) Use ovs-ofctl to add a flow that sends skbs to dummy0 and then eth1,
>> this is effectively the same as one of my hacks to the datapath
>> that I mentioned in an earlier mail. The result is that eth1
>> "paces" the connection.
>
> Further to this, I wonder if there is any interest in providing
> a method to switch the action order - using ovs-ofctl is a hack imho -
> and/or switching the default action order for mirroring.
I'm not sure that there is a way to do this that is correct in the
generic case. It's possible that the destination could be a VM while
packets are being mirrored to a physical device or we could be
multicasting or some other arbitrarily complex scenario. Just think
of what a physical switch would do if it has ports with two different
speeds.
next prev parent reply other threads:[~2011-01-13 15:45 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-01-06 9:33 Flow Control and Port Mirroring Revisited Simon Horman
2011-01-06 10:22 ` Eric Dumazet
2011-01-06 12:44 ` Simon Horman
2011-01-06 13:28 ` Eric Dumazet
2011-01-06 22:01 ` Simon Horman
2011-01-06 22:38 ` Jesse Gross
2011-01-07 1:23 ` Simon Horman
2011-01-10 9:31 ` Simon Horman
2011-01-13 6:47 ` Simon Horman
2011-01-13 15:45 ` Jesse Gross [this message]
2011-01-13 23:41 ` Simon Horman
2011-01-14 4:58 ` Michael S. Tsirkin
2011-01-14 6:35 ` Simon Horman
2011-01-14 6:54 ` Michael S. Tsirkin
2011-01-16 22:37 ` Simon Horman
2011-01-16 23:56 ` Rusty Russell
2011-01-17 10:38 ` Michael S. Tsirkin
2011-01-17 10:26 ` Michael S. Tsirkin
2011-01-18 19:41 ` Rick Jones
2011-01-18 20:13 ` Michael S. Tsirkin
2011-01-18 21:28 ` Rick Jones
2011-01-19 9:11 ` Simon Horman
2011-01-20 8:38 ` Simon Horman
2011-01-21 2:30 ` Rick Jones
2011-01-21 9:59 ` Michael S. Tsirkin
2011-01-21 18:04 ` Rick Jones
2011-01-21 23:11 ` Simon Horman
2011-01-22 21:57 ` Michael S. Tsirkin
2011-01-23 6:38 ` Simon Horman
2011-01-23 10:39 ` Michael S. Tsirkin
2011-01-23 13:53 ` Simon Horman
2011-01-24 18:27 ` Rick Jones
2011-01-24 18:36 ` Michael S. Tsirkin
2011-01-24 19:01 ` Rick Jones
2011-01-24 19:42 ` Michael S. Tsirkin
2011-01-06 10:27 ` Michael S. Tsirkin
2011-01-06 11:30 ` Simon Horman
2011-01-06 12:07 ` Michael S. Tsirkin
2011-01-06 12:29 ` Simon Horman
2011-01-06 12:47 ` Michael S. Tsirkin
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='AANLkTimO=5HmTJO1kmHGAWa-HTac+3d0TbrmJX5W4hVu@mail.gmail.com' \
--to=jesse@nicira.com \
--cc=dev@openvswitch.org \
--cc=eric.dumazet@gmail.com \
--cc=horms@verge.net.au \
--cc=kvm@vger.kernel.org \
--cc=mst@redhat.com \
--cc=netdev@vger.kernel.org \
--cc=rusty@rustcorp.com.au \
--cc=virtualization@lists.linux-foundation.org \
--cc=virtualization@lists.osdl.org \
/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).