Historic archive of defunct list bismark-devel@lists.bufferbloat.net
 help / color / mirror / Atom feed
From: Dave Taht <dave.taht@gmail.com>
To: Nick Feamster <feamster@cc.gatech.edu>
Cc: bismark-devel@lists.bufferbloat.net
Subject: Re: [Bismark-devel] incorporating openflow
Date: Wed, 4 May 2011 10:46:58 -0600	[thread overview]
Message-ID: <BANLkTin+YcLMv4qYL_u3pMx2Yq6=jbt7tg@mail.gmail.com> (raw)
In-Reply-To: <398DDA0C-771B-4769-A643-E3CBFC3863D8@cc.gatech.edu>

[-- Attachment #1: Type: text/plain, Size: 2393 bytes --]

OK I compiled it. It built as a package.

On Wed, May 4, 2011 at 10:16 AM, Nick Feamster <feamster@cc.gatech.edu>wrote:

> As Srikanth said, not needed for capetown.
>
> Some possibilities we've been thinking about using it for:
>
> * enforcing per-user or per-device usage caps (in capped settings)
>

In this particular case, using SFB on a per IP address basis may work.

(not that SFB is in the build)


> * rate-limiting and QoS functionality
>

I incorporated the QoS scripts which use tc and the hsfc QoS algorithm + SFQ
to do that job. It works well in iscwrt, in combination with a lowered
txqueue (to 16), I observed latencies peaking at 66ms on a DSL connection
that can't do better than 26ms. (vs over half a second)

(these are disabled by default (I think) as they will mess up your
measurements)

But do try 'em out!

They seem pretty good on the outgoing interface. To do rate limitation on a
per user basis in conjunction with that I would think about SFB on the inner
interfaces.

So I'm back to wondering what to do, exactly, with openflow.

If you aren't already aware of the iscwrt subproject, see:

http://www.bufferbloat.net/projects/iscwrt/wiki

One of the principal maintainers of bind (Evan Hunt) is hacking on it - even
running it now using the wndr3700v2 as his main router.

* remote management, troubleshooting, or security services.  See this, for
> example:
>        http://www.gtnoise.net/papers/2010/feamster:homenets2010.pdf
>
>
nagios + snmp are common for some of that.

I'll read the paper on the plane.


> I'd love to get this into the build sometime soon-ish, so I could encourage
> people to use it in my class this coming fall. :-)
>
> -Nick
>
>
> On May 4, 2011, at 11:33 AM, Dave Taht wrote:
>
> > 3) What is the need for openflow? I am meeting with the folk at stanford
> at 2PM PDT and although I get what openflow is for, I'm not sure of the
> intended use. Is it needed for capetown?
> >
> >
> > --
> > Dave Täht
> > SKYPE: davetaht
> > US Tel: 1-239-829-5608
> > http://the-edge.blogspot.com
> > _______________________________________________
> > Bismark-devel mailing list
> > Bismark-devel@lists.bufferbloat.net
> > https://lists.bufferbloat.net/listinfo/bismark-devel
>
>


-- 
Dave Täht
SKYPE: davetaht
US Tel: 1-239-829-5608
http://the-edge.blogspot.com

[-- Attachment #2: Type: text/html, Size: 3891 bytes --]

      parent reply	other threads:[~2011-05-04 16:42 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-05-04 15:33 Dave Taht
2011-05-04 15:36 ` Srikanth Sundaresan
2011-05-04 16:16 ` Nick Feamster
2011-05-04 16:38   ` Dave Taht
2011-05-04 16:46   ` Dave Taht [this message]

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='BANLkTin+YcLMv4qYL_u3pMx2Yq6=jbt7tg@mail.gmail.com' \
    --to=dave.taht@gmail.com \
    --cc=bismark-devel@lists.bufferbloat.net \
    --cc=feamster@cc.gatech.edu \
    /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