[Codel] Codel, fq_codel in ns3

[Codel] Codel, fq_codel in ns3

[Dave Taht]

Andrew Mcgregor has developed some ns3 models of Kathie Nichols' and
Van Jacobson's new codel algorithm and also incorporated Eric
Dumazet's fq_codel derivative...

VJ's ietf talk last week on codel and fq_codel is here:

http://recordings.conf.meetecho.com/Recordings/watch.jsp?recording=IETF84_TSVAREA&chapter=part_3

we just rebased the ns3 versions against current ns3 head,
and exported it via hg-git to github...

https://github.com/dtaht/ns-3-dev

Certain parts of the algorithms remain under heavy development, but on
the whole results are good.

ideas on how to best move forward and sideways with this are welcomed.
I would argue we need some time out of tree to bake further.

(is anyone working on ecn in ns3?)

(plug: codel is being tracked on the codel mailing list on
lists.bufferbloat.net)

-- 
Dave Täht
http://www.bufferbloat.net/projects/cerowrt/wiki - "3.3.8-6 is out
with fq_codel!"

Re: [Codel] [Ns-developers] Codel, fq_codel in ns3

[Tom Henderson]

On 08/10/2012 01:32 PM, Dave Taht wrote:
> Andrew Mcgregor has developed some ns3 models of Kathie Nichols' and
> Van Jacobson's new codel algorithm and also incorporated Eric
> Dumazet's fq_codel derivative...
>
> VJ's ietf talk last week on codel and fq_codel is here:
>
> http://recordings.conf.meetecho.com/Recordings/watch.jsp?recording=IETF84_TSVAREA&chapter=part_3
>
> we just rebased the ns3 versions against current ns3 head,
> and exported it via hg-git to github...
>
> https://github.com/dtaht/ns-3-dev
>
> Certain parts of the algorithms remain under heavy development, but on
> the whole results are good.
>
> ideas on how to best move forward and sideways with this are welcomed.
> I would argue we need some time out of tree to bake further.

Dave, this looks interesting; to encourage some reviews, would you mind 
posting the patch on Rietveld, where we do our code reviews?

http://code.google.com/p/rietveld/

>
> (is anyone working on ecn in ns3?)

Not that I'm aware of.

- Tom

>
> (plug: codel is being tracked on the codel mailing list on
> lists.bufferbloat.net)
>

Re: [Codel] [Ns-developers] Codel, fq_codel in ns3

[Lalith Suresh]

On Fri, Aug 10, 2012 at 11:32 PM, Dave Taht <dave.taht@gmail.com> wrote:
> Andrew Mcgregor has developed some ns3 models of Kathie Nichols' and
> Van Jacobson's new codel algorithm and also incorporated Eric
> Dumazet's fq_codel derivative...
>
> VJ's ietf talk last week on codel and fq_codel is here:
>
> http://recordings.conf.meetecho.com/Recordings/watch.jsp?recording=IETF84_TSVAREA&chapter=part_3
>
> we just rebased the ns3 versions against current ns3 head,
> and exported it via hg-git to github...
>
> https://github.com/dtaht/ns-3-dev
>
> Certain parts of the algorithms remain under heavy development, but on
> the whole results are good.
>
> ideas on how to best move forward and sideways with this are welcomed.
> I would argue we need some time out of tree to bake further.
>

To add to what Tom said, check this link for using Rietveld:
http://www.nsnam.org/developers/tools/rietveld/



> (is anyone working on ecn in ns3?)
>
> (plug: codel is being tracked on the codel mailing list on
> lists.bufferbloat.net)
>
> --
> Dave Täht
> http://www.bufferbloat.net/projects/cerowrt/wiki - "3.3.8-6 is out
> with fq_codel!"
>



-- 
Lalith Suresh
www.lalith.in

Re: [Codel] [Ns-developers] Codel, fq_codel in ns3

[Tom Henderson]

>
> Dave, this looks interesting; to encourage some reviews, would you mind
> posting the patch on Rietveld, where we do our code reviews?
>
> http://code.google.com/p/rietveld/

At Dave's request, I posted the issue here:

http://codereview.appspot.com/6463048/

and added ns-3-reviews@googlegroups.com to the cc list.

- Tom