[Codel] suggestions for enhancing AQM

[Codel] suggestions for enhancing AQM

[sahil grover]

[-- Attachment #1: Type: text/plain, Size: 648 bytes --]

Hi

For my research work ,i want to enhance AQM.

And i was thinking to improve CoDel by managing UDP flows because it
allocates disproportionate share of bandwidth to tcp's when misbehaving UDP
flows are there(transmitting data at very high rate).

And when i checked with sfq_CoDel,this problem is resolved. It maintains
the throughput of tcp's.

So can anybody tell me limitations of codel ,sfq_codel so that i can work
on it.

or any suggestions to make CoDel,Sfq_CoDel even better.

Please keep in mind, right now i can work only on ns2.

And i want  each and everyone to  share their ideas ,it would be  useful
for all

please reply.

Thanks

[-- Attachment #2: Type: text/html, Size: 917 bytes --]

Re: [Codel] suggestions for enhancing AQM

[Jonathan Morton]

[-- Attachment #1: Type: text/plain, Size: 895 bytes --]

Right now we consider AQM and Fair Queuing to be a solved problem, in terms
of designing and implementing algorithms. The open problem in this area is
DEPLOYMENT of these known solutions in the field. This is not strictly a
technical problem, but a political and economic one.

There are two related areas which we are still working on at a technical
level: Comprehensive Queue Management (CQM) which adds traffic
classification to the existing combination of AQM and FQ, and adapting
AQM/FQ/CQM to Wi-Fi networks, which have some unique characteristics which
require special consideration.

However, making progress in these two areas requires the people involved to
have understood the state of the art at a deep level before they can
usefully contribute. It also requires work with real traffic and
environments, or at least simulators considerably more advanced than ns2.

- Jonathan Morton

[-- Attachment #2: Type: text/html, Size: 971 bytes --]