[Make-wifi-fast] Fwd: Re: berlin?

[Dave Taht <dave.taht@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 1666 bytes --]

---------- Forwarded message ----------
From: Simon Wunderlich <sw@simonwunderlich.de>
Date: Mon, Aug 31, 2015 at 3:09 AM
Subject: Re: Re: berlin?
To: Dave Taht <dave.taht@gmail.com>


Hi Dave,

hope you had a great trip so far! Here is that whitepaper I was talking about.
It is somewhat outdated already obviously, but maybe it will be interesting
(at least our test setup looks nice ;] ). We have created an ath9k patch to
account used txtime per station (including retries etc). The scheduling
function is then adapted to send to the station with the greatest airtime
deficit instead of just doing round robin. The missing communication between
the netdev layer and ath9k layer is also described here which made the concept
fail, but this could have been solved already with that new hook you
described.

Anyway, this research has been done for Open-Mesh but never got implemented in
a product so far. If there is interest and funds, I would be happy to continue
on that - I'm sure we can (re)use that patchset or our "wall of death" test
setup, I'll talk to Open-Mesh but they already did let me know earlier last
year that this would be possible.

Enjoy your trip,
    Simon

On Friday 21 August 2015 12:04:22 Dave Taht wrote:
> > BTW, I'm still trying to get the latest version of that fair airtime
> > analysis/summary we made, I'll send it to you as soon as I got it.
>
> TIA!
>
> > Cheers,
> >
> >     Simon
> >
> > On Friday 21 August 2015 09:24:48 Dave Taht wrote:
> >> I am heading to berlin on the 28th. Where are you based?


-- 
Dave Täht
endo is a terrible disease: http://www.gofundme.com/SummerVsEndo

[-- Attachment #2: Whitepaper-airtime fairness.pdf --]
[-- Type: application/pdf, Size: 914114 bytes --]

[-- Attachment #3: signature.asc --]
[-- Type: application/pgp-signature, Size: 188 bytes --]