[Make-wifi-fast] [RFC/RFT] mac80211: implement fq_codel for software queuing

[Aaron Wood]

On Mon, Mar 21, 2016 at 6:29 PM, David Lang <david at lang.hm> wrote:

> On Mon, 29 Feb 2016, Michal Kazior wrote:
>
> Our intent is to continue to improve the flent test suite to be able
>>> to generate repeatable tests, track relevant wifi behaviors and pull
>>> relevant data back, graphed over time (of test) and time (over test
>>> runs). A problem with udp flood tests is that tcp traffic is always
>>> bidirectional (data vs acks), so a naive thought would be, that yes,
>>> you should get half the bandwidth you get with a udp flood test.
>>>
>>
>> I don't see why you'd be doomed to get only half the bandwidth because
>> of that? Sure, Wi-Fi is half-duplex but transmit time for ACKs is a
>> lot smaller than transmit time for the data.
>>
>
> The difference is actually far less than you think. Each transmission has
> a fixed-length header and quiet times that were designed in the days of
> 802.11b (1-11Mb) and if you are transmitting a wide 802.11ac signal at a
> couple hundred Mb, you can find that the time taken to transmit even full
> packets is a surprisingly small percentage of the total transmit time.
>
> David Lang


A 2-dimensional display of data sent vs. time might be useful, for a couple
packets, to help explain this (although it may need to be at log-scale).
X-axis is time, Y is bandwidth being sent.

-Aaron
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.bufferbloat.net/pipermail/make-wifi-fast/attachments/20160321/046439e7/attachment.html>