On Sun, 9 Aug 2015, Simon Barber wrote:

11ac with it's always on RTS/CTS and mandatory A-MPDUs really performs much better than many 11n implementations. It's finally a fairly sane MAC. The 64 bit limit in the compressed block ack is a limiter though. The really wide channels are another complex area for busy networks.

The aggregation overhead for 11ac is about 222uS, counting EDCA backoff, RTS, CTS, PLCP header and the Block-ACK and all the inter frame spaces. One full size ethernet frame at 1Gb/s = ~12us. Large aggregates are critical to good efficiency and performance, and a certain amount of queuing is required to form them. They have to be completely formed before transmission starts.

do you know the per-transmission overhead for different modes (n and a/g specifically)? Also, what parts of the overhead get extended when the data rate slows? It's all well and good to talk about a full size packet being 12us at 1GHz, but that requires a good signal an 3x3 radios. If instead you are on a 1x1 radio with not as good a signal, you can easily drop your data rate by an order of magnatude or so. At ~100Mb your data packet is now 120us, what is the overhead? if you drop to 10Mb your packet is now 1200us, what is the overhead.

David Lang

Simon

On 8/9/2015 12:31 PM, Jonathan Morton wrote:

The question of whether to aggregate under congested conditions is controversial, probably because it depends on complex conditions.  There are arguments both for and against.

It may be worth considering it as a risk/reward tradeoff.  Given N packets (which for brevity I'll assume are equal MTU sized), the reward is obviously proportional to N. Risk however is calculated as probability * consequence.

Assuming all packets in the aggregate are lost on collision, the risk of collision scales with L*N, where L is N plus the overhead of the TXOP. Under that argument, usually you should not aggregate if the probability of collision is high.

However, if only one packet is lost due to collision with, for example, a small RTS probe which is not answered, the risk scales with L, which is sublinear compared to the reward relative to the amount of aggregation (especially at high data rates where the TXOP overhead is substantial). Under this assumption, aggregation is usually profitable even with a high collision probability, and results in overall higher efficiency whether or not collisions are likely.

This is the difference between the typical 802.11n situation (one checksum per aggregate) and the mandatory 802.11ac capability of a checksum per packet.  As long as you also employ RTS/CTS when appropriate, the possibility of collisions is no longer a reason to avoid aggregating.

- Jonathan Morton



_______________________________________________
Make-wifi-fast mailing list
Make-wifi-fast@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/make-wifi-fast

_______________________________________________

Make-wifi-fast mailing list

Make-wifi-fast@lists.bufferbloat.net

https://lists.bufferbloat.net/listinfo/make-wifi-fast