Re: [Cake] Configuring cake for VDSL2 bridged connection

[Alan Jenkins <alan.christopher.jenkins@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 3081 bytes --]

On 26/08/16 12:29, moeller0 wrote:
> Hi techicist,
>
>> On Aug 26, 2016, at 13:15 , techicist@gmail.com wrote:
>>
>> Is flowblind likely to give better performance?
> 	That depends on your definition of better, I guess. Typically flow-fair queuing seems to be what most people prefer (unless an application either does not respond to AQM signals or open an excessive amount of individual flows flow-fair queueing effectively treats most traffic sources equal, pretty much what people seem to want, add to this a bit of classification to exempt e.g. VOIP traffic from only getting its flow-fair share of the bandwidth and the whole thing also works reasonably well with slow links). People suffering from unruly applications (like mis-configured? bit-torrent clients or recently windows update) often ask for per-application fairness, but that is not something a router will ever be able to deliver in my opinion; the closest we get to this would be fairnes by internal or external end-IP addresses. Luckily cake offers just these modes “dsthost”, “srchost” and even better offers a combination modes that will on a first level attempt per host-IP fairness and within each host IP also per-flow fairness (“dual-srchost” and “dual-dsthost”, and even “triple-isolate” which systematically might be better called “dual-srchost-dsthost” since it offers fist level fairness based on an under-documented mix of src and dst addresses, but I digress). Please note that on a typical homerouter, due to NAT, all the IP addressed based fairness modes will not work for IPv4 on the wan interface, IPv6 traffic should be fine, but IPv4 basically degrades into a computationally more intensive version of flow-fairness (as after NAT cake only sees the routers external IP for all internal hosts). This might have been more than you wanted to know…
>
> Best Regards
> 	Sebastian

flowblind is an option for testing purposes or advanced use cases. The 
design goal for Cake is to avoid understanding and fiddling with options 
to get good performance for common cases.

If you try enabling flowblind, your latency under load will jump by 
5ms+.  "Head of line blocking".  A full queue will be 5ms.  This will 
delay flows which do not need a full fair share of the link, like VOIP 
or gaming.  Lower latency is  better for VOIP or gaming.

You should find this is small compared to the latency increase under 
load without cake.  You wouldn't notice it in web browsing. (Frankly I 
don't seem to notice 100ms extra latency in web browsing.

I run fq_codel for similar performance to cake, mainly to increase my 
confidence that torrent uploads don't have noticable effects for other 
household users.  Torrent downloads still suck, but I haven't seen any 
Cake results promoted on that basis.  It either needs to be fixed at the 
ISP end, or in the torrent software.  QUIC are emulating the 
competitiveness of 2x TCP flows in a single UDP flow. BT should be able 
to emulate half a TCP flow when downloading from two peers simultaneously).

[-- Attachment #2: Type: text/html, Size: 3634 bytes --]