[Bloat] Does VDSL interleaving+FEC help bufferbloat?

[Bloat] Does VDSL interleaving+FEC help bufferbloat?

[cloneman]

[-- Attachment #1: Type: text/plain, Size: 1352 bytes --]

Interleaving on DSL is enabled by many providers which allegedly provides
some packet loss protection at the cost of some latency.

I think it's designed mainly Layer 1 line noise, but now I'm wondering if
it might also translate to other benefits, even on a clean DSL line that
doesn't *need* interleave.

Question:
Is interleaving protective of packet loss due to momentary spikes on the
downloads?

I tested it a bit  with the help of my ISP

VDSL:
Interleave+FEC On (15ms first hop) 3% packet loss on steam download stress
Interleave+FEC OFF (4ms first hop) 5-6% packet loss on steam download
stress

With fq_codel: (-15% bw)
Both had 0.3% packet loss (very good) on steam Download stress

Of note, steam downloads used to break fq_codel for me (20 flows split
evenly is how they designed it *sigh*) but it appears that it's working
well now, perhaps steam *finally* fixed their stuff. Steam stress is
performed by downloading any of their free games with their client software
e.g. Dota2.

tl-dr; interesting to test weather interleave+FEC, a technology designed
for layer 1 noise, can help with layer 3 ingress contention for resources.

Also, I found a guy who designed a layer 3 FEC protocol designed for
terrible connections: https://github.com/wangyu-/tinyfecVPN

If anyone wants to comment on that, I think it's nice to know that it
exists.

[-- Attachment #2: Type: text/html, Size: 2226 bytes --]

Re: [Bloat] Does VDSL interleaving+FEC help bufferbloat?

[Jonathan Morton]

> On 3 Jan, 2019, at 11:38 pm, cloneman <bufferbloat@flamingpc.com> wrote:
> 
> tl-dr; interesting to test weather interleave+FEC, a technology designed for layer 1 noise, can help with layer 3 ingress contention for resources.

I'm reasonably sure it has no effect on bufferbloat.  You're not even trying to measure latency in your tests so far, only packet loss, which is completely different.  Interleave would tend to increase baseline latency, but probably has no effect on the queue.

Only changing the queue length, or inserting an AQM (which shortens the *effective* queue length) would affect bufferbloat.

 - Jonathan Morton

Re: [Bloat] Does VDSL interleaving+FEC help bufferbloat?

[Simon Barber]

Increasing fixed latency can help avoid TCP hitting RTO.

Simon


> On Jan 3, 2019, at 8:01 PM, Jonathan Morton <chromatix99@gmail.com> wrote:
> 

Re: [Bloat] Does VDSL interleaving+FEC help bufferbloat?

[Dave Taht]

dsl interleave was added primarily to make multicast udp tv streams
work better (as they are very intolerant of packet loss). Often (as in
free's implementation) these streams are "invisible" to the overlying
IP applications. It typically adds at least 6ms of delay to an already
slow technology.

Do measure unloaded latency with and without it on, it's been a long
time since I fiddled with it....

Re: [Bloat] Does VDSL interleaving+FEC help bufferbloat?

[Mikael Abrahamsson]

On Fri, 4 Jan 2019, Dave Taht wrote:

> dsl interleave was added primarily to make multicast udp tv streams work 
> better (as they are very intolerant of packet loss). Often (as in free's 
> implementation) these streams are "invisible" to the overlying IP 
> applications. It typically adds at least 6ms of delay to an already slow 
> technology.

ADSL2+ is very prone to short bursts of interference, so setting no 
interleaving means quite high packet loss. Setting interleaving to 16ms 
means FEC has a much better chance of correcting errors and thus reduce 
packet loss.

At several jobs ago we actually had several different profiles for 
customers, they could choose 1, 4 or 16ms interlaving depending on their 
needs for gaming etc. The 1 and 4 ms interleaving targets had different 
SNR margin targets so they were sacrificing speed for lower latency, 
because that's the tradeoff you basically have to do with normal L4 
protocols that end customers typically use.

-- 
Mikael Abrahamsson    email: swmike@swm.pp.se

Re: [Bloat] Does VDSL interleaving+FEC help bufferbloat?

[Jan Ceuleers]

On 04/01/2019 18:10, Dave Taht wrote:
> dsl interleave was added primarily to make multicast udp tv streams
> work better (as they are very intolerant of packet loss). Often (as in
> free's implementation) these streams are "invisible" to the overlying
> IP applications. It typically adds at least 6ms of delay to an already
> slow technology.
> 
> Do measure unloaded latency with and without it on, it's been a long
> time since I fiddled with it....

Interleaving isn't a parameter that mere end-users* can influence - it
is part of the line's profile, and that's determined by the network
operator.

*: you know, the people who pay the bills...