[Bloat] AQM & Net Neutrality

[Stuart Cheshire]

On 24 May 2021, at 06:09, Livingood, Jason via Bloat <bloat at lists.bufferbloat.net> wrote:

> I’m looking for opinions here re bloat-busting techniques like AQM in the context of network neutrality (NN). The worry I have is whether some non-technical people will misunderstand how AQM works & conclude that implementing it may violate NN because it would make interactive traffic perform better than it does today. That is true of course – it’s a design goal of AQM, but non-interactive traffic performs as well as it always has – it is not disadvantaged.

There’s a faulty assumption buried in this, a common misunderstanding that we need to correct.

Consistently lower delay benefits *all* applications. The lower the round-trip time, the better TCP fast-retransmit works. TCP connection setup is faster. The TLS handshake is faster. The less buffering in the network, the less application-layer buffering is required for streaming video, giving quicker startup times, and quicker random access. If you try to make a list of applications that *don’t* benefit from lower delays, you’ll probably end up with an empty list.

This is why I’ve been advocating for making low delay available for *any* traffic that chooses to opt-in to this smarter queue management, not selectively for just some privileged traffic. I’m not making any subjective value judgement that video conferencing traffic is more important or more deserving that streaming video traffic, or weather forecasts, or driving directions, or software downloads. I do not support traffic prioritization schemes that privilege some traffic types over others. I support making low delay available to *all* traffic that agrees to behave properly and cooperate to keep the queues short -- meaning packet pacing, responding appropriately to congestion signals, etc.

Delay reduction is not an either/or choice. In order for some traffic to benefit other traffic doesn’t have to suffer. It’s not a zero-sum game. Eliminating standing queues in network buffers benefits all traffic. This can be hard to communicate because it seems counter to human intuition. It sounds too good to be true. In normal human life this is uncommon. When first class passengers board the plane first, all economy passengers wait a little bit longer as a result. Computer network queueing doesn’t operate like that, which makes it hard to explain by analogy to everyday experiences that most people understand.

I talked about this six years ago in my presentation at the Apple developer conference:
<https://developer.apple.com/videos/play/wwdc2015/719/?time=2702>

There’s also a neat demo a little earlier in that same video:
<https://developer.apple.com/videos/play/wwdc2015/719/?time=2520>

Stuart Cheshire