> On 25 Jul, 2020, at 10:35 pm, David P. Reed <dpr...@deepplum.com> wrote: > > And to be clear, AQM (cake, being an example) is not about bandwidth > allocation. It does focus on latency/queueing-delay, for the most part.
Cake is not *just* an AQM, though I understand your point. It is a qdisc with many interwoven functions. Cake's Diffserv support is neither a pure priority scheme nor a pure bandwidth allocation. By using a hybrid of the two for bandwidth allocation, I was hoping to avoid the main pitfalls that the simple Bell-headed approaches routinely encounter. Each tin also has its own AQM parameters, which feed into the distinction between high-throughput and low-latency classes of traffic. There are doubtless other approaches that could be tried, of course. And there might be endless debate over exactly how many traffic classes are actually needed; I don't think five is the right number, and the symmetry argument is not persuasive. But can we at least agree that Cake's attempt is a step in the right direction? - Jonathan Morton _______________________________________________ Cake mailing list Cake@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cake
