[Cerowrt-devel] Perfection vs. Good Enough

[David Lang]

On Sat, 11 Jan 2014, Sebastian Moeller wrote:

>>> Compared to the orders of magnitude we already get from fq codel, the sum benefit
>>> of these [Link Layer Adaptation] fixes is in the very small percentage points.
>
> 	I do not agree with this sentiment, as I understood Dave was talking 
> about different modifications to fq_codel (nfq_codel and efq_codel), this was 
> not about the link layer; for an ATM link if you get the link layer wrong the 
> shaper does at best work stochastically; and if the shaper does not work well 
> we are back at square one: badly managed buffers out of our control filling up 
> causing delays worth seconds. So unless you shape down to ~50% of link rate, 
> you will get at least temporary buffer bloat on an ATM link, unless you take 
> all the ATM peculiarities into account (basically what link layer ATM is 
> doing).

the question boils down to

compared to stock firmware,

how much of a beneifit is openwrt trunk (and how risky)
how much of a benefit is cerowrt without the link-level tuning

how much additional benefit do we expect to get from the added work.

What we have now is a huge step forward compared to stock firmware, a 
significant portion of this benefit has been pushed upstream to openwrt.

but right now there is a lot of work to make things perfect, and as a result, 
people are stuck using stock firmware or openwrt releases (unless they compile 
their own image from trunk), and so they are missing everything that's been 
done.

There is a lot of value in getting another release out that brings all the gains 
that we have now to everyone. Then development can continue trying to optimize 
things more.

David Lang