[Babel-users] 64k routes, bird, babel rfc, rtod, etc

[Dave Taht]

On Thu, Nov 8, 2018 at 1:51 PM Toke Høiland-Jørgensen <toke at toke.dk> wrote:
>
> Dave Taht <dave.taht at gmail.com> writes:
>
> > On Thu, Nov 8, 2018 at 1:40 PM Dave Taht <dave.taht at gmail.com> wrote:
> >>
> >> On Thu, Nov 8, 2018 at 1:21 PM Juliusz Chroboczek <jch at irif.fr> wrote:
> >> >
> >> > > things fall apart at about 32k in this case.
> >> >
> >> > That's pretty good.  We'll reach your 64k target when we get around to
> >>
> >> It's a worthy goal, but really excessive. :) I basically wanted to
> >> know the limits
> >> of the lowest end hw I had, for real, before daring to try and deploy
> >> ipv6 willy nilly again.
> >>
> >> > fixing the resend data structure
> >>
> >> Got any ideas? I enjoyed poking into the latest trie research (source
> >> specific routing is a new area!) and I really enjoyed learning about
> >> skiplists last week, am liking the uthash potential long term...
> >>
> >> > and implement pacing in the buffer
> >> > flushing code.
> >>
> >> Pacing can be done at the socket layer now and it's been working for
> >> udp and QUIC since, oh, about 4.12? 4.9? can't
> >> remember. I tried this at one point. Easy patch to fiddle with to
> >> "just turn it on"
> >>
> >> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=62748f32d501f5d3712a7c372bbb92abc7c62bc7
> >
> > This no longer requires the fq scheduler, btw. Problem is (as the quic
> > folk are experiencing) is that the kernel doesn't know
> > (at least not with public patches) that you can send udp at one rate
> > one way, and another at another. tcp can do it per connection.
>
> There is also a socket option that allows you to set an "earliest send
> time" on each packet... Can't remember what it's called, but it ties
> into Van's timer wheel idea...

The whole tcp stack got switched over to the core idea in a recent commit.

It's not a sockopt, but a per packet cmsg elsewhere

See:

https://www.mail-archive.com/netdev@vger.kernel.org/msg253554.html

I think it requires the new sch_etx or hardware support (publically,
that's a few
intel cards) and a software nic, presently.

I really am also in love with the whole timerwheel idea in a lot of
cases, as one
of my hopes was to resume work on the "bobbie" policer, but this method can only
work on host-sourced packets. sigh. ddpk here I come...

In babel's case... it might in the long term be a way to pace, or
schedule hellos more in advance,
and or/ the quic driven atomic bind and connect per connection idea:

https://www.mail-archive.com/netdev@vger.kernel.org/msg253530.html

>
> -Toke



-- 

Dave Täht
CTO, TekLibre, LLC
http://www.teklibre.com
Tel: 1-831-205-9740