On Thu, 11 Jun 2015 23:03:18 +0200 Sebastian Moeller <[email protected]> wrote:

> On Jun 11, 2015, at 16:14 , Tristan Seligmann <[email protected]> wrote:
> 
> > On Thu, 11 Jun 2015 at 13:58 Sebastian Moeller <[email protected]> wrote:
> >
> > I sort of did, but it takes a relative long time to gather data
> > (typically around 3 hours) plus 15 minutes for the analysis. I am
> > happy to share this with interested parties...
> > 
> > I would definitely be interested in this for myself.
> 
>       Sure I will send it your way this weekend (need to do a tiny
> bit of clean-up and actually want to put it into git and possibly
> git-hub). It essentially is a shell script that runs ping and a
> matlab/octave script that does attempts the ATM detection, so far it
> worked well, but it is by no way automatic (yet)….

Please also send it my way. (But I don't have time to look at it before
*after* netfilter workshop[1])

One hint when detecting this, use the *minimum* observed latency/ping
for a given packet size, as jitter/other-traffic can only increase the
latency, but cannot lower it.  (This is an idea that I picked up from
Van Jacobsson).

-- 
Best regards,
  Jesper Dangaard Brouer
  MSc.CS, Sr. Network Kernel Developer at Red Hat
  Author of http://www.iptv-analyzer.org
  LinkedIn: http://www.linkedin.com/in/brouer


[1] http://workshop.netfilter.org/2015/
_______________________________________________
Bloat mailing list
[email protected]
https://lists.bufferbloat.net/listinfo/bloat

Reply via email to