[190028] in North American Network Operators' Group
Re: Measuring the quality of Internet access
daemon@ATHENA.MIT.EDU (Joe Provo)
Mon Jun 13 17:23:12 2016
X-Original-To: nanog@nanog.org
Date: Mon, 13 Jun 2016 17:23:08 -0400
From: Joe Provo <nanog-post@rsuc.gweep.net>
To: Max Tulyev <maxtul@netassist.ua>
In-Reply-To: <96821.1465850332@turing-police.cc.vt.edu>
Reply-To: nanog-post@rsuc.gweep.net
Cc: NANOG list <nanog@nanog.org>
Errors-To: nanog-bounces@nanog.org
On Mon, Jun 13, 2016 at 04:38:52PM -0400, Valdis.Kletnieks@vt.edu wrote:
> On Mon, 13 Jun 2016 22:11:47 +0300, Max Tulyev said:
> > Is it possible in general to measure the quality of Internet access? And
> > if yes - how?
>
> First, *define* "quality". Raw bandwidth to a test server? Raw bandwidth
> to a weighted average of the Alexa Top 100? Does RTT/bufferbloat count?
> What about RTT jitter?
...not to mention the definition of "Internet". Aside from well-known
filters, see also
https://ripe72.ripe.net/presentations/6-2016-05-23-connectivity.pdf
and https://ripe72.ripe.net/archives/video/241/
Cheers,
Joe
--
RSUC / GweepNet / Spunk / FnB / CotSG / Usenix / NANOG
