[184471] in North American Network Operators' Group
Re: Inexpensive probes for automated bandwidth testing purposes
daemon@ATHENA.MIT.EDU (Jared Mauch)
Sat Oct 3 18:44:05 2015
X-Original-To: nanog@nanog.org
From: Jared Mauch <jared@puck.nether.net>
In-Reply-To: <37DBA43E-EE76-4323-962C-30BB988D0C2E@hathcock.org>
Date: Sat, 3 Oct 2015 18:43:58 -0400
To: Lorell Hathcock <lorell@hathcock.org>
Cc: nanog@nanog.org
Errors-To: nanog-bounces@nanog.org
If you are going to roll your own something like a raspberry PI would work. Y=
ou can also build your own measurements with a platform like ripe atlas. It a=
ll depends if you want to run iperf3 tests or simple smokeping type of stuff=
to correlate errors.=20
Jared Mauch
> On Oct 3, 2015, at 6:27 PM, Lorell Hathcock <lorell@hathcock.org> wrote:
>=20
> Greetings, NANOG. Happy Saturday to all.
>=20
> I am running a DOCSIS network that has a noisy cable plant. I want to be a=
ble to substantiate and quantify users' bandwidth issues. I would like a se=
t of inexpensive probes that I could place at selected customer's homes/busi=
nesses that would on a scheduled basis perform bandwidth tests.
>=20
> Likely I would need to place a server in the head end or across the intern=
et that would allow me to isolate and test certain network segments.
>=20
> I've looked into these in the past and was presented with some wonderfully=
expensive units that would duplicate my network problems into company finan=
cial problems as well.
>=20
> Any ideas? I know there are other ways to measure noise in the cable plan=
t and I am working on those as well. I will soon be running other, non-DOCS=
IS networks and need to have the same capabilities to test available bandwid=
th on those networks.
>=20
> Thanks!
>=20
> Lorell Hathcock
>=20
> Sent from my iPad
