[177821] in North American Network Operators' Group
RE: mpls over microwave
daemon@ATHENA.MIT.EDU (Daniel Hooper)
Fri Feb 6 01:40:14 2015
X-Original-To: nanog@nanog.org
From: Daniel Hooper <dhooper@gold.net.au>
To: "nanog@nanog.org" <nanog@nanog.org>
Date: Fri, 6 Feb 2015 02:46:54 +0000
In-Reply-To: <5F596032-4D9C-4FBE-B2B2-FB9576EBE6B7@medline.com>
Errors-To: nanog-bounces@nanog.org
Aviat Networks has recently released a microwave router with MPLS features,=
it's basically a router inside the microwave indoor unit.
What we've found what hurts with anything over microwave is when you're run=
ning n+n links over long haul and the RF modulation steps down on one of th=
ose links and decreases your bandwidth, the Aviat solution is apparently me=
ant to solve this with the RF cards talking directly to the on-board router=
and giving your MPLS nodes a kick to shift traffic in the right direction.
-Daniel
-----Original Message-----
From: NANOG [mailto:nanog-bounces@nanog.org] On Behalf Of Naslund, Steve
Sent: Friday, 6 February 2015 10:39 AM
To: nanog@nanog.org
Subject: Re: mpls over microwave
We run Dragonwave systems and have no issues at all. MPLS in itself doesn'=
t make a difference since the gear is a straight Ethernet link. Just make =
sure your gear handles your frame sizes and tagging and you should be good.=
=20
As long as your radio link is engineered right you should have high reliabi=
lity. The key is having enough margin to maintain links during fades. So =
for example our link runs at -34 dbm and our receivers are good down to abo=
ut -65 dbm at this rate. That gives us a roughly 35 dbm margin for degrada=
tion before the modems will change modulation a to a lower speed. Here in =
chicago we have seen maybe a total of an hour of weather related fade in a =
10 years period on a 20 mile link running 600 Mbps. They are very popular=
for low latency since they actually have less latency than fiber. The hig=
h frequency traders pay big bucks to get on the microwaves between markets =
because of that trait. Microwaves through air are faster than photons in a=
fiber cable.
Steven Naslund
Chicago IL
Sent from my iPhone
> On Feb 5, 2015, at 4:03 PM, Mike Hammett <nanog@ics-il.net> wrote:
>=20
> Shouldn't really be any different as long as your gear supports the appro=
priate MTUs.=20
>=20
>=20
>=20
>=20
> -----
> Mike Hammett
> Intelligent Computing Solutions
> http://www.ics-il.com
>=20
>=20
>=20
> ----- Original Message -----
>=20
> From: "Scott Weeks" <surfer@mauigateway.com>
> To: nanog@nanog.org
> Sent: Thursday, February 5, 2015 3:55:04 PM
> Subject: mpls over microwave
>=20
>=20
>=20
> Anyone doing MPLS over microwave radios? Please share your experiences=20
> on list or off.
>=20
> scott
>=20
