[181357] in North American Network Operators' Group
Re: Data Center Network Monitoring with TAPs
daemon@ATHENA.MIT.EDU (Mel Beckman)
Mon Jun 22 01:06:00 2015
X-Original-To: nanog@nanog.org
From: Mel Beckman <mel@beckman.org>
To: Mitch Howards <hbf9121@hotmail.com>
Date: Mon, 22 Jun 2015 05:04:19 +0000
In-Reply-To: <BAY173-W44D926F4F6C2287340E014A8A30@phx.gbl>
Cc: "nanog@nanog.org" <nanog@nanog.org>
Errors-To: nanog-bounces@nanog.org
Ultimately this is one of the things that SDN schemes such as OpenFlow brin=
g a data center for free. Distributed flow statistics collection through Oe=
nFlow's extensible infrastructure gives you a huge range of reporting and a=
nalysis capabilities, with no taps needed. Every network port is in essence=
a tap.
Here's an interesting paper on one open source OF tool:
https://www.nas.ewi.tudelft.nl/people/Fernando/papers/MonitoringOpenFlow.pd=
f
-mel beckman
> On Jun 21, 2015, at 9:50 PM, Mitch Howards <hbf9121@hotmail.com> wrote:
>=20
> Hello All,=20
>=20
> Was wondering what folks are using to monitor traffic
> on their networks. Looking into Ixia and APCON devices for dedup and=20
> other filtering features as well as passive fiber TAPs to capture the=20
> traffic.=20
>=20
> How are folks handling TAP'ing large data center=20
> networks? TAPs at the "distribution layer" would be the best fit for my=20
> network but that would require a ton of passive fiber TAPs for the=20
> incoming fibers to the distribution switches. The end goal is to not=20
> only capture the north-south traffic on the network but also east-west=20
> traffic. It seems more efficient to just use SPANs but there are many=20
> limitations using SPANs.=20
>=20
> Thanks in advance for any suggestions.
>=20
> Mitch =20
