There are a few folks using edgerouter infinity, then tend to be rather quiet.. 
some of them can be found on Facebook groups.
I have not heard of any bgp convergence performance issue with the edgerouter 
infinity
I have one, it is not in production yet, due to time constraints.
Baltic Networks / Brian & Josh had a test shoot out setup for bgp convergence 
test at the last WISPAmerican Show..
 (I thought they had down either a presentation or a video on it.. contact them 
ask them for the results..)

With what I know, I will be very surprise to see an Edgerouter infinity taking 
15 min to re-converge..
(unless someone is confusing it with a Mikrotik CCR bgp performance )
The only gripe with the edgerouters is lacking support for mpls,vpls, etc ...

The X86 based Mikrotik perform very well for bgp convergence, however they 
don't have the ability to compete with dealing with high pps (packets per 
second) of traffic as the MT CCR's and Edgerouter Infnity can.

Take all of (everyone's) Top end specs with a grain of salt.

Regards.


Faisal Imtiaz
Snappy Internet & Telecom
http://www.snappytelecom.net

Tel: 305 663 5518 x 232

Help-desk: (305)663-5518 Option 2 or Email: [email protected]

----- Original Message -----
> From: "Tim Densmore" <[email protected]>
> To: "Ubiquiti Users Group" <[email protected]>
> Sent: Wednesday, May 2, 2018 5:45:22 PM
> Subject: [Ubnt_users] Edgerouter Infinity BGP

> Hi Folks,
> 
> Can anyone share experiences with the er infinity in regards to BGP
> performance?  I've read several places that it'll take several full
> feeds with no problem, which is obviously great.  However, the few
> reports I've seen that discuss convergence performance are fairly
> pessimal.  Can anyone shed light on their real world experience with
> running the er infinity in an environment with eBGP, iBGP, and OSPF?  Is
> it actually taking 15 min to re-converge after a peer is lost?  That
> seems absurdly long to me, but I have zero experience in the $1500-3000
> BGP router market so maybe my expectations are unreasonable.
> 
> Thanks,
> Tim Densmore
> 
> _______________________________________________
> Ubnt_users mailing list
> [email protected]
> http://lists.wispa.org/mailman/listinfo/ubnt_users
_______________________________________________
Ubnt_users mailing list
[email protected]
http://lists.wispa.org/mailman/listinfo/ubnt_users

Reply via email to