[c-nsp] acl on bvi in ios xr (9k) 4.1.2

[Mack McBride]

For the older boxes that aren't getting larger tcam, FIB compression is the only way to avoid fork lift upgrades to less than optimal boxes.

In a mixed colo managed hosting environment, we need an device that will support RSTP.
MSTP is not an option as changes within a customer environment can affect other customer environments.

7K mpls support is still incomplete.  

The routing table is expected to exceed 640K in between 18 and 24 months.
The routing table growth is expected to go exponential as IPv4 space is exhausted and larger blocks are broken down and sold off.
The actual growth curve is hard to predict but expect over 1 million routes within 5 years.

Mack

-----Original Message-----
From: cisco-nsp-bounces at puck.nether.net [mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Lukasz Bromirski
Sent: Saturday, July 21, 2012 9:28 AM
To: cisco-nsp at puck.nether.net
Subject: Re: [c-nsp] acl on bvi in ios xr (9k) 4.1.2

On 7/19/12 9:45 PM, Mack McBride wrote:

> If cisco deploys FIB compression it might solve some of those
 > concerns but the feedback is that the development on that has stopped  > or is at least not on the road map.

Why? FIB compression is bad. For convergence, for scalability, for anything other than synthetic testing. Some competitors learned this in a hard way, customers also.

> The ASR 9K is a great box but without decent switching support and
 > rapid-pvst it doesn't work well in a managed services/colocation  > environment.

"decent switching support" means...?

ASR 9k supports MST which uses the RSTP. You can also do multi-chassis LAG for redundancy if needed to external boxes.

> The Nexus 7K is stuck at the same place the 6500 and 7600 are
 > routing table wise and lack of MPLS support is still a concern.

7k offers MPLS support:
http://www.cisco.com/en/US/docs/switches/datacenter/sw/5_x/nx-os/mpls/configuration/guide/mp_mpls_overview.html

In terms of FIB size, what is the size you need?

> The Juniper MX series can handle switching and rapid-pvst and upwards 
> to 4 million routes (usual division of IPv4/IPv6 applies but is dynamic) and we are currently testing it for a replacement for the 6500/7600.

Good luck.

-- 
"There's no sense in being precise when |               Łukasz Bromirski
  you don't know what you're talking     |      jid:lbromirski at jabber.org
  about."               John von Neumann |    http://lukasz.bromirski.net
_______________________________________________
cisco-nsp mailing list  cisco-nsp at puck.nether.net https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/