[c-nsp] Campus Network Design advice

[Chris Gauthier]

My previous employer had 14 sites with a building count ranging between 1 and 20. At each site (aka campus), we had 20 vlans or so. Though it was not a Cisco network, the concept is the same, we had an L3 chassis switch in the core as the campus router. From there, we extended L2 vlans to each building's distribution switch and downstream to each access switch. I was careful at the locations I directly managed to keep the vlans pruned and porganized so they touched as few switches as necessary. The campus backbone was 1GB Fiber and the core was based on a 300 Mbps fiber metro ring. We simply use L3 routed connections in the core and L2 at the distribution and access layers. Worked great. In the extended star topology that others implemented (before I came on board), the powers that be did not use STP. Generally, things were acceptable. I wanted RSTP, but that's a gripe for another day. OSPF was the routing algorithm. 2 of the L3 switches interfaced with the border routers. 

At my current employer, we are facing a similar situation as Marc, a flat network (with FAR fewer nodes) for 14 sites interconnected using T-1 circuits. I'm going to this same basic concept as above... a "standardized" ip addressing scheme and a standardization of vlan numbers at each site. We have very little at the access layer that needs anything special other than IP phones. 

Chris 

----- Original Message ----- 
From: "Marc Archer" <marc at archernet.id.au> 
To: cisco-nsp at puck.nether.net 
Sent: Tuesday, January 27, 2009 5:50:06 PM GMT -08:00 US/Canada Pacific 
Subject: [c-nsp] Campus Network Design advice 

Hi Guys, I'm looking for some advice on redesigning our campus network. We have around 2500 devices on our site which are spread across multiple buildings. At present the network runs on a (legacy) single flat VLAN which has caused us more than our fair share of headaches of late. Basically we are looking at 2 design options : The first option we have considered is to have a "router on a stick" at our core and trunk VLANS out to distribution switches in each building (and on to workgroup switches etc), leaving all routing to be done at the core. This would allow us to have all VLANS available in each building but I'm not sure if this is still going to be a problematic design (with VLANS extended all over the site). The other option we have been looking at (see attached) is to have L3 switches as all our distribution switches and contain VLANS to a particular building. This seems to be a neater solution to me, but I'm not sure of the best way to connect the distribution switches back to the core. I would also like to connect adjacent distribution switches together for redundancy, so I'm wondering if I should be looking at a heap of /30 links between distribution switches & the core (and run OSPF) - or just use a L2 network and let STP manage the links. I havent had much playtime on networks this size so any advice would be greatly appreciated. M. 
_______________________________________________ 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/