Campus LAN Design

I have a Campus LAN that has 8 distibution blocks.  The 7th and 8th distribution block belong to the the WAN Edge and Server Block.  The other 6 are user access distribution blocks.  The way I have it setup is as follows.

2 WAN Edge Routers
2 4506 L3 Core switches

The 2 WAN routers connect to the Core via a Cisco 3560 to the Core.  They connect to the 3560 via 100 mbps access links.  Each router has 2 Gig ports.  All 4 of the router interfaces connect to the is 3560.

The 3560 has 2 Trunk uplinks to the Cores; 1 1Gmbps to the primary and the other to the secondary. The 2 Cores have HSRP running between them.  I run EIGRP between the WAN Edge and the Core; between the Core and the user access distribution blocks.  The user access distribution blocks are connected via routed links.

The Server Farm Block consists of 4 Cisco 4948 switches.  2 are connected to the to the 3560 I mentioned above buy 2 1 Gig Trunk Uplinks.  The other 2 are connected via these 2 4948's.  

The reason I have them connected this way is beacuse there is too much distance between the Closet where the 2 cores located and the Closet where the server farm and the WAN gear are located.

I have no problem with the setup between the user access distribution blocks and the Core.  

That said, my goal is to redesign connectivity between the core, the server farm  block and the WAN block.  I understand that the lone Cisco 3560 switch represents a Single Point Of Failure.

Can someone eexpalin to me the design and performance problem casued by having all traffic from the user access blocks; the server farm and the WAN go across this access switch.  I have no means for measuring this so I can explain to stackholders that we need to have all distribution blocks connect to the core directly and not via some access switch?

 How can I justify them to invest in a fiber infrustruture in order to provide direct connectivity since copper distance is way beyond the 300 ft limit?

What theoratical/practical  performance issues are paused by the setup that I have?

Thanks in advance
Who is Participating?

[Product update] Infrastructure Analysis Tool is now available with Business Accounts.Learn More

I wear a lot of hats...

"The solutions and answers provided on Experts Exchange have been extremely helpful to me over the last few years. I wear a lot of hats - Developer, Database Administrator, Help Desk, etc., so I know a lot of things but not a lot about one thing. Experts Exchange gives me answers from people who do know a lot about one thing, in a easy to use platform." -Todd S.

John MeggersNetwork ArchitectCommented:
My opinion is there are no absolutely right or wrong designs.  There are better and worse designs, and what's a better design in one situation isn't necessarily the better design in another situation.

If I'm understanding your description correctly, your server farm and WAN connections are in one building and your core is in another building, and you're questioning why traffic should have to traverse the buildings back to the core to go from the WAN enclave to the server enclave.  While I agree with the benefits of a hierarchical design, in your case it may not be an absolute requirement, especially if there are physical limitations (distance and available media) or financial (the cost of running new connections and upgrading hardware to accommodate the new connection) that make it impractical to take that path.  

That said, it should be obvious that you to make sure your architecture is scaleable and robust.  Maybe you have two switches in the second building with each one having dual connections to the cores in the other building, and you view those switches as a "distribution / secondary core" component of the topology.  Maybe there's another option you can come up with that meets your needs but doesn't break the bank.  There's no absolute requirement that you only have two core switches and all traffic must traverse them, but the challenge is knowing how to work around your limitations and still have a network that isn't out of control because five times someone decided to just directly connect two groups rather than letting traffic traverse a well-designed network, and now you have a network with a variety of paths that complicate the routing table, etc.

I'm sure there are going to be other opinions, but that's my $.02....  HTH

Experts Exchange Solution brought to you by

Your issues matter to us.

Facing a tech roadblock? Get the help and guidance you need from experienced professionals who care. Ask your question anytime, anywhere, with no hassle.

Start your 7-day free trial
It's more than this solution.Get answers and train to solve all your tech problems - anytime, anywhere.Try it for free Edge Out The Competitionfor your dream job with proven skills and certifications.Get started today Stand Outas the employee with proven skills.Start learning today for free Move Your Career Forwardwith certification training in the latest technologies.Start your trial today
Network Architecture

From novice to tech pro — start learning today.