4

I do not understand when will the core layer is needed. All the materials that I have read in general say that core layer should switch traffic quickly and no filtering. None say when or in what situations the core layer is implemented.

7

For smaller deployments a Collapsed Core is usually fine. Collapsed Core means that some devices perform both Distribution and Core layer duties.

With just a few Access switches there's no need to build a large Core.

If you have a large network then you will have many Access switches. Let's say that you have 20 Distribution switches. Connecting all the Distribution switches would be a huge waste of ports and fiber and you would also need a lot of routing peerings. For everytime you add a new Distribution switch there would be a lot of work connecting it to the others and putting in new routing statements.

For those situations it makes sense to use a Core, you would then only add maybe two connections from each Distribution switch to the Core. The number of peerings on the Distribution layer switches would be a lot less as well.

The main factor is scalability but other things could affect as well like speeds and feeds, the distance your network is covering, external connections and factors like that.

7
  • If there are many access switches, why should I even use distribution switch when a router and VLAN can be used to create broadcast domain?
    – Ron Vince
    Apr 11 '14 at 10:07
  • 1
    @RonVince If you have a lot of access switches, it becomes even more imperative to stick with the access/dist/core model. Distribution switches serve as redundant, high-speed systems that should survive a single node failure. Read the Cisco High Availability Design Guide for in-depth reasoning on what your trying to achieve. The crux of it is to eliminate single points of failure, not solely for breaking up broadcast domains.
    – Ryan Foley
    Apr 11 '14 at 11:15
  • @Fizzle What if I just use access switch at distribution and router at core? Of course 2 access switches and 2 routers for high availability. What would be the disadvantage of that?
    – Ron Vince
    Apr 12 '14 at 2:51
  • @RonVince That's actually backwards, you normally shouldn't route at your core. The disadvantages are all dependent on your requirements, if your organization is comfortable with having a substantial amount of users hanging off a one or two devices, then that's a vulnerability they're willing to accept. When you say "there are many access switches", I assume this is > 200. Some organizations explicitly define access/distribution user requirements, example: more than 128 users per access switch requires redundant uplinks.
    – Ryan Foley
    Apr 12 '14 at 7:59
  • @Fizzle > 200 access switches are overkill for me. I'm not at that level. My idea of using access switch at distribution and router at core is by having an idea that there are 4 access switches, 24 ethernet ports per switch. Somehow, this idea seems like to be collapsed core. Access switch is used to connect between user access switch with router to overcome low port density on router to make mesh topology if connecting the user access switch to router directly. Using another access switch for high availability shows the significant of the problem, 8 ethernet ports now required on the router..
    – Ron Vince
    Apr 12 '14 at 11:35
4

Totally agree with Daniel Dib's response towards scalability and other good points.

One of the main issues we experienced towards implementing a layered access ->distribution/core solution is spanning-tree.

Lets say you implement an access-> distribution/core solution and you have many buildings/floors/departments with many access switches. What we have seen is when someone introduces (accidentally) a spanning-tree loop the cpu on your core goes through the roof (nearly). ALL of the users cannot use their computers, internet, file servers, etc... and the whole infrastructure suffers.

To alleviate this issue, I would recommend to implement an access/distribution -> core solution for your zones/buildings. Keep your vlans and the spanning-tree root roles for these vlans on the distribution switches.

Lets say you have 4 distribution switches connected to your core(s) and a routing protocol taking care of your internal routing between the distribution and the core switches, should one of the distribution switches suffer from a spanning tree loop, only that zone is affected and your other three zones/buildings continue to function unaffected.

That makes troubleshooting somewhat easier :-)

3
  • What about my idea that I explained to Fizzle - using access switch to connect between user access switch to router?
    – Ron Vince
    Apr 13 '14 at 0:49
  • @Ron. That works too, depending on your design. If you look at the "Cisco High Availability Design Guide" document provided by Fizzle. Your router connects to your network at the access layer. Or you could do without.
    – user4565
    Apr 13 '14 at 20:32
  • I also think that the logic reason to implement access>distribution>core is frame loop besides port density on router and port speed on router. Using distribution>core seems to be necessary on very large network that implementing router-on-stick is not good anymore. Assume that the computing resource of router and access switch (that connects user access switches to router) are monstrous, it seems that there is no need for L3 switch. Of course by assuming that the port speed on the router is high too. But again, managing and troubleshooting may be difficult if frame loop happens.
    – Ron Vince
    Apr 14 '14 at 5:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.