Question About Design of this Network

[RFS]

Hello everyone,

I've attached a diagram to this discussion. Would a design like this be used in the real world? R1 and R2 are running HSRP. The issue I'm concerned about with the design is SW2. The diagram shows redundancy with the routers, but then SW2 becomes a single point of failure when there is an issue with the switch. Is there a reason a design like this would be implemented? If there is please enlighten me.

[Georg Pauwen]

Hello,

usually, the only 'reason' a network is built as shown in your topology is: money, or the lack thereof. Ideally, you obviously would want redundancy for ALL devices, including the switches. But that is expensive. It all depends how critical a failure would be to the business. 

So,  a design like the one you are showing is not uncommon.

[georgehewittuk1 · Accepted solution]

For L2/Core L3 Switch resiliency we commonly use switch stacks, VSS/Stackwise virtual and vPCs. This allows us to take advantage of technologies such as MEC (so we can have multiple paths forwarding and not blocking [STP]) for downstream switches/hosts & ensure most importantly for enterprises they have resiliency if a component physically failed. The only reason you would see a single switch on LAN is usually because of budget & customer has accepted risks of that component failure outweighting cost of having resilience.