Advice about network design

[reindert]

Hi, my client is starting from scratch with his IT in a new building and I'm designing his new network layout. I would like to have some advice and your thoughts about this design.

General info:

• There will be (only) 50 users, so the switches (probably) won't be tested to the limit.
• Switches I want to use:
  • Core switches: Cisco Business 350 Series (CBS350-24XS)
  • Distribution switches: Cisco Business 250 Series
  • Access switches: Cisco Business 220 Series
• RACK01 and RACK08 are about 300 meters apart so there will be fiber connections between the two racks.
• I want to have everything redundant (core switches, access switches, servers, firewalls,...)
• Routing will be done on the firewall (10Gbit uplinks connected to core-switches)

I actualy made 2 designs. One with core and access layer only and one with core, distribution and access layer (but maybe that's overkill for this customer).

1. Design 1: Core and access layer: https://prnt.sc/22gzxh3
   1. I want to stack the coreswitches over the 300m fibers. My only thought with this: What happens if someone accidently cuts the fiber uplinks for the core-stack? Will they still work as two seperate switches so we have the time to fix the stack again?
   2. I have two servers (in redundant 10Gbit synch with StorMagic SVSAN) that will have redundant uplinks (10Gbit) to each of the core-switches (blue dotted lines).
2. Design 2: Core, distribution and access layer: https://prnt.sc/22h02xh
   1. Same as design 1 but with and extra distribution layer (where the servers are connected to). Is this usefull?

I've attached both designs as attachments as well.

Let me know what you guys think :).

Thank you!

[Pedroxh · Accepted solution]

hi reinert

I believe that what will rule the choice are the costs of having the extra layers (Distribution) or not. I don't see problems in a collapsed Core architecture (Distribution-Access) due to the traffic volume and client profile (Enterprise, 50 clients).

At the operation level, the collapsed core design even further helps the operation to resolve troubleshootings faster. As it is a controllable environment, you make an L3 transit between the SW`s Core in stack with the FWs in HA, and all L3 high availability will be covered.

[reindert]

Thank you for your feedback!
So you don't see any problem in stacking the switches (which are 300 meters apart) over fiber? In the past I always stacked the core switches in the same room and connected the access switches to them.

Is there anything I should consider when stacking switches which are not in the same room?  

[reindert]

Thank you for your feedback! I'll go back to the drawing board

[Leo Laohoo]

What is the budget for this exercise?

[reindert]

There's not really a budget. We just need to make a good price-quality offer.