High availability is only as good as its weakest link. Small businesses today have access to affordable server and storage systems that provide full hardware redundancy in case of an individual part failure. However, when it comes to the network, some IT Managers rely on the 4-hour part replacement SLAs from their service providers.
Guaranteed part replacement within 4 hours onsite 24-hours a day, 7 days a week, 365 days a year: Ahh, now I can relax, my entire IT infrastructure is safe in case of an outage. Or is it? Time and again I’ve walked into beautifully designed server rooms with redundant power, perfect cable management and cooling. A proud IT manager shows off the high-capacity ultra-redundant storage connected to clustered VMware or Hyper-V hosts with redundant aggregated network ports.
But what about the rest of the LAN?
Too many highly talented IT managers overlook redundancy when it comes to end-user connections. It’s understandable since network path redundancy for 10s or 100s of hosts is very expensive and complex to implement. Fortunately, just about every network switch vendor offers 24x7x4x365 part replacement which is great IF:
- Your end users can afford 4 hours of downtime
- Your hardware vendor actually has the part available
If your vendor doesn’t have the part in stock within a 4-hour delivery radius, then they will source it as quickly as possible. Okay so now your end users are down until…some guy in St. Louis finds your part and puts it on the 2am FedEx flight out of Memphis.
SOLUTION: Keep an updated cold spare racked and ready!
When building out your end-user network add an identical extra switch and keep it updated. Put it in the rack and keep backup switch configs close at hand. That way, if you walk into your server room and see 48 ports solid green when they should be blinking – you’ll be 5 feverish cable-patching minutes away from hero status.