• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 593
  • Last Modified:

What is an appropriate Replacement Schedule/Timeline for Cisco Switches and/or Switches in general?

We have about 20 Cisco Catalyst 2950 switches (24 port, for what it is worth). They range in age from 3 to 5 years old. They have been completely trouble free so far.

I'm trying to plan for the next three to five years.  A lot of businesses seem to upgrade pc's every three years and a lot of schools seem to upgrade PC's every five. Is there a rule of thumb for replacing switches, assuming no technical need? Are there technical concerns I should think about in terms of performance?

I hate to replace them just to replace them if they are working fine.

It appears from the web site that Cisco will quit doing firmware supports 12/2007. We don't carry any service contracts on them. Any additional thoughts on the subject would be appreciated.

thanks
Youngslim
0
youngslim
Asked:
youngslim
  • 3
  • 3
  • 2
2 Solutions
 
Don JohnstonInstructorCommented:
There are a couple of approaches:

1) If it ain't broke, don't fix it. Generally, switches and routers are different than PC's in that new applications  aren't coming out every couple of years that require much more processing power. The bottom line with a switch is; a) is it reliable and b) does it move the traffic as quickly as needed.

2) Lots of businesses don't buy their equipment. They lease it. In some cases on a 3 year lease. That means they HAVE to replace it every 3 years. This results in always having he latest equipment.

In summary (from my point of view), if the equipment is not getting overloaded with traffic and you don't need some new feature that's not availble on your current platform and you're happy with it... then keep it.

But that's just my opinion.
0
 
youngslimAuthor Commented:
I agree very much with the if it ain't broke philosophy.

Any experience with the mean time of failure for a switch? Do people even use that term (I've heard it mainly with hard drives).

How would you budget for contigencies? We do have a "hardware" bucket that I use to replace hardware that breaks before a planned replacement, but it would be nice to be a little more scientific about it.

thoughts?
0
 
Don JohnstonInstructorCommented:
Cisco publishes between 159,000 to 480,000 MTBF with 159,000 being about 18 years. If you subscribe to the whole MTBF issue.

http://www.cisco.com/anz/channels/pnr/simplycisco/pdf/c29ie_ds.pdf

That said, they can fail at anytime. So if uptime is what you need, then build in redundancy by adding another switch.
0
Concerto's Cloud Advisory Services

Want to avoid the missteps to gaining all the benefits of the cloud? Learn more about the different assessment options from our Cloud Advisory team.

 
pseudocyberCommented:
Talk to your finance people and figure out what the depreciation is.  I would budget for replacements when they're at $0 and they're supposed to be written off - unless there's a pressing business need they're incapble of meeting before that.

Other than that, keep a level of support you feel comfortable with - either pre configured spare, 9x5 NBD, 4 hour response, etc.
0
 
youngslimAuthor Commented:
Guys - am out of the office today, but sneaking onto the web. Thanks for the response so far. I guess my concern would be to be hit all of the sudden with some pressing need to replace all twenty switches, which, upon reflection would be pretty unlikely (again, as opposed to the real pressing need to replace twenty six or seven year old pcs). Having one or two spare in the racks makes pretty good sense.

Any last thoughts? I'll wrap this one up tomorrow.
-Pat
0
 
Don JohnstonInstructorCommented:
Keep a spare on the shelf.

If you want faster failover, hook up the spare and let spanning-tree do it's thing.

0
 
pseudocyberCommented:
>> I guess my concern would be to be hit all of the sudden with some pressing need to replace all twenty switches,

In my experience, this kind of scenario is usually only driven by an executive being sold on a new technology for which the infrastructure can't support - such as VOIP.  In that case, it's the role of the person in charge of the IT Infrastructure to point out that the current infrastructure can't support something like that and a network refresh would have to be a part of any future VOIP implementation.

If you have reasonable Executives, they'll understand that and either have some kind of ROI in mind and initiate a project to study and/or implement, or they'll just think it's neat and then choke when they see how much it will cost and the project evaporates and the Executive conveniently forgets they ever thought of it.

Another scenario would be very old - as in outdated technology which faced with a new business requirement cannot handle the additional network load - such as a flat segment network with hubs.  However, this is really just another example of changing business needs and your responsibility to point out the corresponding IT needs to support the business.

Hope this helps.
0
 
youngslimAuthor Commented:
Thanks guys. The whole stream was ahelpful. I gave Don the extra point because he was first. Hope that doesn't offend.
Both answers gave me both the assurance that hanging in with the current technology was OK and provided some ready answers if anyone asks why our switches are all 5,7,xx years old.


Thanks again.
0

Featured Post

What does it mean to be "Always On"?

Is your cloud always on? With an Always On cloud you won't have to worry about downtime for maintenance or software application code updates, ensuring that your bottom line isn't affected.

  • 3
  • 3
  • 2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now