What's the MTBF of the server at 3 years? The vendor should be able to tell you that.
I typically consider the life-time of a server at 3 to 4 years; after that, they tend to start breaking down more often. That doesn't mean you won't find the odd outlier that has been running for 6 years with "nary a problem". Regards, Michael B. Smith Consultant and Exchange MVP http://TheEssentialExchange.com From: Jeremy Anderson [mailto:jer...@mapiadmin.net] Sent: Tuesday, February 09, 2010 12:36 PM To: NT System Admin Issues Subject: OT ? Server ROI - Reuse ? I am not even sure what the subject of this should be. I have a server, it's about 3 years old, the warranty expires in 15 days. It runs a %mission critical App%. This App is going to be replaced with %new mission critical app%. This server meets the hardware requirements for %new app% just fine. (it does require a BIOS update) Its been a stable and reliable server for the last 3 years. I can purchase an extended warranty for around $500, or I can purchase a new server for around $4500.00. The bean counters say, buy the warranty, run %new app% on it, life is good and we save 4 grand. My instinct is that this is a horrible idea, and we should just buy a new server. If we run %new app% on %old server% we will be completely wiping and reloading the OS. My question for everyone here is: How do I convince the bean counters that this is a bad idea. Or, is it not a bad idea, and is a 3 year old server not really that old? How do I justify spending 4k on a server when technically we have a perfectly good server sitting there to be reused? Am I just getting distracted by bright shiny things? %NewApp% is mission critical. If %NewApp% is down, the company is dead in the water. To put this in prospective however, %NewApp% will not be redundant, or even highly available and we are not even considering those options. Think of %newApp% like an Exchange server, for a company that relies on Email for all their communication. And yes, I know %newapp% should be clustered or highly available, but its not going to happen. Does this email make sense? Any help, or insight on the matter would be appreciated. Thanks Jeremy ~ Finally, powerful endpoint security that ISN'T a resource hog! ~ ~ <http://www.sunbeltsoftware.com/Business/VIPRE-Enterprise/> ~