Should Companies Repurpose Older Servers as Virtualization Hosts?
April 10th, 2009 | 
Author: As IT budgets decline, old servers called to virtualization duty is an article written by Alex Barrett and Colin Steele on SeachVirtualization.com. Other then asking some pretty good virtualization sources their opinions 
, I think the article does a great job explaining why the expense of buying modern hardware still makes sense even in today’s economy. Whether to buy new servers is a common decision being made today as more and more companies are conducting capacity planning studies for consolidation to virtual infrastructure analysis.
Read the entire article for several great points, but here is one quote as an example argument:
“The first thing to determine is whether installing virtualization on older hardware actually saves money. Today’s servers are equipped with quad- and six-core CPUs that can deliver much greater consolidation ratios that single- or dual-core models, he said. “VI3 [VMware Infrastructure 3] is licensed per socket, so you’re paying the same cost for older CPUs but you’re getting less bang for your buck.”But if the budget for new hardware simply isn’t there, consider the age of a server and how much life it has left in it. When identifying a candidate for a virtualization host, identify a box that is three years or younger and has “had a relatively low runtime,” the systems administrator said.”
Posted in 
Tags: 
I've also noticed more repurposing of older equipment at my clients with VMware ELAs (Enterprise Licensing Agreements) in place which allows them to take the licensing costs out of the equation.
CARLO.
Carlo,
No doubt. An Enterprise License agreement tends to encourage a company
to virtualize everything they can! There is not much better than seeing
all those available ESX licenses in VirtualCenter!!
I have never seen servers re-used in a primary datacenter yet. Tons of folks re-using server hardware in their DR sites though.
It all boils down to basic if statements and common sense.
-Find the Machines that are up to the task. Generally if you ask the question “does this machine support VT?” you get the answer which machines are capable.
-Find Underutilized Servers. If you try to get something out of a bloated machine you will most probably get nothing.
-Estimate upgrade costs to bring them up to the task. Yes it is very probable that servers less than 3 years old can be up to the task but upgrades (a bit of memory here and there, storage over here) can not be avoided.
-Calculate the ROI for performing this task including man hours, extra costs, cost savings etc.
There is a purpose to repurpose older servers as virtualization hosts only if they are newish, underutilized and total cost of upgrade does not make sense both now and in the long run.
Sometimes we do get indecisive and do get mixed up in our choices. The choice parameters off the above question are known to us since we started virtualizing stuff. Some of those parameters and choices are the essence of the virtualization argument.
I would be happy to hear arguments or something i missed out of the equation.
nettraptor,
Great criteria breakdown. Can't argue with any of those.
Mark over at Lanamark was inspired to list factors that influence
reusing older hardware as hosts too. Check out
http://blog.lanamark.com/2009/04/how-to-determi...
*Rich Brambley*
rbrambley@gmail.com
VM /ETC <http://feeds2.feedburner.com/%7Er/vmetc/%7E6/2>
? Grab this Headline Animator
<http://feedburner.google.com/fb/a/headlineanima...>
nettraptor,
Great criteria breakdown. Can't argue with any of those.
Mark over at Lanamark was inspired to list factors that influence
reusing older hardware as hosts too. Check out
http://blog.lanamark.com/2009/04/how-to-determi...