SearchDataCenter.com has an article about how Stanford University shuts down utilities saving them $250,000, but their IT infrastructure stays up and running.

Over the Christmas holiday vacation, Stanford University shuts down the utilities in more than three-quarters of its 220 buildings on campus but like most organizations, its IT infrastructure stays up and running, said Susan Kulakowski, campus energy manager.
The Stanford, Calif.-based university saves about $250,000 in utility costs during the annual shutdown. It could save more if a portion of its IT infrastructure shut down, but that's not an option, Kulakowski said.
"There would be a big outcry if we tried to shut down our servers. We scale down IT staff and shut off utilities in other areas, but our students and staff still use the system over the break, so we have to have it available," Kulakowski said.
SearchDataCenter.com searched in vain for an enterprise that shuts down servers over the holiday break and came up with this: Even the most idle servers are kept awake at all times because the prospect of shutting down is just plain scary.

I understand SearchDataCenter.com's frustration finding an enterprise that shuts down servers. I got around this problem, by finding someone who was in the right position to implement the idea, and discussed the benefits of demo'ing the concept. Cornell Medical school's BioMedicine has been turning off servers and the users don't even notice.

This facility is one of the only places I know of that turns off servers when they are not needed. For IT Pros they do the equivalent of turning off the lights when they leave the office this holiday weekend (thanksgiving). Think about how many servers are running these next 4 days from Thurs – Sun with no load on them. Would anyone notice if they were turned off?

The amazing thing is the Biomedicine department has been turning off their servers in a high performance compute cluster for the past 6 months and the users don’t notice a change in service, because they turn off and on the compute nodes in response to the job queue. There aren’t going to be that many research scientist submitting jobs on Thanksgiving day. And, as each compute job is completed and sits idle, there is an automated system that turns off the servers. When new compute resources are required as new jobs are submitted on Monday, the machines are turned back on.