On Mon, 24 Jul 2006, Richard A Steenbergen wrote:
I'm surprised nobody said anything about the (apparently regional) utility outage in NoVA on Saturday.
There have been several around the country. Secaucus, NJ, various cities near Los Angeles and the SF Bay Area, not to mention the power quality goes to hell when the grid gets very hot. I've been a bit surprised we haven't heard about more equipment failures.
Equinix Ashburn was running on generator for several hours, but apparently the SAVVIS facility down the road a few miles in Sterling (old Exodus facility) didn't fare so well. The latest story I heard was that they lost 14 out of 16 chillers and customers had to send techs in in shifts because it reached over 130F inside.
I've always been a fan of being able to force 100% economizer and chiller loop bypass emergency operation; it won't keep you "cool" but will help keep your data center from turning into an Easy-Bake Oven(tm). But that failure operating mode is rarely part of the standard HVAC programming. You are probably still going to need to replace some toasted disks and chips afterwards. There is a reason why NEBS equipment costs more.
Come on Sean, this "very few disruptions" stuff is below your usual standards. The least you can do to help us pass the time in this damn heat is to recount a few good stories about routers you could scramble eggs on. :)
I remember back in the old days, when all we had were AGS+ routers in wiring closets without any air conditioning .... We had to carry them uphill, both ways.