Perhaps I am being naive, but I thought data centers had back-up power to cover this exact problem? I.e. battery power until the generators kick in... or is this how it used to be done and these days no one bothers and just relies on fail-over to other sites????
Power failures also occur sometimes when the issue is on the wrong side of where power is distributed - we've lost power to our servers when maintenance has gone wrong for example.
I thought major data centres had redundant power from the grid too (not just battery and generator), so the UPS smooths the power transition as the switching gear fails over to the alternate grid connection; then generators come up if that fails too? Is that considered over-engineering?
Components inside the DC should be redundant, but the customer can decide whether to have dual power feeds from redundant areas of the DC provided to their rack/cage.
Even with all this, issues do happen, Telecity had a very high profile and prolonged issue at the end of last year that affected AWS & Azure amongst others: http://www.theregister.co.uk/2015/11/18/telecity_outage_fix_...
08:12
- We are currently experiencing an outage impacting some of our equipment located at Telecity/Equinix Harbour Exchange in London. It is impacting both the Juniper and Extreme LANs.
- We are still investigating to get further information, but we have lost all management access to the devices at the moment.
08:45
- We have had confirmation from Telecity/Equinix that there was indeed a partial loss of power impacting one of the power feeds.
- Power failed at 07:55 BST and was restored at 08:17 BST for the majority of our equipment.
- Unfortunately one of our edge routers (edge4-tch) on the Juniper LAN is still down at the moment.
09:33
- Power to edge4-tch has also been restored now. The router has been back up since 9:15 BST.
- All member services should be restored.
- We have an engineer on site and are working with Telecity/Equinix to understand why we lost all power to that device.
- We are also waiting for a full report from Telecity/Equinix about the cause of the power failure.
---
We don't peer with BT so ou traffic was OK but we did notice loss of service from AWS to BTNet