Sandy takes out NYC colo facility
The diesel's run out by now
Cloud storage: Lower cost and increase uptime
Post-tropical storm Sandy, as the event is now known, has taken out at least one piece of significant information infrastructure.
Internap, which operates a global content delivery network and data centres to make it work, has reportedly emailed customers the following warning earlier today about one of its facilities in New York City:
Please be advised that Internap's LGA11 facility is experiencing significant flooding in the sub-basement of the 75 Broad Street building as a result of Hurricane Sandy. The flooding has submerged and destroyed the site's diesel pumps and is preventing fuel from being pumped to the generators on the mezzanine level.
The available fuel reserves on the mezzanine level are estimated to support customer loads for approximately 5-7 hours. Once this fuel supply has been exhausted the generator will no longer be able to sustain operation and critical customer power loads will be lost.
The Reg understands the email was sent before midnight GMT on the 30th, meaning the diesel has run out. Happily no lives were at risk in the facility.
Internap shut down servers for clients for whom it was contractually obliged to do so. The email suggests others should log on ASAP to do it themselves.
Little else is known about Sandy's effect on the technology world. Twitterer Skeeve Stevens has reported a friend who works at Microsoft reports submarine cables have gone dark.
Amazon Web Services Service Help Dashboard reported all is well at the time of writing, while Rackspace also reports business as usual after earlier detailing how it prepared for the storm.
COMMENTS
Re: wow...
Diesel is very, very unlikely to explode.
Under the right conditions, it might burn fiercely.
Re: wow...
Don't diesel supplies tend to be underground? The whole pump thing from the trucks and all and the added safety of concrete?
The building in not a dedicated colo
The tanks were likely underground as that was where building codes required them to be. Since fire is much more likely than a 100-year flood event, it's pretty sensible. Still, I fail to see why the pumps should fail...
Anyway, I have servers in that colo - they went offline at exactly 8:33 PM PST, according to our monitoring. Here's a link to the location https://maps.google.com/maps?q=75+broad+st,+nyc&hl=en&sll=40.72586,-73.957644&sspn=0.050671,0.129175&gl=us&hnear=75+Broad+St,+New+York,+10004&t=m&z=17
Also, it seems that trans-Atlantic cables are starting to go dark - https://twitter.com/skeevestevens/status/263137865578450944
IT infrastructure monitoring strategies
What you need to know about cloud backup
Enabling efficient data center monitoring
Agentless Backup is Not a Myth
Top 10 SIEM implementer’s checklist
