Google lightning strike

[comrade]

http://www.bbc.co.uk/news/technology-33989384

Google says data has been wiped from discs at one of its data centres in Belgium - after it was struck by lightning four times.

Some people have permanently lost access to their files as a result.

A number of disks damaged following the lightning strikes did, however, later became accessible.

 

 

Unlucky.

Edited August 19, 2015 by elsiegee40

[mthomas08]

Perhaps they can add that in to their TOS, "In the event of 4 lightning strikes you may lose your data".

[Dos_Box]

How tall was that data centre!!

[SteveT]

.... and backups???

[jamesb]

Seems recently written data was affected, so stuff that was waiting to be committed for backup.

 

Honestly looking at the numbers, I don't think they've done badly.

[PotNoodleTech]

Just goes to show - never skimp on UPS's!

[Andrew_C]

never skimp on UPS's!

Show me a UPS that will protect against a (or multiple) direct MVolt strikes?

[Wubbalubbadub]

Show me a UPS that will protect against a (or multiple) direct MVolt strikes?

 

 

Yeah i would be interested in seeing one of these! Even if it existed... pretty sure our education budgets wouldn't cut it!

[Arthur]

never skimp on UPS's!

Google certainly doesn't. Each server has its own dedicated UPS.

 

www.cnet.com/uk/news/google-uncloaks-once-secret-server-10209580/

 

 

Show me a UPS that will protect against a (or multiple) direct MVolt strikes?

If the data is replicated between multiple data centers in different locations surely it doesn't matter if a few UPSs get fried in the process?

 

Honestly looking at the numbers, I don't think they've done badly.

Given that only a few gigabytes were lost out of multiple petabytes, I think they did pretty well considering there were four lightning strikes to the local power grid.

 

http://arstechnica.co.uk/information-technology/2015/08/lightning-strike-permanently-zaps-data-in-european-google-datacenter/

 

The company said that less than 0.000001 percent of storage in its persistent disk system was affected. That doesn't sound like much—it's ten kilobytes for every 1 terabyte—but with the datacenter likely holding many petabytes, it all adds up, and Google could well have lost as much as a few gigabytes of data.

 

Google was already working on an upgrade program to make its caching system more resilient to power issues; the affected storage was still waiting to be upgraded.

 

Also interesting is that the SSDs were unaffected.

 

https://status.cloud.google.com/incident/compute/15056

 

The issue only affected Standard Persistent Disks that existed when the incident began at 09:19 PDT. There was no effect on Standard Persistent Disks created after 09:19. SSD Persistent Disks, disk snapshots, and Local SSDs were not affected by the incident. In particular, it was possible at all times to recreate new Persistent Disks from existing snapshots.

Edited August 23, 2015 by Arthur