On: Backblaze online backup almost acquired

[barmstrong]

They give some details here on how they got the cost of a Petabyte down:

http://blog.backblaze.com/2009/09/01/petabytes-on-a-budget-h...

[rb2k_ → barmstrong]

Finally somebody using JFS... I could never figure out why you read so little about the filesystem. Seems to be fast, mature and not all that CPU intensive. I also wonder why they're not using something like gluster fs to replicate data over several nodes. data seems to be pretty local at backblaze.