News for hackers. Stuff that matters.

A comment on Farming hard drives: how Backblaze weathered the Thailand drive crisis. Parent comment

Re:Farming hard drives: how Backblaze weathered the Thailand... by bruceboughton on Tuesday October 09, 2012 @08:33AM (#4631122)
My guess would be that the 50TB/day figure is before de-duplication and not at all relevant to the actual storage, but even at 10% of that I would expect them to be sourcing them wholesale, not retail.

In fact, this whole article is making me doubt my faith in Backblaze as an off-site backup solution.

[ Parent | on HN ]
Re:Farming hard drives: how Backblaze weathered the Thailand... by recuter on Tuesday October 09, 2012 @08:38AM (#4631152)
I remember a reddit AMA where they mentioned that they don't actually do any de-duplication. This isn't Dropbox and nowhere as reliable as S3 - its literally as if you're renting remote consumer external drives for offsite backup, for that usage pattern Backblaze makes some sense.

And those pods are cool. But I agree, a somewhat flawed approach.

[ Parent | on HN ]
Re:Farming hard drives: how Backblaze weathered the Thailand... by zorlem on Tuesday October 09, 2012 @08:55AM (#4631217)
>This isn't Dropbox and nowhere as reliable as S3 - its literally as if you're renting remote consumer external drives for offsite backup

What are their reliability promises? I think that they're doing some mirroring of the data, but I'm still not sure about the rest of their service details.

I'm considering using them for backups, but have not decided yet.

I came upon this [1] discussion here, from 3 years ago. I'm diving in :)

Yeah, the pods are really sweet and they provide all the design documents [2]!

[1]: http://news.ycombinator.com/item?id=803136

[2]: http://blog.backblaze.com/2011/07/20/petabytes-on-a-budget-v...

edit: added additional details about the design

[ Parent | on HN ]
Re:Farming hard drives: how Backblaze weathered the Thailand... by setec on Tuesday October 09, 2012 @08:52AM (#4631206)
I don't think it should be assumed that there is no data redundancy between pods, afterall from prior blog posts it seems like the pods can be taken offline. I'd also posit that a properly constructed RAID array can mitigate the risks of using consumer parts; afterall that is the "I" in RAID.
[ Parent | on HN ]