Hacker News Reader

Comment by DanBC | Hacker News Reader

Were they the same as the internal drives? Or just the same capacity?

I get the impression they just kludged as many 3 Tb drives in as they could, and are ignoring any other specs. Maybe I'm missing something.

View on HN

Gleb from Backblaze here. The drives were literally the same inside the external enclosures. We bought a few to start with to check.

We use consumer-grade drives across our entire 40+ petabyte storage farm because we do not need high-performance drives. Instead, we need low-cost, low-power, high-density drives to put into our custom-designed Storage Pods: http://blog.backblaze.com/2011/07/20/petabytes-on-a-budget-v...

We tested "enterprise drives" and found that from a pure reliability perspective, they actually performed worse in our testing than the consumer drives. With about 5 years and 40+ petabytes of data, we're fairly confident this holds true.

Finally, we assume that all drives will fail and replace about 10 drives every week. The software is written such that the system takes drive failures into account and data is stored redundantly and continuously cross-checked.

It's interesting to find that "enterprise drives" performed worse than consumer. Recently, a report came from a Microsoft datacenter, where they found most errors weren't from faulty hardware, but firmware incompatibilities between array controller and drive controller. I wonder if this type of incompatibility occurs more often for enterprise drives than consumer.

The prices on external drives lagged the prices on internal. The externals are generally sold with reduced warranty -- 1 year, or sometimes as low as 90 days -- and my suspicion has always been that many of them failed an internal longevity test and were binned.

So Backblaze is facing a huge increase in drive failure rates in the near future? That's reassuring.

That could be a lot of liability they're holding, since they voided the warranties on the externals by shucking them. So they could wind up paying twice for failed HDDs = $130 + $130 + $25 in gas and wage to pay someone to drive around to pick these up = $285 or so per HDD?

Plus 2 people got banned from Costco.

Failure rates are still likely to be low (<5% per annum)

I've only ever bought five external drives in my lifetime. Incidentally, all of them have been "shucked" over the years, and none of them have failed. They vary from 3-5 years old.

In fact, the only two drives I've had fail on me were OEM packaged, so based on my anecdotal, statistically insignificant personal experience you could even say the opposite, that external drives are more reliable than their OEM counterparts.

You never know. Perhaps the additional packaging prevents damage to the drives in transit, and the process of installing software on the drives (as I believe some manufacturers do with external drives) provides another opportunity to quality check the drives.

*Yev from Backblaze here: I can now purchase from Costco again, so all is well on that front. Even with a price at or around your calculated $285, that is still half as expensive as the $600 per drive we were once quoted during the crisis.

Another way to look at it is they potentially found an inexpensive form of credit. They save money now and might have to repay it back later -- the interesting question will be how much "credit" they'll have to "repay" because of a higher failure rate.

I hope they're keeping some basic metadata about the shucked drives, so they can compile some statistics for us. It would be fascinating to see what the lifespan of these drives will be over the next few years!

(Backblaze engineer here.) I have no magic crystal ball, but we do have more than 10,000 hard drives spinning in our datacenter, some for 5 years now. We have collected some pretty detailed statistics regarding failure rates vs. drive brand (and 20 other metrics). We really don't see the correlation you imply (shucked consumer drives failure rates). I'm hoping to do a blog post about what we have found regarding drive failure rates, but here is a TL;DR summary: drives fail approximately 1 - 5 % each year, your application and RAID architecture and monitoring strategy MUST compensate for that with no data loss.

I'd love to see more information from people with many drives about reliability.

The well known Google study was interesting.

They may have different branding or model numbers, but in terms of technical specifications they are no different to internal drives.

The article implies that all internal drives rocketed in price. The drives used in the external drives will most likely be a subset of the internal drives available.

(Backblaze engineer here.) Exactly, we chose the particular external drives to contain the exact type of drives that we prefer in our datacenter.

MyBook Essentials (pictured) use the same Caviar Green internal hard drive, with the same warranty, which is still priced $10 more on amazon.

Yes, think of the bare drive (if you don't need the enclosure) as "frustration free packaging" - ie, the kind that Amazon charges a good % markup - and people willingly pay for.

Reply on news.ycombinator.com