Film-Tech Forum ARCHIVE: Hard drive reliability

This is topic Hard drive reliability in forum Digital Cinema Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=16;t=001799

Posted by Carsten Kurz (Member # 5396) on 02-20-2014, 03:16 PM:

This may be interesting to some, as the topic of drive reliability/failures and historic experience ('deathstar') comes up frequently.

Okay, they are not Google or Amazon, but they run a decent number of disc drives in huge RAID configs and naturally DO care about cost/reliability aspects. And they NAME the brands and drives they qualify.

Outcome is - Hitachi and Western Digital in general outperform Seagate drives in reliability, and their history goes back 4 years.

http://blog.backblaze.com/2014/01/21/what-hard-drive-should-i-buy/

Other than the link suggests, they are not selling harddiscs, but operate a huge datacenter for online backup services. They quote to have had 27.000 drives spinning at the end pf 2013.

- Carsten

(Don't know wether I should actually cite the whole page for archiving reasons?)

Posted by Marin Zorica (Member # 1716) on 02-20-2014, 03:49 PM:

Interesting fact to say, and I would agree with them in term of disc realibility. But, one thing is different here, if you lose one hdd you can rebuilt RAID and data loss is not critical as is with web industry since you always have dcp's other servers or you just can call local distributor to get you hdd for ingest movie, but on other hand when some server loses important data, that can huge mess.

Posted by John Thomas (Member # 6509) on 02-27-2014, 02:15 AM:

That's a good read, Carsten. On the same topic of drive reliability, something that doesn't often occur to people is the notion of correlated failures:

Wikipedia

quote:
In practice, the drives are often the same age (with similar wear) and subject to the same environment. Since many drive failures are due to mechanical issues (which are more likely on older drives), this violates the assumptions of independent, identical rate of failure amongst drives; failures are in fact statistically correlated.[4] In practice, the chances of a second failure before the first has been recovered (causing data loss) is not as unlikely as four random failures. In a study of about 100,000 drives, the probability of two drives in the same cluster failing within one hour was four times larger than predicted by the exponential statistical distribution—which characterizes processes in which events occur continuously and independently at a constant average rate. The probability of two failures in the same 10-hour period was twice as large as predicted by an exponential distribution.

That is to say, if all of your drives were manufactured in the same place, around the same time, and are operating in the same environment, it might not be so weird to lose more than one in a short period of time.

Posted by Carsten Kurz (Member # 5396) on 02-27-2014, 04:32 AM:

Right. That goes against the general rule to only operate same brand/type drives within a classic RAID config.

- Carsten