Hard Drive Endurance

[Lennart Sorensen]

On Thu, Jan 30, 2014 at 05:42:13AM -0500, Giles Orr wrote:
> Can't trust anything these days:

Certainly not this article.

> http://www.enterprisestorageforum.com/storage-hardware/selecting-a-disk-drive-how-not-to-do-research-1.html
> 
> Also from Slashdot, it's a refutal - by someone who claims no association
> with Backblaze, Seagate, or any other HD manufacturer.  I admit to not
> having read it, but Slashdot's summary claims that Backblaze's Seagate
> drives were all significantly older than the other drives and some had
> known problems - so of course they failed first.

Seagate has had a lot of "known problems", many of which Seagate spent
a long time denying at first.  They still failed.  They still put drives
on the market that didn't last.

And that article is amazingly wrong.  It confuses read error rate with
drive failure rate.  And it doesn't understand that 120% failure per
year means the drives on average lasted less than a year (and in fact
is listed as 0.8 years on average).  So no understanding of statistics
or hard drives in general it seems.

-- 
Len Sorensen
--
The Toronto Linux Users Group.      Meetings: http://gtalug.org/
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists