[mythtv-users] BackBlaze hard drive study

[Craig Treleaven]

At 2:23 AM +0000 1/30/14, Gary Buhrmaster wrote:
>On Tue, Jan 21, 2014 at 4:14 PM, Craig Treleaven <ctreleaven at cogeco.ca> wrote:
>>  Very interesting analysis from running 25,000 consumer-class drives over 5
>>  years:
>>
>  > http://blog.backblaze.com/2014/01/21/what-hard-drive-should-i-buy/
>
>A rather scathing response regard the statistical accuracy of this report:
>http://www.enterprisestorageforum.com/storage-hardware/selecting-a-disk-drive-how-not-to-do-research-1.html

Thanks for the link but I think Newman is essentially setting up a 
strawman argument.  His concluding paragraph:

>You must understand the manufacturer's specifications in 
>relationship to the planned usage. If, let's say, you buy a 
>commodity product like a home washing machine, but put it in a 
>laundromat and it breaks after 3 months of usage, should you have 
>expected it to last 3 years?  I think the same question should be 
>asked of anyone using consumer technology and a commercial 
>application.

BackBlaze isn't complaining, per se, about hard drive failures.  They 
are providing their experience in running a lot of drives in a 
demanding environment.  Indeed, the first paragraph of the BackBlaze 
blog is:

>My last two blog posts were about expected drive lifetimes and drive 
>reliability. These posts were an outgrowth of the careful work that 
>we've done at Backblaze to find the most cost-effective disk drives. 
>Running a truly unlimited online backup service for only $5 per 
>month means our cloud storage needs to be very efficient and we need 
>to quickly figure out which drives work.

I take it that they are willing to trade off various factors to 
minimize the overall cost of providing a big pool of storage.  They 
apparently think that the low purchase cost of consumer-grade drives 
offsets the ongoing costs of replacing failed drives.  Reworking 
Newman's analogy, if home washing machines costs $50 and commercial 
units $5,000, hell yes BackBlaze equip the laundromat with them!

Newman's other complaint seems to be the level of detail and 
statistical rigour of the report.  Except it is not a report...it's a 
blog posting.  A pretty detailed blog that follows up on other 
similarly-detailed postings.  Not a consulting report or an academic 
paper; a summary of some internal work that was made available for 
free even though it could be of use to competitors.  Seems 
disingenuous to complain that free information wasn't extensive 
enough!

Craig