BackBlaze drive failure rate for 2017

[cbn]

https://www.backblaze.com/blog/hard-drive-stats-for-2017/

HGST still doing the best.

Glad to see Seagate ST8000DM02 doing well. (This is just a regular consumer desktop drive)

 

[Elixer]

Helium drives seem to be doing pretty good... and none of those images worked inline.

 

[crashtech]

Strange that I can't see these images in the OP, but when I quote the message, I can see them. Some kind of parsing issue with the editor, maybe...?

Actually, having the article open in a tab is what grants the right to view the hotlinks, I think. I've properly hosted them below.

 

[Elixer]

Could be hot-link protection for the images. *shrug*

 

[crashtech]

You're right about the hotlinking.

 

[dave_the_nerd]

Backblaze only has 93,000 drives? Huh.

 

[cbn]

Backblaze only has 93,000 drives? Huh.

That sounds about right for 500 Petabytes of data:

https://www.backblaze.com/blog/500-petabytes-and-counting/

 

[bononos]

Whats up with the Seagate M001 and M005? Looks like those drives are having problems.

 

[aigomorla]

Whats up with the Seagate M001 and M005? Looks like those drives are having problems.

what else did u expect from seagate?
Seriously im done with them as a company, and i probably will never touch another seagate drive again.
Ive had 6 fail on me in 1 year right after the 3yr mark when the warranty all expired.
They all died one month after each other to make things worse.

But that's besides the point that on 2 of the drives, the warranty actually expired 3 months before the 3yr mark, because seagate has like one of the worst RMA policys on warranty starting from drive manufactor date and not purchase date.

So yeah, i refuse to get seagate, i refuse to recommend seagate unless you bought it as a disposable/temp drive.

That being said HGST for the WIN!
This is why i only buy HGST now....

Could be hot-link protection for the images. *shrug*

i think its something about the xenfono....
You now need to use the forums tools located up top and push that img button .

 

[cbn]

Whats up with the Seagate M001 and M005? Looks like those drives are having problems.

The percentage (29%) is very high for M005 because it only covers 60 drives used for about 20 days each. So therefore having only one drive failure in such a short time can give a very high annualized failure rate.

 

[ElFenix]

Whats up with the Seagate M001 and M005? Looks like those drives are having problems.

i would take any data based on less than 1000 drives with a big box of salt.

 

[bononos]

The percentage (29%) is very high for M005 because it only covers 60 drives used for about 20 days each. So therefore having only one drive failure in such a short time can give a very high annualized failure rate.

i would take any data based on less than 1000 drives with a big box of salt.

True that the stats aren't based on '1000s of drives but WDC with similar drive counts don't have the same problem with high failure rates.

I think how backblaze pods work is that it is only fully on and working hard for a short time until it fills up with data and then it goes into standby mode (or switched off?) so it has a much higher probably of dying in the first few short weeks while it is doing alot of heavy writes.