topbanner_forum
  *

avatar image

Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
  • Thursday March 28, 2024, 11:33 pm
  • Proudly celebrating 15+ years online.
  • Donate now to become a lifetime supporting member of the site and get a non-expiring license key for all of our programs.
  • donate

Last post Author Topic: How long do hard drives actually live for?  (Read 36752 times)

xtabber

  • Supporting Member
  • Joined in 2007
  • **
  • Posts: 618
    • View Profile
    • Donate to Member
Re: How long do hard drives actually live for?
« Reply #50 on: May 20, 2014, 06:46 AM »
Backblaze has a new report that shows absolutely no correlation between drive temperature and failure, based on  data they collected from 34,000 hard drives.

In other words, keeping drives cool does nothing to keep them running longer.

mouser

  • First Author
  • Administrator
  • Joined in 2005
  • *****
  • Posts: 40,896
    • View Profile
    • Mouser's Software Zone on DonationCoder.com
    • Read more about this member.
    • Donate to Member
Re: How long do hard drives actually live for?
« Reply #51 on: May 20, 2014, 10:26 AM »
that is pretty darn important news if true -- for the last few years i've been operating under the belief that keeping hard drives cool was my #1 priority -- and that this would dictate buying more expensive cases, and running tools to display and alert hard drive temperatures in the system tray where i could keep an eye on them.

IainB

  • Supporting Member
  • Joined in 2008
  • **
  • Posts: 7,540
  • @Slartibartfarst
    • View Profile
    • Read more about this member.
    • Donate to Member
Re: How long do hard drives actually live for?
« Reply #52 on: May 20, 2014, 07:43 PM »
^^ Yes, when I read that news, I immediately wondered what your reaction to the report might be. It's suggesting a kinda non-intuitive or unconventional conclusion in the report (i.e., not in line with conventional wisdom), but it doesn't necessarily flat-out contradict conventional wisdom.
Me, I could only see that it was inconclusive as to what the actual provable causes of hard drive (disk surface) deterioration/failure were - they will probably be found to be a little more complex than just heat. I mean, for example, what about other things, such as (say) the effect of the earth's magnetic field?

You know, where you say that "i've been operating under the belief that...", I really would recommend caution.
From hard-won personal experience, it will be the "belief" thing that gets you every time. As a recent example, for several years I had been operating under/in the belief that I could trust a business partner implicitly and without question, only to recently discover (last month) that she had stolen $19,400 from our partnership in 2007 right at the start of our partnership. She covered it up with an old trick - she put herself in charge of the accounts administration and filed away all the bank statements that would have revealed the theft by showing her syphoning off the funds in increments every month over a 13-month period, so I never saw them.
Talk about gullible. I feel so embarrassed for being trusting dumb.    :-[

It was an incredibly dumb thing for my partner to do though, so I also completely misjudged her intellectual and ethical capacity, let alone her character.
« Last Edit: May 20, 2014, 07:50 PM by IainB »

40hz

  • Supporting Member
  • Joined in 2007
  • **
  • Posts: 11,857
    • View Profile
    • Donate to Member
Re: How long do hard drives actually live for?
« Reply #53 on: May 20, 2014, 08:01 PM »
No correlation with heat? Now that is really interesting...

I wonder, however, if there's a missing factor at work since Backblaze is operating in a climate-controlled data center environment with what I'd suspect is cleaner and better regulated power.

Are the ambient heat levels or fluctuations similar to those experienced in a less controlled environment? I'm wondering if they've engineered an environment such that heat doesn't factor significantly in drive failure - which might suggest it's not just heat, but heat plus an additional factor  - or factors.

Yup. This is getting real interesting.


mouser

  • First Author
  • Administrator
  • Joined in 2005
  • *****
  • Posts: 40,896
    • View Profile
    • Mouser's Software Zone on DonationCoder.com
    • Read more about this member.
    • Donate to Member
Re: How long do hard drives actually live for?
« Reply #54 on: May 20, 2014, 08:08 PM »
I wonder, however, if there's a missing factor at work since Backblaze is operating in a climate-controlled data center environment with what I'd suspect is cleaner and better regulated power.


ding ding ding! i think we have a winner.

look at the chart on that page:
blog-temp-totals.jpg

and we have this comment: " And almost all of the drives are in the nice comfortable range from 15˚ to 30˚."

SO.. it looks like the lesson is, IF YOUR DRIVES ARE ALWAYS RUNNING RELATIVELY COOL, trying to run them cooler won't improve reliability.

But this doesn't at all address the danger of running drives at warm temperature ranges like above 36 degrees or so.

So... Since my drives are running currently at 36,37,32 with no load and in mild weather, and i used to regularly see drive temps of 45 or more, it seems my paranoia of watching my drive temperatures are still warranted.

Target

  • Honorary Member
  • Joined in 2006
  • **
  • Posts: 1,832
    • View Profile
    • Donate to Member
Re: How long do hard drives actually live for?
« Reply #55 on: May 20, 2014, 08:37 PM »
I'm no expert here but unless it's extreme I suspect temperature isn't necessarily a good measure.

obviously it needs to be monitored but I was always of the understanding that it was the hot/cold cycle that does the damage, so using a data centre where everything runs at a pretty steady rate 24/7 isn't going to give you a good comparison (unless perhaps you're looking to build a data centre ;D)

And while there are going to be natural variations between components (see the graph in the previous post) I would assume a higher but steady temp in a stable system was 'acceptable' unless it was extreme.  Sudden spikes or significant increases over time would be cause for concern, otherwise...

superboyac

  • Charter Member
  • Joined in 2005
  • ***
  • Posts: 6,347
    • View Profile
    • Donate to Member
Re: How long do hard drives actually live for?
« Reply #56 on: May 20, 2014, 10:34 PM »
it seems my paranoia of watching my drive temperatures are still warranted
This is a good moment to enjoy.

IainB

  • Supporting Member
  • Joined in 2008
  • **
  • Posts: 7,540
  • @Slartibartfarst
    • View Profile
    • Read more about this member.
    • Donate to Member
Re: How long do hard drives actually live for?
« Reply #57 on: June 06, 2014, 12:47 PM »
Well, from an engineering perspective alone, theory would presumably support the significance/relevance of heat to hard drive life.

ewemoa

  • Honorary Member
  • Joined in 2008
  • **
  • Posts: 2,922
    • View Profile
    • Donate to Member
Re: How long do hard drives actually live for?
« Reply #58 on: June 06, 2014, 08:09 PM »
SO.. it looks like the lesson is, IF YOUR DRIVES ARE ALWAYS RUNNING RELATIVELY COOL, trying to run them cooler won't improve reliability.

Nice obvservation!  So there's a data point that in favor of keeping drive temperatures between 14 and 38 C -- so if it's too hot for us, perhaps it's too hot for them?

f0dder

  • Charter Honorary Member
  • Joined in 2005
  • ***
  • Posts: 9,153
  • [Well, THAT escalated quickly!]
    • View Profile
    • f0dder's place
    • Read more about this member.
    • Donate to Member
Re: How long do hard drives actually live for?
« Reply #59 on: February 05, 2016, 04:17 PM »
Uh-oh...here comes f0dder!
Where did he get off to? I haven't seen him in forever.
Real life - and it looks like I'm only a couple of years late to the party ;)

I got the impression that it could/would be an extremely time-consuming process, since, in bad drive sectors, Spinrite apparently goes down to the level of individual bits of data on the disk, and then resorts to an almost analogue-type approach where the bit has an indistinct magnetic polarity.
This is nonsense, but it's what Gibson wants you to believe, supported by the nonsensical "ooh! magic!" display of the user interface.

Fact is that you could do some interesting tricks back when harddrives where MFM - but that's so long ago that I haven't ever seen one. Claiming you can tweak bit patterns and repair the drive on anything made in the last, oh I dunno, 30 or so years, is snake oil.

What SpinRite can do is repeatedly trying to read a sector, which might eventually succeed. This is one of the most dangerous things you can do to a failing drive, though, since it involves a lot of head movement, and that's not a very nice thing if your drive mechanics are bust - you want to do a quick single-pass ignore-error read of the disk to an image file before doing anything else, if you're interested in saving data.

Another thing SpinRite can do is trigger the drives sector reallocation, but that's not magic either - drives do that automagically when you try to write to a bad sector. This doesn't mean the bad sector is "repaired", simply that this sector on the drive is remapped to a (relatively small) pool of reallocation sectors. And as I mentioned, it's not magic - all it takes is a write to the sector.

Friends don't let friends use SpinRite.
- carpe noctem