MTBF: A bunch of bs

  • Thread starter Thread starter John
  • Start date Start date
Peter said:
Once you have it, it helps you to find out how much it will cost you to do a
maintenance for it.
Click to expand...

How do you calculate the maintenance cost of a single drive? Not the
average maintenance cost of one drive that's part of a set of hundred or
thousand drives; the actual maintenance cost of a single drive. Let's
assume an MTBF of 4M4 h, or a failure probability of 1% in 5 y, and a price
of $200 for the drive. For simplicity let's not consider working time, and
call it replacement cost instead of maintenance cost.

What you will come up with is that you may or may not have to spend another
$200 in those 5 y. The probability for having to spend another $200 is
small, but it's greater than 0. The only thing we know for sure is that the
replacement cost in those 5 y won't be $2 (1% of $200) -- it will be either
$0 (probable), $200 (unlikely), $400 (even more unlikely), or a still
higher multiple of $200 (still more unlikely).

Now apply the same MTBF to an array of 1000 drives. You'll come up with a
high probability that the replacement cost over the course of 5 y will be
around $2000 (1% of $200,000), or very likely in the range of, say, $1000
to $3000. (To give probabilities for any given replacement cost range you
need to assume a failure probability distribution, in addition to the
MTBF.)

That probability of a failure applies the same to a single as to a huge
number of components.
Click to expand...

Of course; I never said anything different. But what you can do with the
probability, its relevancy for a given situation, is quite different for
small and large numbers.

Gerhard
 
Peter said:
In that case DIY.
Click to expand...

Sorry, but you're the one asserting that there is some calculation that one
can perform involving MTBF of a single drive in a single machine, that has
some real world utility, so it's up to you to demonstrate that utility,
which you have failed to do.

In other words, put up or shut up.
 
Of course it may help for the purchase decision, but what does it help
How do you calculate the maintenance cost of a single drive? Not the
average maintenance cost of one drive that's part of a set of hundred or
thousand drives; the actual maintenance cost of a single drive. Let's
assume an MTBF of 4M4 h,
Click to expand...

That is quite high for a 200$ drive.
or a failure probability of 1% in 5 y, and a price
of $200 for the drive. For simplicity let's not consider working time, and
call it replacement cost instead of maintenance cost.
Click to expand...

If time spent on diagnosing a problem, getting a new drive, installing it,
restoring data from a backup; is free, it is hardly a business case. Also
you did not count how much of business you have lost because of a downtime.
What you will come up with is that you may or may not have to spend another
$200 in those 5 y. The probability for having to spend another $200 is
small, but it's greater than 0. The only thing we know for sure is that the
replacement cost in those 5 y won't be $2 (1% of $200) -- it will be either
$0 (probable), $200 (unlikely), $400 (even more unlikely), or a still
higher multiple of $200 (still more unlikely).
Click to expand...

Based on your assumptions, expected maintenance cost is $2.
Now apply the same MTBF to an array of 1000 drives. You'll come up with a
high probability that the replacement cost over the course of 5 y will be
around $2000 (1% of $200,000), or very likely in the range of, say, $1000
to $3000. (To give probabilities for any given replacement cost range you
need to assume a failure probability distribution, in addition to the
MTBF.)



Of course; I never said anything different. But what you can do with the
probability, its relevancy for a given situation, is quite different for
small and large numbers.
Click to expand...

Sure it is. A bad decision for only one drive is not as painfull as a bad
decision with 1000 drives.
 
Which just tells how to calculate the probability, and tells you
Sorry, but you're the one asserting that there is some calculation that one
can perform involving MTBF of a single drive in a single machine, that has
some real world utility, so it's up to you to demonstrate that utility,
which you have failed to do.

In other words, put up or shut up.
Click to expand...

What utility? Do you have a brain?
You may look up this thread for some ideas.

You want me to shut up? You've got it.
 
Maybe it doesn't help YOU.
If I had to chose between otherwise identical drives, one with 300,000 MTBF,
other with 1,500,000 MTBF; I would pick a second one. Even if that would be
for a single hard disk in my PC.
Click to expand...


I'd pick the one with the longer warranty.

It's true that MTBF doesn't say much about a particular drive - for
exactly the same reason that the observed lifetime of a single drive
says little about the MTBF of that model.
 
Peter said:
That is quite high for a 200$ drive.
Click to expand...

Quite irrelevant for the issue at hand. One needs to pick /some/ numbers.
Feel free to pick your own, but try to come up with a real world
calculation.
If time spent on diagnosing a problem, getting a new drive, installing it,
restoring data from a backup; is free, it is hardly a business case. Also
you did not count how much of business you have lost because of a downtime.
Click to expand...

Of course. But if you can't even calculate the simple component replacement
cost, why start with all of that?
Based on your assumptions, expected maintenance cost is $2.
Click to expand...

Now my financial officer is /so/ glad you're not working here :)

You can't get much of a drive for $2. Maybe a coffee to help activating the
brain (some need that :), but not a drive. Either you don't need to replace
it, then it's $0. Or you do need to replace it, then it's $200. But /never/
$2.

You made the same mistake many beginners do when applying statistical
numbers: you apply an average (1% of the purchase price) where there's not
high enough numbers for an average to make sense.

What you calculated is the expected /average/ maintenance cost if you take
/many/ such cases. But we were talking about applying these numbers to
/one/ such case...

But I guess that's enough...

Gerhard
 
Peter said:
What utility?
Click to expand...

YOU are the one claiming that there is utility in having the information.
Now you're asking what that utility might be yourself? Are you changing
your story or is it starting to dawn on you that maybe there isn't any?
Do you have a brain?
Click to expand...

What relevance does that have to the utility of knowing the MTBF of a single
drive?
You may look up this thread for some ideas.
Click to expand...

All I've found is you blustering and everybody else asking you to
demonstrate the utility.
You want me to shut up? You've got it.
Click to expand...

Goody, goody.
 
Peter said:
That is quite high for a 200$ drive.


If time spent on diagnosing a problem, getting a new drive, installing it,
restoring data from a backup; is free, it is hardly a business case. Also
you did not count how much of business you have lost because of a
downtime.


Based on your assumptions, expected maintenance cost is $2.
Click to expand...

So you budget $2 and the drive fails. Now what? What have you gained by
budgeting $2?
Sure it is. A bad decision for only one drive is not as painfull as a bad
decision with 1000 drives.
Click to expand...

Suppose you buy a drive with an MTBF of x instead of one with an MTBF of 2x.
The probability of either failing during its rated service life is very
small, and the one with the longer MTBF can still fail if you are unlucky.
 
If one owns too few HDs to make a statistically significant sample, MTBF can
still mean something if you have to buy warranty or service, though I'm not
sure if any retail warranty providers take it into account.

Think about a car warranty or insurance. The car (or its parts) MTBF is also
thoushands of hours, and probability of you getting into an accident is low
(means MTBA is big), but you still may be paying different premiums
depending on the car MTBF and your (estimated) MTBA.
 
Alexander said:
If one owns too few HDs to make a statistically significant sample, MTBF
can still mean something if you have to buy warranty or service, though
I'm not sure if any retail warranty providers take it into account.
Click to expand...

The price of a service contract is generally a fixed percentage of the
purchase price of the product.
 
Alexander said:
Think about a car warranty or insurance. The car (or its parts) MTBF is also
thoushands of hours, and probability of you getting into an accident is low
(means MTBA is big), but you still may be paying different premiums
depending on the car MTBF and your (estimated) MTBA.
Click to expand...

Yes, of course -- from the insurance provider's point of view, it's not
about individuals (even though their ads may want to make you believe that
:), it's about a /large/ number of individual risks, large enough to make
statistics work.

And you contract insurance so that you get the benefit of the large numbers
of individual risks joined together in your insurance provider. According
to Peter, you could just multiply your home's value (say $400k) with the
risk of it burning down in the next ten years (say 0.1%), set aside the
resulting $400 and not pay fire insurance. Not a procedure I'd recommend :)

Gerhard
 
Back
Top