Re: MTTF of a Disk Array

From: Pest (pest_at_nuisancenospam.org)
Date: 02/23/05

  • Next message: examachine_at_gmail.com: "Re: an true information theory" 
    Date: Wed, 23 Feb 2005 06:45:12 +0000 (UTC)

    stephen@nomail.com wrote:
    > berndlosert@netscape.net wrote:
    > :> The formula from "A case for RAID" is based on the assumption that
    > : each
    > :> single disk has a constant probability of failure during each day
    > : that it is
    > :> in use. Estimating MTTF using this formula is conservative, in that
    > : the
    > :> result you calculate will be shorter than the real MTTF, because
    > : disks
    > :> generally fail less often when they are new.

    > : You say the formula is conservative but what is this based on exactly?
    > : Is there any experimental evidence?

    > It is based on probability theory. Consider a coin tossing
    > experiment with a coin that comes up tails with probability p.
    > The expected number of tosses before a tail appears is 1/p.
    > If you think of a tail as a failure, then the MTTF is 1/p.

    Right. And what is the probability _distribution_?

    > If you toss n coins, the probability of seeing a tails is
    > 1-(1-p)^n which for small p is approximately np. The expected number
    > of tosses before a tail appears is now approximately 1/np.

    > Stephen

  • Next message: examachine_at_gmail.com: "Re: an true information theory"

    Relevant Pages

    • Re: A challenge for Pitman-math
      ... So the chance of getting one head in two tosses would be ... the cases have exactly one head so the probability is actually 0.5. ... are 10 cases that have exactly one tail. ... of the ten tosses so the chance of getting one tail and nine heads is ...
      (talk.origins)
    • Re: MTTF of a Disk Array
      ... Consider a coin tossing ... > experiment with a coin that comes up tails with probability p. ... > The expected number of tosses before a tail appears is 1/p. ...
      (comp.theory)
    • Re: Findining a P value from a T distribution
      ... Jerry pointed out the crux of the issue in our discussion. ... probability calculator, that's a separate issue. ... >> The one thing I find lacking is that it can't handle handle tail areas ... But why present the two-tail probability as .000465 ...
      (sci.stat.math)
    • Re: [LogoForum] Re: coin tosses
      ... Probability to have no run longer than r in a seris of n heads and tails ... if I toss 100 fair coins in sequence ...
      (comp.lang.logo)
    • Re: Quantum Gravity-Dark Energy as Zero-Infinity (Coded as 0-1) Duals 2: Quantum/Macrosc
      ... the univariate cumulative distribution functions of X ... when x, y are very small and positive for X, Y nonnegative, we get Rare ... this involves the left tail of the ... In terms of probability, the left tail ...
      (sci.physics)