SSD Lifetimes

  • Steve Jones - SSC Editor (11/26/2012)


    Jeff Moden (11/26/2012)


    Michael Valentine Jones (11/25/2012)


    I think they quote most mechanical disks with a MTBF of 1 million+ hours. I have to assume that they get these numbers from the reported failure rate of large numbers of disks, but maybe they just make them up.

    Yeah... I don't believe those numbers, either. 😛

    They're statistical extrapolations. There's some science, but "mean" is "mean", not likely or expected. It means that half fail later, which implies that yours might fail tomorrow.

    Precisely. I was in the service so I definitely know what MTBF stands for (Most Troubles Begin way more Frequently :-D)... especially the ones that claim a million + hours. 😛

    --Jeff Moden


    RBAR is pronounced "ree-bar" and is a "Modenism" for Row-By-Agonizing-Row.
    First step towards the paradigm shift of writing Set Based code:
    ________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.

    Change is inevitable... Change for the better is not.


    Helpful Links:
    How to post code problems
    How to Post Performance Problems
    Create a Tally Function (fnTally)

  • Jeff Moden (11/26/2012)


    Steve Jones - SSC Editor (11/26/2012)


    Jeff Moden (11/26/2012)


    Michael Valentine Jones (11/25/2012)


    I think they quote most mechanical disks with a MTBF of 1 million+ hours. I have to assume that they get these numbers from the reported failure rate of large numbers of disks, but maybe they just make them up.

    Yeah... I don't believe those numbers, either. 😛

    They're statistical extrapolations. There's some science, but "mean" is "mean", not likely or expected. It means that half fail later, which implies that yours might fail tomorrow.

    Precisely. I was in the service so I definitely know what MTBF stands for (Most Troubles Begin way more Frequently :-D)... especially the ones that claim a million + hours. 😛

    If they have enough drives in service, MTBF should not be hard to calculate:

    5,000 drives in service * 8,760 hours/year at 100% duty cycle = 43,800,000 total drive operating hours/year, so 43 failures/year would give you 1,000,000 MTBF

    Of couse, there may be a big difference in the failure rate of new drives vs. 6 year old drives, and they probably don't have any history for that.

  • Michael Valentine Jones (11/26/2012)


    Jeff Moden (11/26/2012)


    Steve Jones - SSC Editor (11/26/2012)


    Jeff Moden (11/26/2012)


    Michael Valentine Jones (11/25/2012)


    I think they quote most mechanical disks with a MTBF of 1 million+ hours. I have to assume that they get these numbers from the reported failure rate of large numbers of disks, but maybe they just make them up.

    Yeah... I don't believe those numbers, either. 😛

    They're statistical extrapolations. There's some science, but "mean" is "mean", not likely or expected. It means that half fail later, which implies that yours might fail tomorrow.

    Precisely. I was in the service so I definitely know what MTBF stands for (Most Troubles Begin way more Frequently :-D)... especially the ones that claim a million + hours. 😛

    If they have enough drives in service, MTBF should not be hard to calculate:

    5,000 drives in service * 8,760 hours/year at 100% duty cycle = 43,800,000 total drive operating hours/year, so 43 failures/year would give you 1,000,000 MTBF

    Of couse, there may be a big difference in the failure rate of new drives vs. 6 year old drives, and they probably don't have any history for that.

    Let us (or is that "lettuce") hope they're not tossing the salad with those kinds of calculations. 🙂

    Still, that wouldn't be so bad... it's a 0.86% failure rate.

    --Jeff Moden


    RBAR is pronounced "ree-bar" and is a "Modenism" for Row-By-Agonizing-Row.
    First step towards the paradigm shift of writing Set Based code:
    ________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.

    Change is inevitable... Change for the better is not.


    Helpful Links:
    How to post code problems
    How to Post Performance Problems
    Create a Tally Function (fnTally)

  • It seems those SSDs are performing much better, and I do know some people that have gotten multiple years in SQL Servers with SSDs. Not a representative sample by any means, but they are better than some I heard about 3-4 years ago that measured the tempdb SSD lifetime in months.

Viewing 4 posts - 16 through 18 (of 18 total)

You must be logged in to reply to this topic. Login to reply