• GiuseppeAndTheYeti@midwest.social
    link
    fedilink
    English
    arrow-up
    20
    ·
    edit-2
    2 days ago

    I just purchased a 28TB hard drive for $230 $330. It would have taken 5.6 million of these IBM 350 units to equal that.

    To put it into perspective, that would be more than 2 football fields in height, width, and depth (725ft³). And buying all of those units would have cost $896 billion in 1956. Adjusted for inflation that’s $10.48 trillion.

    Edit: Sorry to get anyone’s hopes up. I mistyped $330 but if you’re wanting to get a mass storage drive at the price I did, I got it from Server Part Deals on eBay. They’re manufacturer recertified so essentially brand new and come with a 2 year warranty. (At least mine did.) My drive had 3 hours of spin time and had been spun up 4 times according to the drive health report. The way they can sell these for so cheap is by buying deprecated spares from massive data centers in bulk and recertifying them to resell.

      • GiuseppeAndTheYeti@midwest.social
        link
        fedilink
        English
        arrow-up
        2
        ·
        2 days ago

        I mistyped. It was $330 and it’s a manufacturer recertified drive with a 2 year warranty and was only spinning for 3 hours and spun up 4 times. So I don’t plan on it failing for awhile. I’ll eventually buy more in the future so they can be configured for RAID.

        • Jerkface (any/all)@lemmy.ca
          link
          fedilink
          English
          arrow-up
          2
          ·
          2 days ago

          I just lost a 12TB Toshiba X300 that was mere months out of its 2 year warranty. Never spin up a single drive! They will always make you wish you mirrored, one day.

          • Toribor@corndog.social
            link
            fedilink
            English
            arrow-up
            4
            ·
            2 days ago

            RAID is still no replacement for a backup. Single drives are fine as long as you have automated backups and can handle the interruption when someone goes wrong.

            • Jerkface (any/all)@lemmy.ca
              link
              fedilink
              English
              arrow-up
              1
              ·
              3 hours ago

              The real world failure rate of single drives is just too high. A second drive makes bad days tolerably rare.

              I’m down a lot longer while I pull everything back down my internet connection from the cloud than I am stuffing a new drive in an enclosure and letting it re-silver in the background.

            • InputZero@lemmy.world
              link
              fedilink
              arrow-up
              1
              ·
              1 day ago

              It depends on what’s on the drive. I have a large library of games stored on striped spinning rust. What it does is let me play very old games without downloading them every time. When one fails which isn’t very often, I just buy a new one, rebuild the array and download again. Usually I’m downloading the library cause I did something stupid and broke it.

              Any data I value at all is at least in a redundant array and anything that I don’t want to ever lose is in a proper 3-2-1 solution. Keeps the costs down, cause I’d be sad if I lost my jellyfin stuff but screwed if I lost my pictures or tax stuff.