What Will Be The Next Data Storage Unit?

The volume of data we're storing is growing exponentially. It's growing so fast that we may not even have the right units of measure defined yet.

Photo: Ethan Miller/Getty Images

Yesterday, I chatted with Robin Gilthorp, the CEO of Terracotta. Terracotta's business is all about using in-memory data storage to accelerate access to critical business data. During his presentation he noted the rapid growth was outpacing the terminology we have:

We're running out of ways to express how much data - terabytes, exabytes, petabytes. I'm proposing hellabytes – a hell of a lot of data.

We noted the existence of the yottabyte yesterday, but even that may not be useful for long. What's your proposal for the next data storage unit of measure?


    Jiggabytes. Similar to Doc Brown's Jiggawatts.

      We are only 2 years away from the future (and hoverboards) after all!

      This all over. Jiggabytes is now a thing. Will top all the currently named storage units.

    Uberbytes? But then that's kind of final-sounding like megabytes, and look how wrong they were!

      "Mega" is simply the standard prefix to denote 1,000,000 of something, e.g. 1 megapixel = 1,000,000 pixels.

    Actually, after petabytes will be zottabytes, then yottabytes. The next one should be lottabytes.



    As in a Tonne of Bytes

    I'm thinking Thetabyte, going with the Ancient Greek theme that has been used for most of the SI prefixes.

    After that would be iota, but I can see them going a different name given the common use of iota to meet small.

      You're on the right track here. However theta is the eighth letter of the Greek alphabet, which would have had it compete with yotta (apparently derived from octo, though I can't see how) for 1000^8. For 1000^9, I imagine the first preference SI would use would be the ninth Greek cardinal prefix ennea-, however the letter E conflicts with exa-, which is already in use. So I think their second preference will be to look at the ninth letter of the Greek alphabet as they've done previously, which is iota as you mentioned. The capital letter i is free, but risks immediate confusion, as an iotabit would be written as Ib, which is far too confusingly similar to the pound abbreviation lb. I doubt they'll switch to the Latin nona-, as that would break consistency. So with that in mind, I think they'll fall back on what they've done in the past, and alter the Greek cardinal prefix instead, as with octo- to yotta.

      Here's my guess: nea-, with the short notation N.

    Maybe algorithms that sniff out crap content and make it disappear. Also known as Bytesthedust.

    At some point, we need a unit officially named "shitload". As in: "I just downloaded ten shitloads of data today."

    Last edited 14/03/13 11:58 am

      I think the logical step after that would be fuckton. "I accidentally a whole fuckton of data at work, now I'm in for it.."

        Ahh, the good old Metric Fucktonne, as opposed to the Imperial Shitload.

          For some reason I just had a vision of an X-Wing pilot: "Watch your back, Red 6! Imperial Shitloads coming in high!"

            "We have discovered a small exhaust port on the fucktonne. With a carefully aimed fuckload, we can cause a chain reaction that should destroy the entire thing"

Join the discussion!

Trending Stories Right Now