Internet/Telecom/Social Media

What must the world’s data growth curve look like?

Vertical, pretty much?

In 1993, during my first semester of doctoral work at the University of Colorado, we had a guest speaker from one of the federal administrations in a class talking about this newfangled thing called “the Internet.” (There are a number of US agencies in Boulder, and I can’t remember which one he was from. NIST, maybe.) The thing he said that really stuck out was that at that moment in time, the single largest repository of stored electronic data in the world was at the National Center for Atmospheric Research (NCAR), located maybe four miles from campus (and just up the street from where I would later live for a couple of years). How big was this massive bright spot of data? Two terabytes.


So I’m visiting some friends this week. Let’s call them “Bob” and “Jane.” Bob and I got to talking about massive concentrations of stored data because I’m insufferably curious and he works at a very large company that you have heard of (and have probably done business with) but that I can’t name here. I told him my two terabyte story and asked him how much stored data he thought his company was sitting on. Now, Bob is an engineer/developer type who, when posed with a question like this, tends to figure it out and give you an answer. This one, though, was clearly a stumper. He said that by the end of next year, he expects the four-person group that reports to him to own maybe four petabytes. A petabyte is a thousand terabytes.

Then he tried to extrapolate. They have these data centers that are over a million square feet, and he thinks if you figured one blade per square foot you wouldn’t be too far off. And each of those blades is between one and two terabytes.

Wow. How many of these centers are there? He isn’t sure. He knows of one state where there are eight of them.

So, over 8M terabytes of stored data in just these data centers in this one state. And this is an international company. But it’s just one company.

Not sure I have a real point here, other than damn, my head hurts. If you’d like your head to hurt, too, try processing some of these numbers.

Have a nice day.

2 replies »

  1. This doesn’t really detract from your point, but the data collected from an experiment at Fermilab (not a collider experiment everybody hears about, but one on a beamline spur) amounted to about a terabyte around 1990. I imagine the whole lab had some multiple of that.

  2. I was at that lecture, too. I remember that comment about NCAR’s storage (mostly solar research data, weather modeling, etc., as I recall).

    I have to wonder: Given the dramatic — astonishingly dramatic — and still rising increase in digitally stored data, as a species has this been good … or not so good … for us?

Leave us a reply. All replies are moderated according to our Comment Policy (see "About S&R")

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s