Forget About the Exaflood, The Peta-plosion is Already Here!

I've written about the exaflood more than once in the past, about how we're facing an oncoming explosion of demand for bandwidth to support all sorts of broadband applications.

Most recently I wrote about the need to stop talking about the exaflood as a maybe and start recognizing it as the reality it is.

Well if you're still on the fence about this, consider the findings discussed in this Computerworld article.

While I'm not sure if I can recommend reading this article as it's filled with pretty technical info, the gist of it is that everyone from entertainment companies to universities to hospitals are dealing with data overload as they attempt to establish digital archives.

Here's a startling statistic: "According to Milford, Mass-based analyst firm Enterprise Strategy Group Inc., private sector archive capacity will hit an eye-popping 27,000 petabytes by 2010."

For those keeping track at home 27,000 petabytes equals 27 exabytes, but for now that's beside the primary point that this is a lot of data and that finding ways to efficiently store all this data has already become a significant challenge.

What's driving this growth is things like archiving email, video libraries, medical test results, and beyond. What do all these applications have in common? Their usage is all on the upswing, further driving demand for storage.

Now you may ask what this all has to do with the need for broadband. Well, simply put all this data needs to get into its intended archives in the first place, which means either being carried in on a physical disk or sent over a network. Additionally, the data in these archives is most often meant to be accessible by others, and while they may not all be connected directly to the Internet many of them are and arguably all of them eventually should be.

So here's an example of one broad category of applications that by the end of the decade will potentially be sending exabytes of data over the network. And this application largely has nothing to do with real-time delivery of data, and the projections don't rely on mass adoption by new users. This is simply an extension of what's already happening today.

When I look at the exaflood I do so in terms of the amount of data people want to send over the network, and from that perspective there can be no doubt the exaflood is on its way, assuming it's not already here.

The unresolved question is whether or not the exaflood will overwhelm the networks that have created the opportunities to send data around at high speeds. That's a more contentious issue and one that, unfortunately, we'll either A. wait for but not hear anything from and therefore ignore, or B. have crash over our unsuspecting, unprepared heads.

This reality isn't intended as a threat so much as, in my opinion, an effort to raise the dialogue about how important getting more capacity into the Internet is.

And I'm hopeful that at least on this issue we can all be in agreement: more bandwidth is a good thing and we should support every effort to get it.