The great Big-Data scare of 2012
![]()
It's coming and it's big. I mean really, really big. It's so big you should be afraid, very afraid because your database can't handle it. It's the attack of "Big Data" and it's coming soon to an enterprise near you.
That's the message that was being fed to us at the Gilbane Conference last week, but as far as I could tell, Big Data could be the biggest hype fest we've seen since, well, cloud computing. Now, I know I've written about both these subjects quite often, but I'm growing skeptical over the Big Data idea and here's why.
As I sat and listened to the experts wax away about Big Data in a session last week at Gilbane, I was taken aback. While the speakers--Peter O'Kelly, principal analyst at O'Kelly Associates, and Hadley Reynolds, managing director at Next Era Research--were highly knowledgeable on the ins and outs of Big Data, neither ever actually defined it.
At one point, O'Kelly actually made fun of the idea, saying it was "really, really big." The only hint of what that meant in real terms was a Forrester report which stated anything under 50 terabytes wasn't big--anything over, well, that's big.
But when we got to the Q&A session at the end, moderator Kathleen Reidy from the 451 Group asked the key question. Since every CMS uses an underlying relational database to manage content, and since the speakers had made it clear that relational databases couldn't scale to meet the needs of really Big Data, what did it mean to content management companies?
Well, it probably doesn't mean much because when you think about it, how many companies are dealing with more than 50 terabytes of data. Oh sure, the Googles and the Facebooks surely are, but your average enterprise? Probably not so much. And if most companies aren't dealing with it, surely the content management vendors don't have to be worrying about their databases all that much--at least in the context of expanding to accommodate Big Data requirements.
In fact, Lubor Ptacek, vice president of strategic marketing at OpenText, thinks there are probably maybe 50 to 100 true Big Data customers and companies like IBM are probably targeting them. He says that the big customers will probably be popular online services like Google and Facebook who need to understand this data to drive their business models. Most other companies won't be dealing in numbers like that any time soon.
That's not to say that companies won't want access to Big Data. They probably will, but they won't need to deal with it on that scale in their own data centers. Instead, they will probably use Big Data services like data.gov and data.com. And if Tim Berners Lee has his way, we will all being sharing data in the same fashion we share links on the web.
However it plays out, this much I can predict: Big Data will be the big buzz word of 2012 and pretty soon we're all going to be sick of hearing about it (if we aren't already).
For all the hype, however, much like the cloud, there's buzz around it for a reason. How we end up using that data will probably have less to do with your IT and data center, however, and much more to do with tapping into public and private Big Data stores to combine it with your own data to try and build advantages for your business. - Ron




Comments