Cloud computing: The telcos' game to lose

You're probably sick of hearing about cloud computing. I can't say I blame you. The buzz seems to be nearly deafening these days. And the most annoying part is that for those of us old enough to remember, it's "back to the future" all over again.

True old-timers can recall the days when "cloud computing" meant a mainframe and associated terminals. And younger veterans of the dot-com years still remember application service providers.  I won't say which I am -- but I have to agree that the current incarnation looks eerily familiar.

FAQ: Cloud computing, demystified

To continue reading, register here to become an Insider. You'll get free access to premium content from CIO, Computerworld, CSO, InfoWorld, and Network World. See more Insider content or sign in.

You're probably sick of hearing about cloud computing. I can't say I blame you. The buzz seems to be nearly deafening these days. And the most annoying part is that for those of us old enough to remember, it's "back to the future" all over again.

True old-timers can recall the days when "cloud computing" meant a mainframe and associated terminals. And younger veterans of the dot-com years still remember application service providers.  I won't say which I am -- but I have to agree that the current incarnation looks eerily familiar.

FAQ: Cloud computing, demystified

However, things really are different these days. Here's why: Cloud computing offers an unprecedented ability for users to specify exactly what they want, in two ways.

First, they can define exactly they want to buy. Software-as-a-service offers network-based delivery of a specific application (Salesforce.com being one of the best-known examples) with the underlying software and hardware hidden from the user. Platform-as-a-service delivers a "compute platform," typically an operating system plus a range of software tools. And infrastructure-as-a-service delivers storage and compute cycles on demand.

Second, users can control exactly how much they want to consume, and when. Providers typically offer services (whether software, platform or infrastructure) based on a virtualized architecture, which lets users auto-provision and quickly scale up (or down) to address demand spikes.

For these reasons, cloud computing is more than just hype. In fact, although fewer than 2% of my clients say they're using it this year, nearly 20% say they're looking into cloud-based services.

So where do telcos come in? Basically, they're beautifully positioned to capitalize on the drive towards cloud computing -- particularly infrastructure-as-a-service -- if they don't screw it up.

Why? A couple of reasons. First is that delivering anything across the net consumes bandwidth -- quite a lot in the case of platform- and infrastructure-as-a-service. It makes logical sense for the folks who provide infrastructure and platform services to supply the bandwidth that goes with it -- otherwise, there's the risk of under or over-provisioning.

Second is the fact that telcos are, above all else, service providers. That means their core businesses involve significant investments in customer service -- unlike, say, Google, whose core business is selling eyeballs to advertisers. Although Google also offers a host of innovative applications and services to users, the users aren't customers (in the sense that they pay for, and expect to receive, a certain quality of service).

Finally, there's the fact that effectively delivering infrastructure is a different challenge than, say, delivering applications. Telcos have decades of experience in delivering infrastructure. They're not the only ones -- companies such as IBM, for instance, can say the same. But they've got certainly got the experience.

All that said, the telcos also have a track record of making unbelievably boneheaded moves, so there's no guarantee they'll pull this one off. As I said up front -- it's their game to lose.

Read more about lans & wans in Network World's LANs & WANs section.