7 Dumb Cloud Computing Myths
(click image for larger view and for slideshow)
It's that time of year again, when analysts write all those dreadful forecasts-for-the-coming-year columns that are always at least mildly entertaining but, well, often wrong. Such is the nature of predicting the future, but there is one forecast for 2013 that I am absolutely, positively certain of: The wireless LAN (WLAN), which has evolved over the past two decades from curiosity to essential -- and from slow, expensive, unreliable, unmanageable security hole to primary access for just about everyone -- is about to undergo yet another transformation. This time it goes right to the core of enterprise IT strategies: the app. Wireless LANs will not just enable access to but optimize and even provision applications.
Yes, you read that right: Apps will be implemented within the WLAN. I know, I know. This is reminiscent of that "network is the computer" slogan of a decade ago from a certain now-defunct computer company. But such is pretty much what is happening here regardless. We're going from access to the apps themselves. Such a development is actually easy to predict. It's perfectly natural for such functional consolidations to take place.
What, after all, is IT really about? Collecting, managing, transforming, and disseminating information. And how is all this done? Over a network, increasingly wireless both indoors and out, with the back end consisting of routers, switches, servers, storage and assorted appliances. So it makes a lot of sense for the network to have knowledge of apps, and, increasingly, to be the vehicle that not only distributes but also hosts and manages these apps.
[ Read Wireless LAN's New Standard, 802.11ac: Prep Time. ]
When you consider that today's enterprise-class WLANs include all manner of management services across Layers 1-3, extending these facilities all the way to Layer 7 isn't really that big of a deal. We're already seeing advances in identity management and other applications driven by -- and, again resident in – wireless LANs. I've gone so far as to predict that the unified wired/wireless network strategy I've long advocated -- and implemented largely in, yes, applications -- will be extended to the domain of application management directly. The primary driver here? Ease of use. Overburdened and underfunded IT and network operations staffs will look upon this consolidation of function as a godsend.
Still skeptical? Here are a couple of examples. Let's start with Aruba's just-announced 7200 series of controllers, featuring what the company calls AppRF, a facility that enhances application delivery via deep packet inspection, airtime fairness, enhanced QoS, and controller-optimized traffic delivery. Knowledge of applications needs isn't quite the same as provisioning those applications, but it's easy to see where this facility might go over time. Xirrus also just added deep packet inspection to its Wi-Fi arrays, and such might become quite common over the next few years.
Even closer to my thesis is Cisco's Mobility Services Engine (MSE), a wireless/mobility-centric application platform that recently received a slew of enhancements that were mostly lost in the news of the Meraki acquisition. (That itself is an application play as well, by the way; cloud-based network management is in fact an application.) Cisco recently showcased a number of novel location-based apps and enhancements to the MSE's SDK and analytics capabilities. Analytics is still picking up steam but is also a clear trend going forward.
Then, of course, there is the holy grail of LAN traffic: video. It's abundantly clear that caching this bandwidth hog as close as possible to the ultimate consumer is an excellent way to optimize network capacity and throughput. Big disks are cheap, and locating them inside a WLAN controller is also, I believe, going to become common. We're already really close to it, what with the virtualization of WLAN controllers already here.
I find it encouraging that apps and networks are finally coming together. We've reached a certain sufficiency in wireless today, meaning that we can address essentially any WLAN performance requirement. With radio-related innovation consequently slowing a bit -- 2013 won't be the year of 802.11ac, and I'm not expecting any big developments in PHYs beyond this anyway -- the focus shifts quite naturally to applications, again, the raison d'etre for IT in the first place. Even if radio performance were peaking, WLANs remain key to overall IT success. It's a place that, after 20 years of evolution, the WLAN most certainly deserves.
On a different note, this will be my final column for InformationWeek. I've truly enjoyed the privilege of being here for the past four years, but it's time to move on. I wish everyone all the best with their wireless and mobile endeavors, and, of course, a pleasant holiday season and a productive new year.