The explosion in broadband video consumption is creating a significant and growing hairball for broadband Internet Service Providers, content providers, regulators and others. The core problem is that ISPs' networks are getting overwhelmed by the sheer volume of video being consumed each day.

ISPs have several ways to address the situation, but unfortunately none are perfect. For example, Comcast's approach until recently has been to use network management tools to block or slow certain kinds of traffic, such as peer-to-peer. P2P is a particular issue for cable ISPs because it uses scarce "upstream" bandwidth. Network management is highly technical, making it hard for policy-makers to understand it, let alone legislate it. So Comcast is now facing a sanction from the FCC over its network management practices (which it says it's moving away from anyway), because the FCC didn't consider them "reasonable" by its own vague definition.

Time Warner Cable is experimenting with another approach: tiers of service carrying bandwidth caps for users. This is a little bit like today's cell phone model - you buy a package of minutes, and if you go over, you pay extra. Though that may sound reasonable, it invites all kinds of confusion for consumers (e.g. "do I watch that show on CBS.com? Maybe I'd better not, I think my kids have watched a lot of YouTube clips this week and I don't want to go over my cap."). Content providers are justifiably concerned about this potential scenario. Separately, for its part, AT&T recently tried to clarify what its users can and cannot expect from their broadband subscriptions.

Yet another route is for broadband ISPs to adopt a much more expansive technical approach to how content is hosted in their networks and delivered to their users. Equipment vendors like Alcatel-Lucent and Cisco believe that ISPs could convert the current bandwidth problem into a full-fledged business opportunity. This would involve ISPs deploying hardware and software that would enable "managed services," each to be delivered at a specified quality level and for a specified price. So rather than a consumer buying a tier, they would buy a specific service offering (e.g. unlimited Hulu, with HD delivery guaranteed).

This wouldn't be a totally unfamiliar concept. Content providers have been buying managed hosting/delivery services for years from CDNs like Akamai, Limelight, Level 3 and others which guarantee certain delivery metrics. But these CDNs' guarantees can't reach into the "last mile" the ISPs' networks serve. So as ever-more bandwidth intensive content is launched such as HD and long-form, content providers should have an increasing motivation to see last mile ISPs offer comparable managed services offerings from ISPs as well.

However, ISP managed services would require fundamental changes in how these companies currently work together, and also invites concerns from "net neutrality" advocates that ISPs could bias in favor of one content provider or another when making their deals. Though compelling in concept, there are many details to sort out in the managed services approach, making it a longer-term option.

All of this just scratches the surface of the growing bandwidth hairball. Layer on the free-speech advocates like Free Press and Public Knowledge and the politicians looking to make hay with constituents and it's evident that the debate over bandwidth is only going to intensify.

What do you think? Post a comment now.