Increasing Demand Demands Increasing Supply of Bandwidth

Yesterday I wrote about the inevitability of the exaflood, citing recent news that ISPs are starting to invest in their backbone/transport networks.

Well there was another announcement yesterday that plays into this conversation: AT&T; has turned on its high speed "ultra-long haul" network.

This new network boasts a capacity of 40Gbps, as opposed to the 10Gbps commonly found in other networks. And AT&T; has shared their plans for eventually upgrading this further to 100Gbps.

AT&T; claims that this new network will give its customers a better user experience, especially when delivering video applications, but what I'm more interested is what this means on a macro-level.

The truth of the matter is that whether or not you agree or disagree with the decisions of a corporation like AT&T;, there's no denying that their networks carry a huge percentage of Internet traffic.

Additionally, while AT&T; is pushing hard to expand its suite of services, at its core the company is a network operator. It builds networks and makes money by delivering traffic on these networks.

So when they say their backbone's getting filled up and needs more capacity, that potentially says a lot about the larger Internet as they're not likely to invest in and light up gobs of new capacity if the demand isn't there.

Whether you think demand for bandwidth is growing 50% a year, or 100% or 1000%, the simple truth is that whatever the growth rate it's great enough today to force the hand of network operators to continue investing in their networks to keep up.

The primary question in my mind isn't whether or not investment is needed and demand is growing, it's how can we make sure that the necessary investment happens ahead of the cresting wave of demand.

That's my biggest fear: that everyone will know we need the investment, that everyone will have the best of intentions to make that investment, but that because we're not able to get the deployment of broadband firing on all cylinders that the demand for bandwidth will begin to outstrip supply.

The problem is, once we pass that point, we'll be forced to try and play catch up. And I don't know about you, but I wouldn't want to be the one trying to chase demand for bandwidth with a fiber optic cable.

It's not a race we're likely to win without a head start, and we need to be realistic about the fact that while the over-capacity built into networks in the late '90s got us out ahead of the curve, we're now starting to come back to the pack. We're losing that head start.

So if we don't want to lose this race, we need to stop looking backwards--questioning how close we are to the pack, how fast the race is being run, who's to blame for our current situation, and who has the ultimate answer to win the race--and start focusing on getting to the finish line of a country where everyone has affordable, reliable access to broadband as quickly as possible.