Internet Performance Delivered right to your inbox

Plenty of Bandwidth?

A few days ago Om Malik suggested that consumers don’t need any more bandwidth. He made a couple of interesting claims about the new round of speed upgrades being offered by networks in the US (and being mirrored by adsl2+ rollouts in Europe). Does anyone really need 6Mb/s (or 30 Mb/s) at their house, or is it just a big ruse by the communications companies to rip off your money?


Malik discussed a few different aspects of this issue:

  • “It costs the network operators a tiny bit more to offer more bandwidth, but they can sell higher speed connections for a premium price.”
  • “Can your eyes tell the difference between a web-page loading in one second or 0.27 seconds. I guess not. “
  • “The real bandwidth question is when are we going to see an increase in the uplink speeds?”

The last issue is the only real one.

It is patently false that it only costs network operators a tiny bit to offer more bandwidth. Traditional ADSL networks top out at single-digit megabit downloads at reasonable distances. Fiber-coax cable networks with the original design cannot offer speeds more than a low-to-mid-single-digit speeds without upgrades. PON (Passive Optical Network) networks split bandwidth in a finite way based on the amount provisioned at the start and the amount delivered to the end site, and that’s not even counting getting that fiber into the house or business, which is the real problem. The bottom line is this: yes, companies are deploying network that provision much more bandwidth to the end user, but those networks represent significant capital investment. The marginal cost per user for a speed bump might be minimal once the network is deployed, but the CAPEX to get there is a real bitch.

The second point is also blindingly false. Users may not tell the difference between a 1s download of a 1MB webpage and a 0.27s download of the same page. But why continue to think that looking at 1MB web pages is what users will do. The uses of the network expand to fill the capacity of the network.

Recall the plan that Netflix had to deliver movies with little delay to your home across a broadband connection. Instead of using the US Postal Service, they would use your Internet connection to deliver the mail. This plan appears to have been shelved by a combination of rights issues with content providers and the brutal competition with Blockbuster, but something like it (from Netflix or someone else) will come back. When it does, the holy grail is going to be to deliver someone a movie fast enough that they don’t have to wait (long) to watch it. 1 Mb/s won’t be anywhere near enough for that. Plus users want to transfer content to other devices: Sony PSPs, ipods, etc. You can’t stream that from your home connection when you’re driving down the street.

Malik is certainly right about the final point, though: access networks are not provisioning uplink capacity, core network capacity or peering capacity at anything near the rate that they’re provisioning end-user capacity right now. There are some serious technical and cost problems in doing so (10Gb/s worth of access network ports is dramatically less expensive than a 10 Gb/s core network routed port–by about three or four orders of magnitude).

It will be interesting to see how long it takes for consumers who buy a 30Mb/s connection to complain about 1-3 Mb/s throughput to the things they care about when they care about them.


Share Now

Whois: Dyn Guest Blogs

Oracle Dyn is a pioneer in managed DNS and a leader in cloud-based infrastructure that connects users with digital content and experiences across a global internet.