Friday, April 16, 2010

User Choice or Imposed Limits: What is Best Way to Manage Bandwidth?

Should the internet treat all data equally, regardless of whether it is part of a multi-gigabyte video file or a short email? Gareth Morgan asks the question in an article at New Scientist.com, after interviewing Johan Pouwelse, a peer-to-peer researcher at the Delft University of Technology in the Netherlands.
Though many will reflexively object, Morgan says one way to deal with demands placed on the network by very-active users is to bill for consumption of bandwidth, not for access at certain speeds. That charging principle would allow people to alter their own behavior, rather than imposing fixed limits to use.
Nearly two years ago, the US Federal Communications Commission (FCC) censured network operator Comcast for trying to impose restrictions on "bandwidth hogs" who use BitTorrent and other file-sharing software. These systems eat up huge amounts of data capacity, and so can degrade the service to other customers, he says.
But the key problem remains unresolved: when large numbers of customers want to access the internet simultaneously, how can traffic be managed in a way that prevents those who are transferring huge multimedia files clogging up the network?
Pouwelse suggests that a different kind of charging tariff could help. Instead of charging customers on the basis of download speeds, network operators should charge users and content providers according to how much data they download or upload. "They could do that without interfering with traffic, in an entirely net neutral way," he says.
This proposal would be opposed by internet giants such as Google and Facebook, who generate large volumes of web traffic and so could face higher charges. But with high-speed broadband stimulating an ever-growing appetite for bandwidth, some way must be found to fairly share out the internet's limited resources.
Bandwidth hogs eat away at principle of net neutrality - tech - 16 April 2010 - New Scientist

No comments:

Generative AI Will NOT have the Impact Many Expect

Generative artificial intelligence, to say nothing of machine learning or neural networks (and eventually general AI), might collectively re...