Researchers at the University of Cambridge Computer Lab...
Researchers at the University of Cambridge Computer Lab are trying a new Internet architecture they say will replace today’s system with a model similar to peer-to-peer file-sharing “but on a massive scale,” the university said Wednesday. The prototype, developed as…
Sign up for a free preview to unlock the rest of this article
Timely, relevant coverage of court proceedings and agency rulings involving tariffs, classification, valuation, origin and antidumping and countervailing duties. Each day, Trade Law Daily subscribers receive a daily headline email, in-depth PDF edition and access to all relevant documents via our trade law source document library and website.
part of an EU-funded project called “Pursuit,” will overhaul the existing structure of the Net’s Internet Protocol (IP) layer, through which isolated networks are connected, it said. That could enable “a more socially-minded and intelligent system” that lets users obtain information without needing direct access to servers where content is initially stored, it said, and could make the Internet faster, more efficient and more able to withstand rapidly escalating levels of global user demand. The new architecture has implications for net neutrality, said Dirk Trossen, senior researcher at the Computer Lab and Pursuit’s technical manager, in an interview. In addition, because it will likely replace the Internet Protocol, it will do away with the need for IPv4 or IPv6 addresses, he said. On today’s Internet, communications are directed to a particular IP address atop which various activities and applications ride, Trossen said. Pursuit focuses on the “stuff” users want to have or do, which would have labels much like IP addresses, he said. Someone who wants to view a particular video, for example, would ask the network to find the video, not to take her to a specific location, he said. Someone seeking a video on YouTube doesn’t care if it comes from a company server in Oregon or from the laptop of someone in the next room who’s watching it, he said. Individual computers would be able to copy and republish content on receipt, providing other users with the option to access data, or fragments of data, from a range of locations rather than the source itself, the university said. “Essentially, the model would enable all online content to be shared in a manner emulating the ‘peer-to-peer’ approach taken by some file-sharing sites, but on an unprecedented, internet-wide scale,” it said in a news release. This would be similar to BitTorrent in how applications operate, but with some differences, Trossen said. In today’s system, many content providers use BitTorrent to distribute legal content, he said. Similarly, a criminal who gets hold of decrypted content can also republish it there, he said. But with Pursuit, a criminal who republishes legal content won’t see any gain from it because the content will have been protected by its owner, he said. This issue is more about encryption than the Pursuit network, because the architecture doesn’t deal with content protection, he said. Pursuit would make it harder to block traffic, changing the concept of net neutrality, Trossen said. Net neutrality today is a regulatory issue and an “arms race” by ISPs to determine what users are doing with the “shredded” bits and pieces of information crossing their networks, he said. In the Pursuit architecture, because every piece of information can be explicitly labeled and located, ISPs could differentiate prices, service quality and other aspects at a very fine-grained level, he said. They could set prices for certain classes of information, but in a more transparent way, he said. It’s “difficult to say who might oppose Pursuit, Trossen said. That data won’t go back to servers means Tier 1 service providers, who provide connectivity to large networks, and some other economic players might not be happy, he said. Operators understand that the networks need renewal, but to enable Pursuit they will have to give customers new broadband routers and other equipment, he said. That will help device manufacturers sell new products, he said. Pursuit holds advantages for Google and other search engines, Trossen said. It won’t make searches passé because the search function runs at a different level, but search engines often give users stale or broken links because the Internet is based on the idea of having to go to a location, he said. If the server for that location has moved, the search will return an error message, which is bad for search engines, he said. The Internet Engineering Task Force’s research arm has set up a working group to identify elements of Pursuit that can be standardized, Trossen said. He predicted the shift to the new architecture will be driven not by core network providers but by information-rich industries sitting on massive amounts of data they have trouble communicating. For the retail sector, for instance, to have a network that allows it to uniquely identify each information item and send it is “extremely appealing,” he said. Moreover, the industry is powerful, and as customers of network operators, retailers can create requirements for solutions that look like Pursuit, he said. Pursuit is distinct from the Internet of Things in its focus on information rather than objects, he said.