With just a handful of different pricing packages, Internet service providers (ISPs) can increase profits and better meet demand, says a new paper by Nick Feamster, an associate professor in Georgia Tech’s college of computing, and colleagues. (The paper, presented at the Special Interest Group on Data Communication conference, aka SIGCOMM, last week, is available here.) Tiered pricing is not new—several ISPs already implement it—but Feamster’s team analyzed the effectiveness of tiers using models built from real-world ISP data.
“The research addresses the fundamental tension between the desire for simple billing models … and the economic efficiency of the resulting flow of traffic,” says Jennifer Rexford, a professor in Princeton’s department of computer science. Simple models, like the blended rates based on megabytes of information per second per month that most ISPs use today, are easy to understand and enforce. However, blended rates disregard other factors, such as the distance the packets of data travel, that can influence the costs of providing service. Ideally, says Feamster, ISPs would offer an “infinite” number of tiers in which the price precisely reflected the costs of the service provided. But how much benefit do tiers offer compared to the bundled pricing used in today’s systems? And how many tiers does it take to approach the optimal results?
To test the effectiveness of tiered pricing, Feamster used data gathered from existing networks to construct a theoretically optimal model with infinite tiers based on traffic and the distance that information had to travel. He also created models of pricing schemes with different numbers of tiers, based on the same criteria, and compared the price effects of these models with the optimal model. He found that the three-to-four-tier model was nearly as efficient (80 percent to 90 percent) as the optimal one.
The handful of easy-to-design tiers that Feamster tested offers “a great ‘sweet spot’ in the trade-off between simplicity and efficiency,” says Rexford. “This work should be very useful to ISPs in fine-tuning their pricing models.”
Andrew Odlyzko, a professor of mathematics at the University of Minnesota who studies Internet traffic, adds that this work provides some science-driven support to the emerging trend toward more complicated pricing models. Several ISPs have already switched to a tiered system, he says, but at this point “there are few guidelines for the carriers to follow.”
Feamster points out that the research is not meant to advocate any particular method of designing tiers. The team organized the tiers based on demand and distance traveled because it was relatively simple, not because they expected it to be the most efficient tier design. More efficient tiers may be possible.
The study focused on Internet transit providers: large ISPs with a national or international reach whose networks act as go-betweens, allowing through-traffic to connect to smaller networks such as enterprises and universities. There are some hints, says Feamster, that similar patterns may emerge in the commercial market, where end users pay for access to the Internet; he and his team hope to investigate this area next.
Sascha Meinrath, director of the New America Foundation’s Open Technology Initiative, sees implications for everyday Internet users. “Many of the core practices used by ISPs … may be remarkably inefficient, a problem that would result in higher prices for end users.”
The hype around DeepMind’s new AI model misses what’s actually cool about it
Some worry that the chatter about these tools is doing the whole field a disservice.
The walls are closing in on Clearview AI
The controversial face recognition company was just fined $10 million for scraping UK faces from the web. That might not be the end of it.
A quick guide to the most important AI law you’ve never heard of
The European Union is planning new legislation aimed at curbing the worst harms associated with artificial intelligence.
These materials were meant to revolutionize the solar industry. Why hasn’t it happened?
Perovskites are promising, but real-world conditions have held them back.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.