The internet is very slow.
As a general rule, you need to operate at speeds close to the speed of light, at least 670 million miles per hour. Instead, internet data travels 37 to 100 times slower than that. The terminology for this speed gap is “network delay”. This is a one second delay in the internet connection as the signal travels from the computer to the server and then returns.
Gregory Laughlin, a professor of astronomy at Yale University’s Faculty of Arts and Sciences, says we can do better. According to Laughlin, the Internet can be at least 10 times faster, and perhaps 100 times faster, in the United States.
Laughlin and colleague P. Brighten Godfrey of the University of Illinois at Urbana-Champaign, Bruce Maggs of Duke, and Ankit Singla of ETH Zurich are collaborating on what is slowing down the Internet and what can be done to fix it. I’m a leader. that. Projects funded by the National Science Foundation Speed of light internet..
Researchers say that some important factors are blocking the Internet. For example, the network of underground fiber optic cable routes on which the Internet depends is very chaotic. Zigzags under highways and railroad tracks, bypassing difficult terrain such as mountains, and typically sends hundreds of miles of signals in the wrong direction at some point during transmission.
Second, there is the problem with the fiber optic cable itself, which is essentially glass. Internet data is a pulse of light that passes through a cable. As light passes through the glass, the movement of light is significantly slowed down.
Laughlin and his colleagues say that a network of microwave radio towers across the United States can allow Internet signals to travel linearly through the air, accelerating the Internet.
Moreover, according to Laughlin, the idea has already been successfully tested on a limited scale. For example, equity traders built a microwave network between the Chicago and New Jersey stock exchanges 10 years ago to reduce valuable microseconds from high-frequency trading.
In the final findings they released, 19 19th USENIX Symposium on Network System Design and Implementation In April, Laughlin and his colleagues discovered that microwave networks are definitely faster than fiber networks, even in bad weather, and the economic value of microwave networks is worth the cost of building them. did.
Laughlin recently talked to Yale News about the project.
How did you get to join the internet at the speed of light?
Gregory Laughlin: I was interested in the economic question of where “price formation” occurs in the US financial markets. This required collecting and correlating data from a variety of markets, including the Chicago metropolitan futures market and the New York metropolitan stock market.When I started working on the problem [in 2008] Even with the strong incentive to reduce delays between different locations as much as possible, it was clear that the physical telecommunications infrastructure imposes limits that impede signal transduction at speeds close to the speed of light. ..
Why did this project appeal to you?
Laughlin: I like the problem of intersections of physics, economics and geography. The problem of price formation is the perfect juxtaposition along those lines.
How is this approach different from other surveys of Internet infrastructure?
Laughlin: Bandwidth is often the main concern in the study of the physical structure of the Internet. This concern is the amount of information that can be sent per second on a particular line. Other latency work focuses on the ideas behind the content delivery network, which are related to pre-positioning information. Our work incorporates the perspective of “what would your solution be if you wanted to make small packet traffic as fast as possible across the United States?”
What surprised you most when you saw what was slowing down the internet?
Laughlin: It’s very well known, and it never surprises me, but it’s a huge amount of information that can be carried over fiber optics. By transmitting different color bands of light simultaneously, a single highly specialized multi-core fiberglass can now transmit hundreds of terabits of data per second. Since my formative internet experience occurred in the late 1980s and early 1990s, the current Wifi connection at Yale University offices seems to be very fast. But it’s amazing to find that a single fiber can now send data over a million times faster than my office connection. So it was surprising to find that with the right hybrid infrastructure, the Internet is very fast and can carry an incredible amount of data. However, the internet originated in an organic way rather than a top-down, pre-planned way, so you’ll find all these weird pockets of slow performance.
You and your colleagues have suggested that a national network of microwave wireless transmission towers will make the Internet faster. Why is this?
Laughlin: Microwave wireless transmission tower overlays provide a negligible, seemingly negligible increase in US Internet bandwidth, but overlays are key to the smallest and most delay-sensitive demands. You can process the part. This type of traffic is associated with the procedure for establishing a connection between two sites, each with a large number of round-trip transmissions, each with a small number of bytes. By accelerating these and using the most physically direct routes, you can get a 10 to -100x increase in the most important traffic. On the other hand, applications that can buffer information, such as streaming video, do not need to use the microwave tower. Fiber is best if you have a large block of data that needs to be transferred.
What does it take to build such a network in terms of cost and commitment?
Laughlin: In our paper, we created a detailed model of a national microwave network capable of transmitting 100 gigabits per second between 120 cities in the United States, an average of 5% slower than the speed of light. [which provides the ultimate physical limit]..This network contains approximately 3,000 microwave transmission sites [that use existing towers]And we estimate that it will cost hundreds of millions of dollars to build.
Is the price tag worth it?
Laughlin: After doing a detailed cost analysis, it’s very clear that this type of project will bring economic benefits. The application performs everything from remote surgery to e-commerce and games.
How often do you think about this when you download a document or click on a website?
Laughlin: Only if the site seems to load slowly!
How did you react to the project findings?
Laughlin: The team presented their findings at one of the major conferences in the networking space, and the response was very positive. Of course, it’s a big step from theoretically designing a network and actually implementing it. But we definitely feel it works and is worth building.