In the previous episodes, we mentioned that the internet demand is continuously growing, and the network infrastructure is no longer able to efficiently support the heavy traffic without costly upgrades and extensions. So far, we discussed the tools currently being used to support content delivery, why we think they are no longer efficient, and the solutions we can provide to help the network operate better. However, what we have not yet answered is why. Why has the network evolved this way if it is inefficient? To answer that, let us have look at some of the internet’s history to determine the aspects that made it this way, what challenges it faces from the outcome, and how Benocs fits into this timeline.
Where the internet came from and where it is going.
What started as a military funded project in 1969 that originally linked three institutes in the USA for academic research sharing, the internet has become a necessity in both social and professional lives. The internet as we know it today began its development in the 1970s as the ARPNET, which tested a new technology called “packet switching” via “nodes” to transmit data. From this development, the first email was sent in 1971 and the first international connections in 1973. From there, the internet began connecting more institutions across the USA and Europe, not just academically, but socially as well. In the 1980s, the amount of those connected reached over one hundred, and its increase in popularity led the military to open up the network for public use via TCP/IP addresses. With the amount of users increasing, World Wide Web using HTTP and HTML was developed in 1992, and the internet was commercialized in 1995 with internet service providers (ISP) controlling the backbone. By the year 2000, about half of the United States’ population was online (“40 Maps that Explain the Internet” by Timothy Lee on vox.com).
As a direct result of a growing user population and growing network, the distribution of content became more difficult and less accurate. Content delivery networks (CDN) began to emerge in the late 1990s to resolve these issues using algorithms to accurately route the traffic (“Company History” on Akamai.com). However, the continuous growth of internet users throughout the globe led to more players controlling the network, and, as an outcome, a more complicated network. Given the rise of internet culture, its value proceeded to increase, and complex business deals between ISPs, CDNs, and content producers began to challenge the way content was delivered. The internet backbone today is no longer a simple route from content to back bone to user. Instead, it now consists of multiple players who have created a power struggle in the network backbone with the emergence of business mergers and direct peering. As a result, content delivery has become harder to manage and is traveling blindly across the network, which is creating bottlenecks and the need for expensive network upgrades in order to help evenly distribute traffic.
Benocs fits directly into the future of the network.
One of the largest challenges facing ISPs today – who provide the connection between the backbone and the user – is to keep its infrastructure operating efficiently. This is difficult since the ISP does not control the source, destination, or behavior of content delivery traffic. CDNs on the other hand, are able to control the source in order to deliver content to its destination, the end user. In the first three episodes, we introduced the methods used to deal with the evolution of the network in order to keep it running smoothly. From this, we learned that the products on the market sustain the network, however, based on our research, are outdated or lacking information. As two different entities sharing the same network space, our mission at Benocs is to get these two players playing together to improve traffic for the CDN and reduce costly infrastructure for the ISP. We discovered that by communicating the network topology information from the ISP to the CDN, the CDNs make better delivery decisions, thus reducing traffic and delivery speeds (“Content-aware Traffic Engineering” by Frank, Poese, Smaragdakis, Uhlig, and Feldmann). So instead of trying to cope with the continuously growing network with premature upgrades and costly extensions, it is time to start managing it with information that it already has. It is time to involve Benocs in the evolution of the internet, and start improving the internet’s quality while saving on unnecessary costs!