One Big Idea to Connect the World

By: Emily Haggstrom Issue: Big Ideas, Smart People Section: Business

As East clashed with West and ideologies collided during the Cold War, the 1950’s set the stage for an idea that would change the world in unprecedented ways. During this volatile time, the Soviet Union launched Sputnik, the first satellite into orbit, prompting President Eisenhower to create the Advanced Research Project Agency (Arpa) to secure the United States as a leader in the technological arms race. In an effort to funnel information worldwide from terminal-to-terminal, an elite team was assembled and the vision of a data network was finally being realized.

Once the tensions internationally subsided, the 1960’s bore the youthful vitality of the baby-boomer generation, who were maturing into college graduates and changing the very fabric of our culture through seemingly superficial trends in music, literature, sports, and fashion. In contrast, human inventions and innovation were revolutionizing the world in the newest wave of electronic technology. Ideas that had previously only been studied in theory were becoming a reality—the human mind was breaking resistance points and uttering the words, "it can’t be done." In universities and research labs across the country, students, professors, and researchers addressed the areas of information theory, data networks, and packet switching.

The foremost thought leaders on these subjects were brought in by Arpa to create an interconnected network where schools and research centers across the country could merge knowledge. The Arpanet project, led by chief scientist Lawrence Roberts, developed the first computer network linking the University of California, Los Angeles with the Stanford Research Institute (SRI), the University of Utah and the University of California, Santa Barbara. This core network eventually grew to include Hawaii and institutions in the United Kingdom and Norway.

"Computers at the time were extremely large and very expensive, but I believed our work would be very economic and effective," said Roberts. By linking computers to a network, the group saved excessive computer costs and a tremendous amount of money by sharing a single connection—it was like realizing the cost of paying for an individual consumer phone line. "The cost savings were 15 times less than using a phone line; it seemed like a no brainer," said Roberts.

By 1969, Arpanet facilitated the first successful communication between nodes, or connection points; for the next two years the network grew to 40 computers as it expanded. Not long after, Internet protocol (TCP/IP) addresses were instituted commercializing what we now know as the Internet.

Knowing it would be a large undertaking to manage, Roberts offered the phone companies an opportunity to take over the network for free. But, he was told that the phone networks were not compatible with what Roberts and his team were working on. So, Roberts and his team continued their work and by 1990, Tim Berners-Lee wrote the initial software for the World Wide Web, which would not have been possible without Roberts and his co-founders work on the Arpanet.

Today, with almost two billion people connecting to the Internet worldwide, the system faces new challenges. Network security has become a serious issue with new viruses, spam, software loopholes and authentication problems affecting users around the globe. What most people fail to realize is that as access to the Internet continues to increase, bandwidth is becoming congested, downloads are sluggish, security is vulnerable, and the quality of service to our global village is slowly diminishing.

"The network was designed for the period in which it was built. It really hasn’t changed its structure, only its speed," said Roberts. As the network experiences spikes in performance, packets drop, causing the network to crash and slowing response times. As software becomes enhanced and new products are instituted, demands on the network from downloading, voice calls, and video streaming, which require more bandwidth, will continue to weaken the network.

Recently Roberts, who was a primary architect of packet switching, has developed an effective counter measure device that directs traffic to dramatically improve responsiveness, throughput, and relieves congestion on the networks. His newest venture, Anagram, routes Internet traffic to control bandwidth so that spikes on the network are reduced to increase speed, reduce delays and eliminate jitter, an end-to-end delay variation between packets. By adjusting individual usage rates, Roberts’ device can equalize capacity between peer-to-peer (P2P) users.

The product has been heavily utilized by Internet service providers (ISP) across Asia who have moved quickly to expand their networks by testing the software, realizing its value, and then implementing the product to enhance their current coverage. The product is also very valuable in higher education where Anagram can evaluate what programs are being used over school networks to categorize and prioritize them based on what packets are slowing the system and then redistributing them based on their priority. By looking at every packet and reviewing the traffic level, it is two to five times cheaper for ISP’s to use Anagram’s device than to re-packet their systems.

"Service should be based on equality, not byte caps. It should be equality by price range, that way there are no caps on wireless users based on what they pay for," said Roberts. "That’s a much better model." "The benefits from this new approach to packet traffic control are numerous, and allow networks to finally support extremely demanding new applications, eliminate the artifact and distortion in voice and video, improve the performance of data applications, guarantee capacity for priority applications, relieve networks from the imbalance caused by multi-flow applications like P2P, and allow each project, department, or application to share a common network without being impacted by the other classes of activities," he said.