Today’s use of distributed computing is made possible by the vast network of personal computers and workstations that make up the Internet. The technology is relatively new and just beginning to be widely used because the Internet itself developed rapidly only in the last few decades.
During the Cold War of the 1960s, the United States government began to focus on computer science research as a means toward the latest and most secure communications technology which they would need in the event of war. Communications networks were essential for the military and needed to be able to withstand attack so that chains of command would not be broken. One researcher, Paul Baran, developed the idea of a distributed communications network in which messages would be sent through a network of switching nodes until they reached their destination. The nodes would be computers instead of the telephone switches used at the time, so that they would be intelligent enough to decide the best route for sending each message. Many nodes and connections would create redundancy so that messages could arrive through many different paths, and the messages would be broken up into blocks of uniform length to make transmissions simpler and more efficient. This idea of message block switching, or later packet switching, was his most important innovation and was the basis for the design of networks that would become the Internet.
The Advanced Research Projects Agency (ARPA) was created in 1958 in response to the launching by the USSR of the Sputnik satellite. The agency funded research sites at universities and businesses throughout the country, setting up the latest computer facilities at each. To make the most efficient use of these facilities, a project was started to connect them all into a network, the ARPANET. Baran’s work would be the basis for the network’s design, and so he was recruited in 1967 to advise the planning team. The nodes for the network would be interface message processors (IMPs) connected by 56-kilobits-per-second telephone lines. The first IMP was installed in UCLA in 1969 and the first four nodes, at UCLA, the Stanford Research Institute, UC Santa Barbara and Utah, were connected by the end of that year.
As the network grew and other packet switching networks developed in other countries, researchers began to explore concepts for connecting different types of networks, or internetworking. They worked with projects in Europe to create a common host protocol, a structure of gateways linking local area networks, and a system of addressing for the complex layout of hosts that this created. The Internet was born and grew from there, eventually making ARPANET obsolete by the late 1980s so that it was retired in 1990. The National Science Foundation (NSF) took over as provider of the Internet’s backbone. Soon after the transfer, the NSF started taking measures to privatize the Internet. It had grown so that use would clearly go beyond non-profit research into commercial and other private uses, which couldn’t be financed and facilitated by government agencies. Companies that were being contracted by NSF to run its backbone were permitted to build their own backbones for providing commercial internet service, and the NSFNET was officially terminated in 1995.
By the 1980s the Internet was still only a relatively small group of networks for the most part linked to military research and operations. At this time, however, the field of computer science was growing in universities who wanted the network facilities that the ARPA sites enjoyed. One of the first major expansions of the Internet away from military use was the CSNET, a network funded by the NSF to connect computer science departments across the country. The advent of small, affordable personal computers in the late 1970s and especially in 1981 with IBM’s PC created a growing number of home and office users who also wanted access to the internet. In 1975 Robert Metcalfe created the Ethernet system which made connecting computers in a local area simple and inexpensive, and 3Com put commercial Ethernet products on the market soon after. With this privatization and popularization of Internet access, the Internet spread rapidly into all areas of independent life.
As more and more users subscribed to Internet Service Providers, the demand increased for high-speed access in homes and businesses. Most users were connected by narrowband analog telephone lines, requiring a modem to translate digital computer data into analog data and a dial-up process to use the telephone system. In the 1990s, two separate forms of broadband service came from companies who already were wired to most homes and businesses in the country, the telephone and television services. Companies in both areas wanted to expand into other communications markets but weren’t allowed to until the Telecommunications Act of 1996. After this, cable television companies began upgrading fibers for two-way transmission, and two main companies, Excite@Home and RoadRunner made broadband Internet access available starting in 1998. In the telephone sector, digital subscriber lines (DSL) were put on the market to replace analog connections, and Internet Service Providers like AOL entered into contracts with local phone companies to reach this market. Excite@Home has the most broadband subscribers to date, with over 3 million as of June 2001. These always-on connections have made independently-owned computers even more accessible for distributed computing, and make such features as returning results to servers as soon as they are calculated much easier.
The magnitude to which the number of private Internet users has grown in the past decade, and the broadband technology which is currently expanding to many of those users, has made distributed computing practical on a large scale. The world wide web is searched by so many users that research projects have been able to accumulate millions of volunteers to make the use of this technology incredibly beneficial. Because network technology plays such a large role in distributed computing, the evolution of these projects has closely followed the growth of the Internet, so that many groups have begun to take part in just the last few years.
The idea of harnessing the unused CPU cycles of a computer is as old as the first networks that later became the Internet. The actual application of this idea has evolved with these networks, based on what terminals were available for use and what kinds of interactions could be facilitated. The kinds of distributed research projects in operation today were only created in the past few years with the growth of private Internet users and always-on high-speed access.
The first distributed computing programs were a pair of programs called Creeper and Reaper which made their way through the nodes of the ARPANET in the 1970s, the predecessor of the Internet. The Creeper came first and was a worm program, using the idle CPU cycles of processors in the ARPANET to copy itself onto the next system and then delete itself from the previous one. It was modified to remain on all previous computers and the Reaper was created which traveled through the same network and deleted all copies of the Creeper. In this way Creeper and Reaper were the first infectious computer programs and are actually often thought of as the first network viruses. They did no damage, however, to the computers they passed through and were instrumental in exploring the possibility of making use of idle computational power.
Another worm program which expanded on this process was the worm created by John F. Shoch and Jon A. Hupp of Xerox’s Palo Alto Research Center in 1973 to move through the computers in a local Ethernet network. It moved similarly to Creeper and Reaper through about 100 computers and used the idle CPU cycles for rendering computer graphics.
The first Internet-based distributed computing project was started in 1988 by the DEC System Research Center. The project sent tasks to volunteers through email, who would run these programs during idle time and then send the results back to DEC and get a new task. The project worked to factor large numbers and by 1990 had about 100 users.
In the 1990s DEC and other groups began to use an Internet interface to distribute the material to be calculated, driven by challenges sponsored by RSA Security, Inc.’s research center, RSA Laboratories. These were puzzles like factoring and prime number searching, and encryption cracking, and came with cash rewards as incentives since the computation and time effort would presumably be large in order to solve any of the problems. RSA sponsored these challenges as research tools, so that they could solve large computational problems and do research in cryptology by having others groups compete to find the answers. Since computational power was key in these problems, groups pooled their efforts to cover more area.
The most prominent group, considered the first to actually use the internet to distribute data for calculation and collect the results, was a project founded in 1997 called distributed.net. They used independently owned computers as DEC had, but allowed the users to download the program that would utilize their idle CPU time instead of emailing it to them. Distributed.net completed several cryptology challenges by RSA Labs as well as other research facilities with the help of thousands of users.
The project that truly popularized distributed computing and showed that it could work was SETI@Home, an effort by the Search for Extraterrestrial Intelligence (SETI) at the University of California at Berkeley. The project was started in May 1999 to analyze the radio signals that were being collected by the Arecibo Radio Telescope in Puerto Rico. It has gained over three million independent users who volunteer their idle computers to search for signals that may not have originated from Earth. This project has really brought the field to light, so that other groups and companies are quickly following their lead. (See Current Projects)
|[Home] [What Is It?] [History] [Future] [Concerns] [Efficiency] [Curr. Projects] [Resources]|