Bookmark and Share
Home » Internet History » Packet Switching

Packet Switching

There has been a great deal of contention over who first came up with this idea, who "invented" the process, and where the credit should be given. No one can argue, however, that the few who had a hand in perfecting this technology truly played a role in the creation of the internet as we know it today.

Packet switching is a network communications process that separates data traffic (digital representations of text, sound, or video data) into clumps, called packets, which are then routed across a shared network. To achieve this, the original message/data is divided into several smaller packets. Each packet is then tagged with its destination or connection ID. In each network node, packets are queued or buffered, resulting in variable delay and throughput, dependent on the traffic burden in the network. This contrasts with the other main paradigm, circuit switching, which arranges a particular circuit with a fixed number of constant bit rate and constant delay connections between nodes for unshared use during the communication session.

Interestingly, the development of packet switching occurred just a few decades after the development of quantum mechanics in physics, which began when Albert Einstein demonstrated that waves of light could also be depicted as streams of individual photons.

Leonard Kleinrock carried on early research in queueing theory which would be significant in packet switching. At MIT, in 1959, he submitted a Ph.D. proposal to study data networks, thus establishing the technology which in time led to the Internet. He finished his work in 1962 which was later published in 1964 by McGraw-Hill as an MIT book entitled "Communication Nets". In this study, he formulated the basic principles of packet switching, thus supplying the foundation for that technology. These principles (along with his later research) continue to provide a groundwork for today's Internet technology and our modern high speed Internet Providers.

Kleinrock is arguably a leading authority and researcher in the field of computer network modeling, analysis and design and a father of the Internet. But the commercial world wasn't ready for data networks and his work was not put to use for most of the 1960's as he continued to publish his results on networking technology while simultaneously climbing quickly through the professorial ranks at UCLA where he had joined the faculty in 1963.

Like the development of hypertext, packet switching seems to have been an idea that craved to be discovered. The packet switching concept was first discovered by Paul Baran in the early 1960's, and then independently a couple of years later by Donald Davies. Leonard Kleinrock's early research in the related field of digital message switching is thought to be a basis for both.

Baran formulated the concept of packet switching as a young electrical engineer at RAND when he was required to perform an investigation into survivable communications networks for the US Air Force, expanding on one of the first wide area computer networks developed for the SAGE radar defense system. His results were first submitted to the Air Force in the summer of 1961 as briefing B-265, and then as paper P-2626, and later in 1964 as a series of eleven astonishingly thorough, comprehensive reports called On Distributed Communications.

Baran's 1964 reports go far beyond documenting the pioneering concept of packet switching and describe an elaborate architecture for a large-scale, spread-out, survivable communications network designed to resist near any level of destruction to individual components without loss of end-to-end communications. Baran also presumed that any connection of the network could break down at any time, and so the network was configured with no central control or governing body.

Baran's groundbreaking work served to convince the U.S. Military that wide area digital computer networks were a promising technology. Baran also spoke with Bob Taylor and J.C.R. Licklider at the IPTO about the idea because they were also working to establish a wide area communications network. Baran's compositions then influenced Roberts and Kleinrock to embrace the technology when they joined the the IPTO for development of the ARPANET, laying the foundation that led to its incorporation into the TCP/IP network communications protocol utilized on the Internet today.

In one of many fascinating such synchronicities in the history of science, Baran's packet switching study was strikingly similar to the work performed independently a couple of years later by Donald Davies at the National Physical Laboratory, including shared details like a packet size of 1024 bits. The term "packet switching" itself was adopted from Davies work, since Baran had named the concept the bit less charming "distributed adaptive message block switching".

At the time in the early 1960's, active communication networks were built from dedicated, analog circuits primarily used for voice telephone connections which were always on once activated. Packet switching completely changed this perspective by viewing networks as discontinuous, digital systems that transmit data in small packets only when required.

At first sight this appears like it introduces two compromises in design: Discontinuity. It gives up the advantage of an always-on, constant connection. Conversions. Analog communications like voice have to go through analog-to-digital encoding to get onto the network and then digital-to-analog decoding at the destination to be read - additional work.

However, as always the details make the difference, and in the end packet switching introduces four concrete advantages that far outweigh any theoretical disadvantages:

The existing communications establishment -- mainly telecommunications companies -- was distrustful about the idea initially, but it was quickly shown that a packet switching network generally worked better, faster, and cheaper than a dedicated circuit network. Since the network apportioned all of the available bandwidth on a packetized basis, many communications could take place simultaneously. This was a major breakthrough, and the key concept that made wide-area communication networks and the Internet itself cost-effective and conceivable.

In the mid-1960's, the Advanced Research Projects Agency (ARPA) - which was created in 1958 as the United States' reaction to Sputnik - became interested in networks. ARPA had been backing numerous computer scientists around the country and, as new researchers were brought in, they of course expected ARPA to supply a computer on which they could do their research; however, ARPA concluded that this community of scientists would be able to share a smaller number of computers if these computers were linked together by means of a data network. The specification for the ARPANET was devised in 1968, and in January 1969, a Cambridge-based computer company, Bolt, Beranek and Newman (BBN) acquired the contract to design, implement and deploy the ARPANET. It was their job to take the specification and produce a computer that could work as the switching node for the packet-switched ARPANET. BBN had chosen a Honeywell minicomputer as the foundation on which they would build the switch.

The packet switching concept was a revolutionary paradigm shift from the predominant model of communications networks using dedicated, analog circuits principally built for audio communications, and instituted a new framework of noncontinuous, digital systems that break messages into individual packets that are sent independently and then assembled back into the original message at the far end. The revolutionary concept formed the foundation for the blueprint of the ARPANET, and then the Internet Protocol, supplying the key enabling technology that has led to the success of the Internet today.