The NPL network, or NPL Data Communications Network, was a local area computer network operated by a team from the National Physical Laboratory (NPL) in London that pioneered the concept of packet switching.
Based on designs first conceived by Donald Davies in 1965, development work began in 1968. Elements of the first version of the network, the Mark I, became operational during 1969 then fully operational in January 1970, and the Mark II version operated from 1973 until 1986. The NPL network followed by the ARPANET in the United States were the first two computer networks that implemented packet switching and the NPL network was the first to use high-speed links. It, along with the ARPANET and the CYCLADES project, laid down the technical foundations of the modern Internet.
During 1965-66, Donald Davies, who was later appointed to head of the NPL Division of Computer Science, proposed a commercial national data network based on packet switching in Proposal for the Development of a National Communications Service for On-line Data Processing. [1] After the proposal was not taken up nationally, he headed a team which produced a design for a local network to serve the needs of NPL and prove the feasibility of packet switching. [2] The design was the first to describe the concept of an "interface computer", today known as a router. [3]
A written version of the proposal entitled A digital communications network for computers giving rapid response at remote terminals was presented by Roger Scantlebury at the Symposium on Operating Systems Principles in 1967. The design involved transmitting signals ( packets) across a network with a hierarchical structure. It was proposed that "local networks" be constructed with interface computers which had responsibility for multiplexing among a number of user systems ( time-sharing computers and other users) and for communicating with "high level network". The latter would be constructed with "switching nodes" connected together with megabit rate circuits ( T1 links, which run with a 1.544 Mbit/s line rate). [4] [5] [6] [7] [8] In Scantlebury's report following the conference, he noted "It would appear that the ideas in the NPL paper at the moment are more advanced than any proposed in the USA". [9] [10] [11]
The first theoretical foundation of packet switching was the work of Paul Baran, in which data was transmitted in small chunks and routed independently by a method similar to store-and-forward techniques between intermediate networking nodes. [12] Davies independently arrived at the same model in 1965 and named it packet switching. [13] He chose the term "packet" after consulting with an NPL linguist because it was capable of being translated into languages other than English without compromise. [14] Davies gave the first public presentation of packet switching on 5 August 1968. [15] His original ideas influenced other research around the world. Larry Roberts incorporated these concepts into the design for the ARPANET. [16] [17] [18] The NPL network initially proposed a line speed of 768 kbit/s. [19] Influenced by this, the planned line speed for ARPANET was upgraded from 2.4 kbit/s to 50 kbit/s and a similar packet format adopted. [20] [21] Louis Pouzin's CYCLADES project in France was also influenced by Davies' work. [22] [6] [23]
The NPL team used their packet switching concept to produce an experimental network using a Honeywell 516 node. Construction began in 1968. Coincidentally, this was the same computer chosen by the ARPANET to serve as Interface Message Processors. [24]
Elements of the first version of the network, Mark I NPL Network, became operational during 1969 (before the ARPANET installed its first node). [25] [26] The network was fully operational in January 1970. [6] The local area NPL network followed by the wide-area ARPANET in the United States were the first two computer networks that implemented packet switching. [27] [28] The network later used high-speed links, the first computer network to do so. [29] [30] The Mark II version operated from 1973. [6] [31]
The NPL team also carried out simulation work on the performance of wide-area packet networks, studying datagrams and network congestion. [6] [32] [33]
The NPL network was later interconnected with other networks, including the Post Office Experimental Packet Switched Service and the European Informatics Network (EIN) in 1976. [6]
In 1976, 12 computers and 75 terminal devices were attached, [34] and more were added. The network remained in operation until 1986. [35]
Alongside Donald Davies, the NPL team included Derek Barber, Roger Scantlebury, Peter Wilkinson, Keith Bartlett, and Brian Aldous. [36] [29] [37]
The first use of the term protocol in a modern data-commutations context occurs in a memorandum entitled A Protocol for Use in the NPL Data Communications Network written by Roger Scantlebury and Keith Bartlett in April 1967. [29] [38] [39] A further publication by Bartlett in 1968 introduced the concept of an alternating bit protocol (later used by the ARPANET and the EIN) [40] [41] and described the need for three levels of data transmission (roughly corresponding to the lower levels of the seven-layer OSI model that emerged a decade later). The Mark II version, which operated from 1973, used such a "layered" protocol architecture. The NPL team also introduced the idea of "protocol verification". [29]
The NPL network was a testbed for internetworking research throughout the 1970s. Davies, Scantlebury and Barber were active members of the International Networking Working Group (INWG) formed in 1972. Vint Cerf and Bob Kahn acknowledged Davies and Scantlebury in their 1974 paper A Protocol for Packet Network Intercommunication, which DARPA developed into the Internet protocol suite used in the modern Internet. [42]
Derek Barber was appointed director of the European COST 11 project played a leading part in proposing the European Informatics Network (EIN) and led the project while Scantlebury led the UK technical contribution. [29] [43] [44] [45] The EIN protocol helped to launch the INWG work, [41] [46] which proposed an international end to end protocol in 1975, although this was not widely adopted. [47] [48] [49] [50]
NPL investigated the "basic dilemma" involved in internetworking; that is, a common host protocol would require restructuring existing networks if they were not designed to use the same protocol. NPL connected with the European Informatics Network by translating between two different host protocols while the NPL connection to the Post Office Experimental Packet Switched Service used a common host protocol in both networks. This work confirmed establishing a common host protocol would be more reliable and efficient. [51]
Davies and Barber published Communication networks for computers in 1973 and Computer networks and their protocols in 1979. [52] [53] [54] They spoke at the Data Communications Symposium in 1975 about the "battle for access standards" between datagrams and virtual circuits, with Barber saying the "lack of standard access interfaces for emerging public packet-switched communication networks is creating 'some kind of monster' for users". [55] For a long period of time, the network engineering community was polarized over the implementation of competing protocol suites, commonly known as the Protocol Wars. It was unclear which type of protocol would result in the best and most robust computer networks. [56]
Davies' later research at NPL focused on data security for computer networks. [57]
The concept of packet switching developed at the NPL became the primary means of data communication in modern computer networks including the Internet. [4] [58] [59] [60]
NPL sponsors a gallery, opened in 2009, about the "Technology of the Internet" at The National Museum of Computing at Bletchley Park. [37]
Then in June 1966, Davies wrote a second internal paper, "Proposal for a Digital Communication Network" In which he coined the word packet,- a small sub part of the message the user wants to send, and also introduced the concept of an "Interface computer" to sit between the user equipment and the packet network.
Both Paul Baran and Donald Davies in their original papers anticipated the use of T1 trunks
the ARPA network is being implemented using existing telegraphic techniques simply because the type of network we describe does not exist. It appears that the ideas in the NPL paper at this moment are more advanced than any proposed in the USA
they lacked one vital ingredient. Since none of them had heard of Paul Baran they had no serious idea of how to make the system work. And it took an English outfit to tell them.
Roger actually convinced Larry that what he was talking about was all wrong and that the way that NPL were proposing to do it was right. I've got some notes that say that first Larry was sceptical but several of the others there sided with Roger and eventually Larry was overwhelmed by the numbers.
In 1965, Davies pioneered new concepts for computer communications in a form to which he gave the name "packet switching." ... The design of the ARPA network (ArpaNet) was entirely changed to adopt this technique.; "A Flaw In The Design". The Washington Post. 30 May 2015.
The Internet was born of a big idea: Messages could be chopped into chunks, sent through a network in a series of transmissions, then reassembled by destination computers quickly and efficiently. Historians credit seminal insights to Welsh scientist Donald W. Davies and American engineer Paul Baran. ... The most important institutional force ... was the Pentagon's Advanced Research Projects Agency (ARPA) ... as ARPA began work on a groundbreaking computer network, the agency recruited scientists affiliated with the nation's top universities.
Although there was considerable technical interchange between the NPL group and those who designed and implemented the ARPANET, the NPL Data Network effort appears to have had little fundamental impact on the design of ARPANET. Such major aspects of the NPL Data Network design as the standard network interface, the routing algorithm, and the software structure of the switching node were largely ignored by the ARPANET designers. There is no doubt, however, that in many less fundamental ways the NPL Data Network had and effect on the design and evolution of the ARPANET.
In nearly all respects, Davies' original proposal, developed in late 1965, was similar to the actual networks being built today.
Leonard Kleinrock: Donald Davies ... did make a single node packet switch before ARPA did
The first packet-switching network was implemented at the National Physical Laboratories in the United Kingdom. It was quickly followed by the ARPANET in 1969.
the first occurrence in print of the term protocol in a data communications context ... the next hardware tasks were the detailed design of the interface between the terminal devices and the switching computer, and the arrangements to secure reliable transmission of packets of data over the high-speed lines
This was the first digital local network in the world to use packet switching and high-speed links.
As Kahn recalls: ... Paul Baran's contributions ... If you look at what he wrote, he was talking about switches that were low-cost electronics. The idea of putting powerful computers in these locations hadn't quite occurred to him as being cost effective. So the idea of computer switches was missing. The whole notion of protocols didn't exist at that time. And the idea of computer-to-computer communications was really a secondary concern.
The authors wish to thank a number of colleagues for helpful comments during early discussions of international network protocols, especially R. Metcalfe, R. Scantlebury, D. Walden, and H. Zimmerman; D. Davies and L. Pouzin who constructively commented on the fragmentation and accounting issues; and S. Crocker who commented on the creation and destruction of associations.