Packet switching

In telecommunications, packet switching is a method of grouping data into short messages in fixed format, i.e. packets, that are transmitted over a digital network. Packets are made of a header and a payload. Data in the header is used by networking hardware to direct the packet to its destination, where the payload is extracted and used by an operating system, application software, or higher layer protocols. Packet switching is the primary basis for data communications in computer networks worldwide.

During the early 1960s, American engineer Paul Baran developed a concept he called distributed adaptive message block switching, with the goal of providing a fault-tolerant, efficient routing method for telecommunication messages as part of a research program at the RAND Corporation, funded by the United States Department of Defense. His ideas contradicted then-established principles of pre-allocation of network bandwidth, exemplified by the development of telecommunications in the Bell System. The new concept found little resonance among network implementers until the independent work of Welsh computer scientist Donald Davies at the National Physical Laboratory in 1965. Davies coined the modern term packet switching and inspired numerous packet switching networks in the decade following, including the incorporation of the concept into the design of the ARPANET in the United States and the CYCLADES network in France. The ARPANET and CYCLADES were the primary precursor networks of the modern Internet.

A simple definition of packet switching is:

The routing and transferring of data by means of addressed packets so that a channel is occupied during the transmission of the packet only, and upon completion of the transmission the channel is made available for the transfer of other traffic.^[5][6]

Packet switching allows delivery of variable bit rate data streams, realized as sequences of short messages in fixed format, i.e. packets, over a computer network which allocates transmission resources as needed using statistical multiplexing or dynamic bandwidth allocation techniques. As they traverse networking hardware, such as switches and routers, packets are received, buffered, queued, and retransmitted (stored and forwarded), resulting in variable latency and throughput depending on the link capacity and the traffic load on the network. Packets are normally forwarded by intermediate network nodes asynchronously using first-in, first-out buffering, but may be forwarded according to some scheduling discipline for fair queuing, traffic shaping, or for differentiated or guaranteed quality of service, such as weighted fair queuing or leaky bucket. Packet-based communication may be implemented with or without intermediate forwarding nodes (switches and routers). In case of a shared physical medium (such as radio or 10BASE5), the packets may be delivered according to a multiple access scheme.

Packet switching contrasts with another principal networking paradigm, circuit switching, a method which pre-allocates dedicated network bandwidth specifically for each communication session, each having a constant bit rate and latency between nodes. In cases of billable services, such as cellular communication services, circuit switching is characterized by a fee per unit of connection time, even when no data is transferred, while packet switching may be characterized by a fee per unit of information transmitted, such as characters, packets, or messages.

A packet switch has four components: input ports, output ports, routing processor, and switching fabric.^[7]

The concept of switching small blocks of data was first explored independently by Paul Baran at the RAND Corporation during the early 1960s in the US and Donald Davies at the National Physical Laboratory (NPL) in the UK in 1965.^{[1][2][3][11]}

In the late 1950s, the US Air Force established a wide area network for the Semi-Automatic Ground Environment (SAGE) radar defense system. Recognizing vulnerabilities in this network, the Air Force sought a system that might survive a nuclear attack to enable a response, thus diminishing the attractiveness of the first strike advantage by enemies (see Mutual assured destruction).^[12] In the early 1960s, Baran invented the concept of distributed adaptive message block switching in support of the Air Force initiative.^[13][14] The concept was first presented to the Air Force in the summer of 1961 as briefing B-265,^[12] later published as RAND report P-2626 in 1962,^[8] and finally in report RM 3420 in 1964.^[9] The reports describe a general architecture for a large-scale, distributed, survivable communications network. The proposal was composed of three key ideas: use of a decentralized network with multiple paths between any two points; dividing user messages into message blocks; and delivery of these messages by store and forward switching.^[13][15] Baran's network design was focused on digital communication of voice messages using switches that were low-cost electronics.^[16][17][18]

Christopher Strachey, who became Oxford University's first Professor of Computation, filed a patent application in the United Kingdom for time-sharing in February 1959.^[19][20] In June that year, he gave a paper "Time Sharing in Large Fast Computers" at the UNESCO Information Processing Conference in Paris where he passed the concept on to J. C. R. Licklider.^[21][22] Licklider (along with John McCarthy) was instrumental in the development of time-sharing. After conversations with Licklider about time-sharing with remote computers in 1965,^[23][24] Davies independently invented a similar data communication concept, using short messages in fixed format with high data transmission rates to achieve rapid communications.^[25] He went on to develop a more advanced design for a hierarchical, high-speed computer network including interface computers and communication protocols.^[26][27][28] He coined the term packet switching, and proposed building a commercial nationwide data network in the UK.^[29][30] He gave a talk on the proposal in 1966, after which a person from the Ministry of Defence (MoD) told him about Baran's work.^[31]

Roger Scantlebury, a member of Davies' team, presented their work (and referenced that of Baran) at the October 1967 Symposium on Operating Systems Principles (SOSP).^{[28][32][33][34][35]} At the conference, Scantlebury proposed packet switching for use in the ARPANET and persuaded Larry Roberts the economics were favorable to message switching.^{[36][37][38][39][40][41]} Davies had chosen some of the same parameters for his original network design as did Baran, such as a packet size of 1024 bits. To deal with packet permutations (due to dynamically updated route preferences) and datagram losses (unavoidable when fast sources send to a slow destinations), he assumed that "all users of the network will provide themselves with some kind of error control",^[28] thus inventing what came to be known as the end-to-end principle. Davies proposed that a local-area network should be built at the laboratory to serve the needs of NPL and prove the feasibility of packet switching. After a pilot experiment in early 1969,^{[42][43][44][45]} the NPL Data Communications Network began service in 1970.^[46] Davies was invited to Japan to give a series of lectures on packet switching.^[47] The NPL team carried out simulation work on datagrams and congestion in networks on a scale to provide data communication across the United Kingdom.^{[45][48][49][50][51]}

Larry Roberts made the key decisions in the request for proposal to build the ARPANET.^[52] Roberts met Baran in February 1967, but did not discuss networks.^[53][54] He asked Frank Westervelt to explore the questions of message size and contents for the network, and to write a position paper on the intercomputer communication protocol including “conventions for character and block transmission, error checking and re transmission, and computer and user identification."^[55] Roberts revised his initial design, which was to connect the host computers directly, to incorporate Wesley Clark's idea to use Interface Message Processors (IMPs) to create a message switching network, which he presented at SOSP.^{[56][57][58][59]} Roberts was known for making decisions quickly.^[60] Immediately after SOSP, he incorporated Davies' and Baran's concepts and designs for packet switching to enable the data communications on the network.^{[38][61][62][63]}

A contemporary of Roberts' from MIT, Leonard Kleinrock had researched the application of queueing theory in the field of message switching for his doctoral dissertation in 1961–62 and published it as a book in 1964.^[64] Davies, in his 1966 paper on packet switching,^[26] applied Kleinorck's techniques to show that "there is an ample margin between the estimated performance of the [packet-switched] system and the stated requirement" in terms of a satisfactory response time for a human user.^[65] This addressed a key question about the viability of computer networking.^[66] Larry Roberts brought Kleinrock into the ARPANET project informally in early 1967.^[67] Roberts and Taylor recognized the issue of response time was important, but did not apply Kleinrock's methods to assess this and based their design on a store-and-forward system that was not intended for real-time computing.^[68] After SOSP, and after Roberts' direction to use packet switching,^[61] Kleinrock sought input from Baran and proposed to retain Baran and RAND as advisors.^[69][70][71] The ARPANET working group assigned Kleinrock responsibility to prepare a report on software for the IMP.^[72] In 1968, Roberts awarded Kleinrock a contract to establish a Network Measurement Center (NMC) at UCLA to measure and model the performance of packet switching in the ARPANET.^[69]

Bolt Beranek & Newman (BBN) won the contract to build the network. Designed principally by Bob Kahn,^[73][74] it was the first wide-area packet-switched network with distributed control.^[52] The BBN "IMP Guys" independently developed significant aspects of the network's internal operation, including the routing algorithm, flow control, software design, and network control.^[75][76] The UCLA NMC and the BBN team also investigated network congestion.^[73][77] The Network Working Group, led by Steve Crocker, a graduate student of Kleinrock's at UCLA, developed the host-to-host protocol, the Network Control Program, which was approved by Barry Wessler for ARPA,^[78] after he ordered certain more exotic elements to be dropped.^[79] In 1970, Kleinrock extended his earlier analytic work on message switching to packet switching in the ARPANET.^[80] His work influenced the development of the ARPANET and packet-switched networks generally.^[81][82][83]

The ARPANET was demonstrated at the International Conference on Computer Communication (ICCC) in Washington in October 1972.^[84][85] However, fundamental questions about the design of packet-switched networks remained.^[86][87][88]

Roberts presented the idea of packet switching to communication industry professionals in the early 1970s. Before ARPANET was operating, they argued that the router buffers would quickly run out. After the ARPANET was operating, they argued packet switching would never be economic without the government subsidy. Baran had faced the same rejection and thus failed to convince the military into constructing a packet switching network in the 1960s.^[10]

The CYCLADES network was designed by Louis Pouzin in the early 1970s to study internetworking.^[89][90] It was the first to implement the end-to-end principle of Davies, and make the host computers responsible for the reliable delivery of data on a packet-switched network, rather than this being a service of the network itself.^[91] His team was thus first to tackle the highly-complex problem of providing user applications with a reliable virtual circuit service while using a best-effort service, an early contribution to what will be the Transmission Control Protocol (TCP).^[92]

Bob Metcalfe and others at Xerox PARC outlined the idea of Ethernet and the PARC Universal Packet (PUP) for internetworking.^[93]

In May 1974, Vint Cerf and Bob Kahn described the Transmission Control Program, an internetworking protocol for sharing resources using packet-switching among the nodes.^[94] The specifications of the TCP were then published in RFC 675 (Specification of Internet Transmission Control Program), written by Vint Cerf, Yogen Dalal and Carl Sunshine in December 1974.^[95]

The X.25 protocol, developed by Rémi Després and others, was built on the concept of virtual circuits. In the mid-late 1970s and early 1980s, national and international public data networks emerged using X.25 which was developed with participation from France, the UK, Japan, USA and Canada. It was complemented with X.75 to enable internetworking.^[96]

Packet switching was shown to be optimal in the Huffman coding sense in 1978.^[97][98]

In the late 1970s, the monolithic Transmission Control Program was layered as the Transmission Control Protocol (TCP), atop the Internet Protocol (IP). Many Internet pioneers developed this into the Internet protocol suite and the associated Internet architecture and governance that emerged in the 1980s.^{[99][100][101][102][103][104]}

For a period in the 1980s and early 1990s, the network engineering community was polarized over the implementation of competing protocol suites, commonly known as the Protocol Wars. It was unclear which of the Internet protocol suite and the OSI model would result in the best and most robust computer networks.^{[105][106][107]}

Leonard Kleinrock's research work during the 1970s addressed packet switching networks, packet radio networks, local area networks, broadband networks, nomadic computing, peer-to-peer networks, and intelligent software agents.^[108][109] His theoretical work on hierarchical routing with student Farouk Kamoun became critical to the operation of the Internet.^[110][111] Kleinrock published hundreds of research papers,^[112][113] which ultimately launched a new field of research on the theory and application of queuing theory to computer networks.^[80][114]

Complementary metal–oxide–semiconductor (CMOS) VLSI (very-large-scale integration) technology led to the development of high-speed broadband packet switching during the 1980s–1990s.^{[115][116][117]}

Roberts claimed in later years that, by the time of the October 1967 SOSP, he already had the concept of packet switching in mind (although not yet named and not written down in his paper published at the conference, which a number of sources describe as "vague"), and that this originated with his old colleague, Kleinrock, who had written about such concepts in his Ph.D. research in 1961-2.^{[58][36][59][118][119]} In 1997, along with seven other Internet pioneers, Roberts and Kleinrock co-wrote "Brief History of the Internet" published by the Internet Society.^[120] In it, Kleinrock is described as having "published the first paper on packet switching theory in July 1961 and the first book on the subject in 1964". Many sources about the history of the Internet began to reflect these claims as uncontroversial facts. This became the subject of what Katie Hafner called a "paternity dispute" in The New York Times in 2001.^[121]

The disagreement about Kleinrock's contribution to packet switching dates back to a version of the above claim made on Kleinrock's profile on the UCLA Computer Science department website sometime in the 1990s. Here, he was referred to as the "Inventor of the Internet Technology".^[122] The webpage's depictions of Kleinrock's achievements provoked anger among some early Internet pioneers.^[123] The dispute over priority became a public issue after Donald Davies posthumously published a paper in 2001 in which he denied that Kleinrock's work was related to packet switching. Davies also described ARPANET project manager Larry Roberts as supporting Kleinrock, referring to Roberts' writings online and Kleinrock's UCLA webpage profile as "very misleading".^[124][125] Walter Isaacson wrote that Kleinrock's claims "led to an outcry among many of the other Internet pioneers, who publicly attacked Kleinrock and said that his brief mention of breaking messages into smaller pieces did not come close to being a proposal for packet switching".^[123]

Davies' paper reignited a previous dispute over who deserves credit for getting the ARPANET online between engineers at Bolt, Beranek, and Newman (BBN) who had been involved in building and designing the ARPANET IMP on the one side, and ARPA-related researchers on the other.^[75][76] This earlier dispute is exemplified by BBN's Will Crowther, who in a 1990 oral history described Paul Baran's packet switching design (which he called hot-potato routing), as "crazy" and non-sensical, despite the ARPA team having advocated for it.^[126] The reignited debate caused other former BBN employees to make their concerns known, including Alex McKenzie, who followed Davies in disputing that Kleinrock's work was related to packet switching, stating "... there is nothing in the entire 1964 book that suggests, analyzes, or alludes to the idea of packetization".^[127]

Former IPTO director Bob Taylor also joined the debate, stating that "authors who have interviewed dozens of Arpanet pioneers know very well that the Kleinrock-Roberts claims are not believed".^[128] Walter Isaacson notes that "until the mid-1990s Kleinrock had credited [Baran and Davies] with coming up with the idea of packet switching".^[123]

A subsequent version of Kleinrock's biography webpage was copyrighted in 2009 by Kleinrock.^[129] He was called on to defend his position over subsequent decades.^[130] In 2023, he acknowledged that his published work in the early 1960s was about message switching and claimed he was thinking about packet switching.^[131] Primary sources and historians recognize Baran and Davies for independently inventing the concept of digital packet switching used in modern computer networking including the ARPANET and the Internet.^{[1][2][38][132][133]}

Kleinrock has received many awards for his ground-breaking applied mathematical research on packet switching, carried out in the 1970s, which was an extension of his pioneering work in the early 1960s on the optimization of message delays in communication networks.^[80][134] However, Kleinrock's claims that his work in the early 1960s originated the concept of packet switching and that his work was a source of the packet switching concepts used in the ARPANET have affected sources on the topic, which has created methodological challenges in the historiography of the Internet.^{[121][123][125][130]} Historian Andrew L. Russell said "'Internet history' also suffers from a third, methodological, problem: it tends to be too close to its sources. Many Internet pioneers are alive, active, and eager to shape the histories that describe their accomplishments. Many museums and historians are equally eager to interview the pioneers and to publicize their stories".^[135]

Packet switching may be classified into connectionless packet switching, also known as datagram switching, and connection-oriented packet switching, also known as virtual circuit switching. Examples of connectionless systems are Ethernet, IP, and the User Datagram Protocol (UDP). Connection-oriented systems include X.25, Frame Relay, Multiprotocol Label Switching (MPLS), and TCP.

In connectionless mode each packet is labeled with a destination address, source address, and port numbers. It may also be labeled with the sequence number of the packet. This information eliminates the need for a pre-established path to help the packet find its way to its destination, but means that more information is needed in the packet header, which is therefore larger. The packets are routed individually, sometimes taking different paths resulting in out-of-order delivery. At the destination, the original message may be reassembled in the correct order, based on the packet sequence numbers. Thus a virtual circuit carrying a byte stream is provided to the application by a transport layer protocol, although the network only provides a connectionless network layer service.

Connection-oriented transmission requires a setup phase to establish the parameters of communication before any packet is transferred. The signaling protocols used for setup allow the application to specify its requirements and discover link parameters. Acceptable values for service parameters may be negotiated. The packets transferred may include a connection identifier rather than address information and the packet header can be smaller, as it only needs to contain this code and any information, such as length, timestamp, or sequence number, which is different for different packets. In this case, address information is only transferred to each node during the connection setup phase, when the route to the destination is discovered and an entry is added to the switching table in each network node through which the connection passes. When a connection identifier is used, routing a packet requires the node to look up the connection identifier in a table.^{[citation needed]}

Connection-oriented transport layer protocols such as TCP provide a connection-oriented service by using an underlying connectionless network. In this case, the end-to-end principle dictates that the end nodes, not the network itself, are responsible for the connection-oriented behavior.

In telecommunication networks, packet switching is used to optimize the usage of channel capacity and increase robustness.^[59] Compared to circuit switching, packet switching is highly dynamic, allocating channel capacity based on usage instead of explicit reservations. This can reduce wasted capacity caused by underutilized reservations at the cost of removing bandwidth guarantees. In practice, congestion control is generally used in IP networks to dynamically negotiate capacity between connections. Packet switching may also increase the robustness of networks in the face of failures. If a node fails, connections do not need to be interrupted, as packets may be routed around the failure.

Packet switching is used in the Internet and most local area networks. The Internet is implemented by the Internet Protocol Suite using a variety of link layer technologies. For example, Ethernet and Frame Relay are common. Newer mobile phone technologies (e.g., GSM, LTE) also use packet switching. Packet switching is associated with connectionless networking because, in these systems, no connection agreement needs to be established between communicating parties prior to exchanging data.

X.25, the international CCITT standard of 1976, is a notable use of packet switching in that it provides to users a service of flow-controlled virtual circuits. These virtual circuits reliably carry variable-length packets with data order preservation. DATAPAC in Canada was the first public network to support X.25, followed by TRANSPAC in France.^[136]

Asynchronous Transfer Mode (ATM) is another virtual circuit technology. It differs from X.25 in that it uses small fixed-length packets (cells), and that the network imposes no flow control to users.

Technologies such as MPLS and the Resource Reservation Protocol (RSVP) create virtual circuits on top of datagram networks. MPLS and its predecessors, as well as ATM, have been called "fast packet" technologies. MPLS, indeed, has been called "ATM without cells".^[137] Virtual circuits are especially useful in building robust failover mechanisms and allocating bandwidth for delay-sensitive applications.

Donald Davies' work on data communications and computer network design became well known in the United States, Europe and Japan and was the "cornerstone" that inspired numerous packet switching networks in the decade following.^{[138][139][140][141][142][143][144][47]}

The history of packet-switched networks can be divided into three overlapping eras: early networks before the introduction of X.25; the X.25 era when many postal, telephone, and telegraph (PTT) companies provided public data networks with X.25 interfaces; and the Internet era which initially competed with the OSI model.^{[145][146][147]}

Research into packet switching at the National Physical Laboratory (NPL) began with a proposal for a wide-area network in 1965,^[23] and a local-area network in 1966.^[148] ARPANET funding was secured in 1966 by Bob Taylor, and planning began in 1967 when he hired Larry Roberts. The NPL network followed by the ARPANET became operational in 1969, the first two networks to use packet switching.^[43][44] Larry Roberts said many of the packet switching networks built in the 1970s were similar "in nearly all respects" to Donald Davies' original 1965 design.^[144]

Before the introduction of X.25 in 1976,^[149] about twenty different network technologies had been developed. Two fundamental differences involved the division of functions and tasks between the hosts at the edge of the network and the network core. In the datagram system, operating according to the end-to-end principle, the hosts have the responsibility to ensure orderly delivery of packets. In the virtual call system, the network guarantees sequenced delivery of data to the host. This results in a simpler host interface but complicates the network. The X.25 protocol suite uses this network type.

AppleTalk is a proprietary suite of networking protocols developed by Apple in 1985 for Apple Macintosh computers. It was the primary protocol used by Apple devices through the 1980s and 1990s. AppleTalk included features that allowed local area networks to be established ad hoc without the requirement for a centralized router or server. The AppleTalk system automatically assigned addresses, updated the distributed namespace, and configured any required inter-network routing. It was a plug-n-play system.^[150][151]

AppleTalk implementations were also released for the IBM PC and compatibles, and the Apple IIGS. AppleTalk support was available in most networked printers, especially laser printers, some file servers and routers.

The protocol was designed to be simple, autoconfiguring, and not require servers or other specialized services to work. These benefits also created drawbacks, as Appletalk tended not to use bandwidth efficiently. AppleTalk support was terminated in 2009.^[150][152]

The ARPANET was a progenitor network of the Internet and one of the first networks, along with ARPA's SATNET, to run the TCP/IP suite using packet switching technologies.

BNRNET was a network which Bell-Northern Research developed for internal use. It initially had only one host but was designed to support many hosts. BNR later made major contributions to the CCITT X.25 project.^[153]

The Cambridge Ring was an experimental ring network developed at the Computer Laboratory, University of Cambridge. It operated from 1974 until the 1980s.

CompuServe developed its own packet switching network, implemented on DEC PDP-11 minicomputers acting as network nodes that were installed throughout the US (and later, in other countries) and interconnected. Over time, the CompuServe network evolved into a complicated multi-tiered network incorporating ATM, Frame Relay, IP and X.25 technologies.

The CYCLADES packet switching network was a French research network designed and directed by Louis Pouzin. First demonstrated in 1973, it was developed to explore alternatives to the early ARPANET design and to support network research generally. It was the first network to use the end-to-end principle and make the hosts responsible for reliable delivery of data, rather than the network itself. Concepts of this network influenced later ARPANET architecture.^[154][155]

DECnet is a suite of network protocols created by Digital Equipment Corporation, originally released in 1975 in order to connect two PDP-11 minicomputers.^[156] It evolved into one of the first peer-to-peer network architectures, thus transforming DEC into a networking powerhouse in the 1980s. Initially built with three layers, it later (1982) evolved into a seven-layer OSI-compliant networking protocol. The DECnet protocols were designed entirely by Digital Equipment Corporation. However, DECnet Phase II (and later) were open standards with published specifications, and several implementations were developed outside DEC, including one for Linux.

DDX-1 was an experimental network from Nippon PTT. It mixed circuit switching and packet switching. It was succeeded by DDX-2.^[157]

The European Informatics Network (EIN), originally called COST 11, was a project beginning in 1971 to link networks in Britain, France, Italy, Switzerland and Euratom. Six other European countries also participated in the research on network protocols. Derek Barber directed the project, and Roger Scantlebury led the UK technical contribution; both were from NPL.^{[158][159][160][161]} The contract for its implementation was awarded to an Anglo French consortium led by the UK systems house Logica and Sesa and managed by Andrew Karney. Work began in 1973 and it became operational in 1976 including nodes linking the NPL network and CYCLADES.^[162] Barber proposed and implemented a mail protocol for EIN.^[163] The transport protocol of the EIN helped to launch the INWG and X.25 protocols.^{[164][165][166]} EIN was replaced by Euronet in 1979.^[167]

The Experimental Packet Switched Service (EPSS) was an experiment of the UK Post Office Telecommunications. It was the first public data network in the UK when it began operating in 1976.^[168] Ferranti supplied the hardware and software. The handling of link control messages (acknowledgements and flow control) was different from that of most other networks.^{[169][170][171]}

As General Electric Information Services (GEIS), General Electric was a major international provider of information services. The company originally designed a telephone network to serve as its internal (albeit continent-wide) voice telephone network.

In 1965, at the instigation of Warner Sinback, a data network based on this voice-phone network was designed to connect GE's four computer sales and service centers (Schenectady, New York, Chicago, and Phoenix) to facilitate a computer time-sharing service.

After going international some years later, GEIS created a network data center near Cleveland, Ohio. Very little has been published about the internal details of their network. The design was hierarchical with redundant communication links.^[172][173]

IPSANET was a semi-private network constructed by I. P. Sharp Associates to serve their time-sharing customers. It became operational in May 1976.^[174]

The Internetwork Packet Exchange (IPX) and Sequenced Packet Exchange (SPX) are Novell networking protocols from the 1980s derived from Xerox Network Systems' IDP and SPP protocols, respectively which date back to the 1970s. IPX/SPX was used primarily on networks using the Novell NetWare operating systems.^[175]

Merit Network, an independent nonprofit organization governed by Michigan's public universities,^[176] was formed in 1966 as the Michigan Educational Research Information Triad to explore computer networking between three of Michigan's public universities as a means to help the state's educational and economic development.^[177] With initial support from the State of Michigan and the National Science Foundation (NSF), the packet-switched network was first demonstrated in December 1971 when an interactive host-to-host connection was made between the IBM mainframe systems at the University of Michigan in Ann Arbor and Wayne State University in Detroit.^[178] In October 1972, connections to the CDC mainframe at Michigan State University in East Lansing completed the triad. Over the next several years, in addition to host-to-host interactive connections, the network was enhanced to support terminal-to-host connections, host-to-host batch connections (remote job submission, remote printing, batch file transfer), interactive file transfer, gateways to the Tymnet and Telenet public data networks, X.25 host attachments, gateways to X.25 data networks, Ethernet attached hosts, and eventually TCP/IP; additionally, public universities in Michigan joined the network.^[178][179] All of this set the stage for Merit's role in the NSFNET project starting in the mid-1980s.

Donald Davies of the National Physical Laboratory (United Kingdom) designed and proposed a national commercial data network based on packet switching in 1965.^[180][181] The proposal was not taken up nationally but the following year, he designed a local network using "interface computers", today known as routers, to serve the needs of NPL and prove the feasibility of packet switching.^[182]

By 1968 Davies had begun building the NPL network to meet the needs of the multidisciplinary laboratory and prove the technology under operational conditions.^{[183][45][184]} In 1969, the NPL, followed by the ARPANET, were the first two networks to use packet switching.^[185][44] By 1976, 12 computers and 75 terminal devices were attached,^[186] and more were added until the network was replaced in 1986. NPL was the first to use high-speed links.^{[187][188][189]}

Octopus was a local network at Lawrence Livermore National Laboratory. It connected sundry hosts at the lab to interactive terminals and various computer peripherals including a bulk storage system.^{[190][191][192]}

Philips Research Laboratories in Redhill, Surrey developed a packet switching network for internal use. It was a datagram network with a single switching node.^[193]

PARC Universal Packet (PUP or Pup) was one of the two earliest internetworking protocol suites; it was created by researchers at Xerox PARC in the mid-1970s. The entire suite provided routing and packet delivery, as well as higher level functions such as a reliable byte stream, along with numerous applications. Further developments led to Xerox Network Systems (XNS).^[194]

RCP was an experimental network created by the French PTT. It was used to gain experience with packet switching technology before the specification of TRANSPAC was frozen.^[195] RCP was a virtual-circuit network in contrast to CYCLADES which was based on datagrams. RCP emphasised terminal-to-host and terminal-to-terminal connection; CYCLADES was concerned with host-to-host communication. RCP influenced the X.25 specification, which was deployed on TRANSPAC and other public data networks.^{[196][197][198]}

Red Especial de Transmisión de Datos (RETD) was a network developed by Compañía Telefónica Nacional de España. It became operational in 1972 and thus was the first public network.^{[199][200][201][202]}

"The experimental packet-switched Nordic telecommunication network SCANNET was implemented in Nordic technical libraries in the 1970s, and it included first Nordic electronic journal Extemplo. Libraries were also among first ones in universities to accommodate microcomputers for public use in the early 1980s."^[203]

SITA is a consortium of airlines. Its High Level Network (HLN) became operational in 1969. Although organised to act like a packet-switching network,^[23] it still used message switching.^[204][205] As with many non-academic networks, very little has been published about it.

A number of computer facilities serving the Science Research Council (SRC) community in the United Kingdom developed beginning in the early 1970s. Each had their own star network (ULCC London, UMRCC Manchester, Rutherford Appleton Laboratory). There were also regional networks centred on Bristol (on which work was initiated in the late 1960s) followed in the mid-late 1970s by Edinburgh, the Midlands and Newcastle. These groups of institutions shared resources to provide better computing facilities than could be afforded individually. The networks were each based on one manufacturer's standards and were mutually incompatible and overlapping.^{[206][207][208]} In 1981, the SRC was renamed the Science and Engineering Research Council (SERC). In the early 1980s a standardisation and interconnection effort started, hosted on an expansion of the SERCnet research network and based on the Coloured Book protocols, later evolving into JANET.^{[209][210][211]}

Systems Network Architecture (SNA) is IBM's proprietary networking architecture created in 1974. An IBM customer could acquire hardware and software from IBM and lease private lines from a common carrier to construct a private network.^[212]

Telenet was the first FCC-licensed public data network in the United States. Telenet was incorporated in 1973 and started operations in 1975. It was founded by Bolt Beranek & Newman with Larry Roberts as CEO as a means of making packet switching technology public. Telenet initially used a proprietary Virtual circuit host interface, but changed it to X.25 and the terminal interface to X.29 after their standardization in CCITT.^[88] It went public in 1979 and was then sold to GTE.^[213][214]

Tymnet was an international data communications network headquartered in San Jose, CA. In 1969, it began install a network based on minicomputers to connect timesharing terminals to its central computers. The network used store-and-forward and voice-grade lines. Routing was not distributed, rather it was established by a central supervisor on a call-by-call basis.^[23]

There were two kinds of X.25 networks. Some such as DATAPAC and TRANSPAC were initially implemented with an X.25 external interface. Some older networks such as TELENET and TYMNET were modified to provide a X.25 host interface in addition to older host connection schemes. DATAPAC was developed by Bell-Northern Research which was a joint venture of Bell Canada (a common carrier) and Northern Telecom (a telecommunications equipment supplier). Northern Telecom sold several DATAPAC clones to foreign PTTs including the Deutsche Bundespost. X.75 and X.121 allowed the interconnection of national X.25 networks.

AUSTPAC was an Australian public X.25 network operated by Telstra. Established by Telstra's predecessor Telecom Australia in the early 1980s, AUSTPAC was Australia's first public packet-switched data network and supported applications such as on-line betting, financial applications—the Australian Tax Office made use of AUSTPAC—and remote terminal access to academic institutions, who maintained their connections to AUSTPAC up until the mid-late 1990s in some cases. Access was via a dial-up terminal to a PAD, or, by linking a permanent X.25 node to the network.^[215]

ConnNet was a network operated by the Southern New England Telephone Company serving the state of Connecticut.^[216][217] Launched on March 11, 1985, it was the first local public packet-switched network in the United States.^[218]

Datanet 1 was the public switched data network operated by the Dutch PTT Telecom (now known as KPN). Strictly speaking Datanet 1 only referred to the network and the connected users via leased lines (using the X.121 DNIC 2041), the name also referred to the public PAD service Telepad (using the DNIC 2049). And because the main Videotex service used the network and modified PAD devices as infrastructure the name Datanet 1 was used for these services as well.^[219]

DATAPAC was the first operational X.25 network (1976).^[220] It covered major Canadian cities and was eventually extended to smaller centers.^{[citation needed]}

Deutsche Bundespost operated the Datex-P national network in Germany. The technology was acquired from Northern Telecom.^[221]

Eirpac is the Irish public switched data network supporting X.25 and X.28. It was launched in 1984, replacing Euronet. Eirpac is run by Eircom.^{[222][223][224]}

Nine member states of the European Economic Community contracted with Logica and the French company SESA to set up a joint venture in 1975 to undertake the Euronet development, using X.25 protocols to form virtual circuits. It was to replace EIN and established a network in 1979 linking a number of European countries until 1984 when the network was handed over to national PTTs.^[225][226]

Hitachi designed a private network system for sale as a turnkey package to multi-national organizations.^[when?] In addition to providing X.25 packet switching, message switching software was also included. Messages were buffered at the nodes adjacent to the sending and receiving terminals. Switched virtual calls were not supported, but through the use of logical ports an originating terminal could have a menu of pre-defined destination terminals.^[227]

Iberpac is the Spanish public packet-switched network, providing X.25 services. It was based on RETD which was operational since 1972. Iberpac was run by Telefonica.^[228]

In 1978, X.25 provided the first international and commercial packet-switching network, the International Packet Switched Service (IPSS).

JANET was the UK academic and research network, linking all universities, higher education establishments, and publicly funded research laboratories following its launch in 1984.^[229] The X.25 network, which used the Coloured Book protocols, was based mainly on GEC 4000 series switches, and ran X.25 links at up to 8 Mbit/s in its final phase before being converted to an IP-based network in 1991. The JANET network grew out of the 1970s SRCnet, later called SERCnet.^[230]

Packet Switch Stream (PSS) was the Post Office Telecommunications (later to become British Telecom) national X.25 network with a DNIC of 2342. British Telecom renamed PSS Global Network Service (GNS), but the PSS name has remained better known. PSS also included public dial-up PAD access, and various InterStream gateways to other services such as Telex.

REXPAC was the nationwide experimental packet switching data network in Brazil, developed by the research and development center of Telebrás, the state-owned public telecommunications provider.^[231]

SITA is a consortium of airlines. Its Data Transport Network adopted X.25 in 1981, becoming the world's most extensive packet-switching network.^{[232][233][234]} As with many non-academic networks, very little has been published about it.

TRANSPAC was the national X.25 network in France.^[136] It was developed locally at about the same time as DATAPAC in Canada. The development was done by the French PTT and influenced by the experimental RCP network.^[195] It began operation in 1978, and served commercial users and, after Minitel began, consumers.^[235]

Tymnet utilized virtual call packet switched technology including X.25, SNA/SDLC, BSC and ASCII interfaces to connect host computers (servers) at thousands of large companies, educational institutions, and government agencies. Users typically connected via dial-up connections or dedicated asynchronous serial connections. The business consisted of a large public network that supported dial-up users and a private network business that allowed government agencies and large companies (mostly banks and airlines) to build their own dedicated networks. The private networks were often connected via gateways to the public network to reach locations not on the private network. Tymnet was also connected to dozens of other public networks in the U.S. and internationally via X.25/X.75 gateways.^[236][237]

UNINETT was a wide-area Norwegian packet-switched network established through a joint effort between Norwegian universities, research institutions and the Norwegian Telecommunication administration. The original network was based on X.25; Internet protocols were adopted later.^[238]

VENUS-P was an international X.25 network that operated from April 1982 through March 2006. At its subscription peak in 1999, VENUS-P connected 207 networks in 87 countries.^[239]

Xerox Network Systems (XNS) was a protocol suite promulgated by Xerox, which provided routing and packet delivery, as well as higher level functions such as a reliable stream, and remote procedure calls. It was developed from PARC Universal Packet (PUP).^[240][241]

Internet
An Opte Project visualization of routing paths through a portion of the Internet
General Access Activism Censorship Data activism Democracy Digital divide Digital rights Freedom Freedom of information Internet phenomena Net neutrality Privacy Right to Internet access Slacktivism Sociology Usage Vigilantism Virtual community Virtual volunteering
Governance IGF NRO IANA ICANN IETF ISOC
Information infrastructure Domain Name System Hypertext Transfer Protocol Internet exchange point Internet protocol suite Internet Protocol Transmission Control Protocol Internet service provider IP address Internet Message Access Protocol Simple Mail Transfer Protocol
Services Blogs Microblogging Email Fax File sharing File transfer Games Instant messaging Podcasts Shopping Television Voice over IP World Wide Web search
History History of the Internet Oldest domain names Pioneers Protocol Wars
Guides Index Outline
Internet portal
v t e

When Internet connectivity was made available to anyone who could pay for an Internet service provider subscription, the distinctions between national networks blurred. The user no longer saw network identifiers such as the DNIC. Some older technologies such as circuit switching have resurfaced with new names such as fast packet switching. Researchers have created some experimental networks to complement the existing Internet.^[242]

The Computer Science Network (CSNET) was a computer network funded by the NSF that began operation in 1981. Its purpose was to extend networking benefits for computer science departments at academic and research institutions that could not be directly connected to ARPANET due to funding or authorization limitations. It played a significant role in spreading awareness of, and access to, national networking and was a major milestone on the path to the development of the global Internet.^[243][244]

Internet2 is a not-for-profit United States computer networking consortium led by members from the research and education communities, industry, and government.^[245] The Internet2 community, in partnership with Qwest, built the first Internet2 Network, called Abilene, in 1998 and was a prime investor in the National LambdaRail (NLR) project.^[246] In 2006, Internet2 announced a partnership with Level 3 Communications to launch a brand new nationwide network, boosting its capacity from 10 to 100 Gbit/s.^[247] In October, 2007, Internet2 officially retired Abilene and now refers to its new, higher capacity network as the Internet2 Network.

The National Science Foundation Network (NSFNET) was a program of coordinated, evolving projects sponsored by the NSF beginning in 1985 to promote advanced research and education networking in the United States.^[248] NSFNET was also the name given to several nationwide backbone networks, operating at speeds of 56 kbit/s, 1.5 Mbit/s (T1), and 45 Mbit/s (T3), that were constructed to support NSF's networking initiatives from 1985 to 1995. Initially created to link researchers to the nation's NSF-funded supercomputing centers, through further public funding and private industry partnerships it developed into a major part of the Internet backbone.

In addition to the five NSF supercomputer centers, NSFNET provided connectivity to eleven regional networks and through these networks to many smaller regional and campus networks in the United States. The NSFNET regional networks were:^[249][250]

• BARRNet, the Bay Area Regional Research Network in Palo Alto, California;
• CERFnet, California Education and Research Federation Network in San Diego, California, serving California and Nevada;
• CICNet, the Committee on Institutional Cooperation Network via the Merit Network in Ann Arbor, Michigan and later as part of the T3 upgrade via Argonne National Laboratory outside of Chicago, serving the Big Ten Universities and the University of Chicago in Illinois, Indiana, Michigan, Minnesota, Ohio, and Wisconsin;
• Merit/MichNet in Ann Arbor, Michigan serving Michigan, formed in 1966,^[251] still in operation as of 2023^[update];^[252]
• MIDnet in Lincoln, Nebraska serving Arkansas, Iowa, Kansas, Missouri, Nebraska, Oklahoma, and South Dakota;
• NEARNET, the New England Academic and Research Network in Cambridge, Massachusetts, added as part of the upgrade to T3, serving Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, and Vermont, established in late 1988, operated by BBN under contract to MIT, BBN assumed responsibility for NEARNET on 1 July 1993;^[253]
• NorthWestNet in Seattle, Washington, serving Alaska, Idaho, Montana, North Dakota, Oregon, and Washington, founded in 1987;^[254]
• NYSERNet, New York State Education and Research Network in Ithaca, New York;
• JVNCNet, the John von Neumann National Supercomputer Center Network in Princeton, New Jersey, serving Delaware and New Jersey;
• SESQUINET, the Sesquicentennial Network in Houston, Texas, founded during the 150th anniversary of the State of Texas;
• SURAnet, the Southeastern Universities Research Association network in College Park, Maryland and later as part of the T3 upgrade in Atlanta, Georgia serving Alabama, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, South Carolina, Tennessee, Virginia, and West Virginia, sold to BBN in 1994; and
• Westnet in Salt Lake City, Utah and Boulder, Colorado, serving Arizona, Colorado, New Mexico, Utah, and Wyoming.

The National LambdaRail (NRL) was launched in September 2003. It is a 12,000-mile high-speed national computer network owned and operated by the US research and education community that runs over fiber-optic lines. It was the first transcontinental 10 Gigabit Ethernet network. It operates with an aggregate capacity of up to 1.6 Tbit/s and a 40 Gbit/s bitrate.^[255][256] NLR ceased operations in March 2014.

TransPAC2 is a high-speed international Internet service connecting research and education networks in the Asia-Pacific region to those in the US.^[257] TransPAC3 is part of the NSF's International Research Network Connections (IRNC) program.^[258]

The Very high-speed Backbone Network Service (vBNS) came on line in April 1995 as part of a NSF sponsored project to provide high-speed interconnection between NSF-sponsored supercomputing centers and select access points in the United States.^[259] The network was engineered and operated by MCI Telecommunications under a cooperative agreement with the NSF. By 1998, the vBNS had grown to connect more than 100 universities and research and engineering institutions via 12 national points of presence with DS-3 (45 Mbit/s), OC-3c (155 Mbit/s), and OC-12 (622 Mbit/s) links on an all OC-12 backbone, a substantial engineering feat for that time. The vBNS installed one of the first ever production OC-48 (2.5 Gbit/s) IP links in February 1999 and went on to upgrade the entire backbone to OC-48.^[260]

In June 1999 MCI WorldCom introduced vBNS+ which allowed attachments to the vBNS network by organizations that were not approved by or receiving support from NSF.^[261] After the expiration of the NSF agreement, the vBNS largely transitioned to providing service to the government. Most universities and research centers migrated to the Internet2 educational backbone. In January 2006, when MCI and Verizon merged,^[262] vBNS+ became a service of Verizon Business.^[263]

• Multi-bearer network
• Optical burst switching
• Packet radio
• Transmission delay
• Virtual private network

• Abbate, Janet (2000). Inventing the Internet. MIT Press. ISBN 9780262511155.
• Gillies, James; Cailliau, Robert (2000). How the Web was born : the story of the World Wide Web. Oxford: Oxford University Press. ISBN 0-19-286207-3. OCLC 43377073.
• Hafner, Katie (1996). Where Wizards Stay Up Late. Simon and Schuster. pp. 52–67. ISBN 9780684832678.
• Norberg, Arthur; O'Neill, Judy E. (2000). Transforming Computer Technology: Information Processing for the Pentagon, 1962-1982. Johns Hopkins University. ISBN 978-0801863691.
• Moschovitis, Christos J. P. (1999). History of the Internet: A Chronology, 1843 to the Present. ABC-CLIO. ISBN 978-1-57607-118-2.
• Lawrence Roberts, The Evolution of Packet Switching (Proceedings of the IEEE, November, 1978)

• Paul Baran et al., On Distributed Communications, Volumes I-XI Archived 2011-03-29 at the Wayback Machine (RAND Corporation Research Documents, August, 1964)
• Paul Baran, On Distributed Communications: I Introduction to Distributed Communications Network (RAND Memorandum RM-3420-PR. August 1964)
• Paul Baran, On Distributed Communications Networks, (IEEE Transactions on Communications Systems, Vol. CS-12 No. 1, pp. 1–9, March 1964)
• D. W. Davies, K. A. Bartlett, R. A. Scantlebury, and P. T. Wilkinson, A digital communications network for computers giving rapid response at remote terminals (ACM Symposium on Operating Systems Principles. October 1967)
• R. A. Scantlebury, P. T. Wilkinson, and K. A. Bartlett, The design of a message switching Centre for a digital communication network (IFIP 1968)

• Pelkey, James L.; Russell, Andrew L.; Robbins, Loring G. (2022). Circuits, Packets, and Protocols: Entrepreneurs and Computer Communications, 1968-1988. Morgan & Claypool. ISBN 978-1-4503-9729-2.
• Russell, Andrew L. (2014). Open Standards and the Digital Age: History, Ideology, and Networks. Cambridge University Press. ISBN 978-1-139-91661-5.

• Wilkinson, Peter (Summer 2020), "Packet Switching and the NPL Network", Computer Resurrection: The Journal of the Computer Conservation Society (90), ISSN 0958-7403
• Oral history interview with Paul Baran. Charles Babbage Institute University of Minnesota, Minneapolis. Baran describes his working environment at RAND, as well as his initial interest in survivable communications, and the evolution, writing and distribution of his eleven-volume work, "On Distributed Communications". Baran discusses his interaction with the group at ARPA who were responsible for the later development of the ARPANET.
• NPL Data Communications Network NPL video, 1970s
• Packet Switching History and Design, site reviewed by Baran, Roberts, and Kleinrock
• Paul Baran and the Origins of the Internet
• 20+ articles on packet switching in the 1970s, archived from the original on 2009-08-01
• "An Introduction to Packet Switched Networks". Phrack. May 3, 1988. Archived from the original on 2023-12-04.