Internet > History >

TCP/IP Internet Protocol

TCP/IP Digest Wednesday, 14 Oct 1981 Volume 1 : Issue 2
From: POSTEL at USC-ISIF
Subject: NCP-to-TCP Transition

It is really very important for everyone to notice the deadline for completing the cutover to IP/TCP and the elimination of NCP from use in the ARPANET. The deadline is: 1 January 1983. That is 14 and a half months from now. Really not much more than a year.

--jon.

TCP/IP Digest Wednesday, 11 Nov 1981 Volume 1 : Issue 6
From: POSTEL at USC-ISIF
Subject: Disabling NCPs

There has been some talk of "forcing" the move to TCP by various administrative and policy measures. There was also a claim that there was no technical way to force the abandonment of NCP. It should be pointed out that a quite simple modification to the IMP program would enable the IMPs to filter out and discard all NCP traffic. As far as i know, there has been no decision to do this, but you should be aware that it is technical feasible.

Jon


The Internet's open and efficient TCP/IP protocol is the foundation of an inter-networking design has made it the most widely used network protocol in the world.

Who invented TCP/IP? The history of TCP/IP is like the protocol -- interdependent design and development conducted by several people and brought together as one. Following the design of TCP/IP by Robert Kahn and Vinton Cerf as described in the previous pages, DARPA contracted with three sites to develop operational versions:

  • BBN. A TENEX/PDP-10 implementation led by Ray Tomlinson and Bill Plummer.
  • Stanford. A BCPL/PDP-11 implementation led by Richard Karp from Cerf's group.

Four increasingly better versions of TCP/IP were developed -- TCP v1, TCP v2, a split into TCP v3 and IP v3 in the spring of 1978, and then stability with TCP v4 and IPv4 -- the standard protocol still in use on the Internet today.

In 1975, a two-network TCP/IP communications test was performed by establishment of a link between Stanford and University College London (UCL) systems through a Satellite Network (SATNET) connection. In November, 1977, a three-network TCP/IP test was conducted, successfully simulating complex, military-grade communications and verifying the inter-networking functionality inherent in the TCP/IP design:

  • An LSI-11 computer in a van driven down the San Francisco Bayshore Freeway sent the original data over a dual-rate 400/100 kbps spread spectrum Packet Radio Net to
  • A stationary ARPANET gateway at BBN in Boston, which sent the traffic over an internal ARPANET satellite link to
  • A computer in Norway, which sent the data over an ARPANET dedicated circuit to
  • UCL in London, England, which sent the data back across the ocean on the 64 kbps Packet Satellite Net using a shared channel on the Intelsat IV satellite to
  • BBN in Boston, when sent the data across the ARPANET to -
  • a DEC KA-10 computer at the University of Southern California Information Sciences Institute in Marina Del Ray.

Between 1978 and 1983, several TCP/IP research prototypes were developed on the systems listed below, proving the protocol's portability and ability to unify different systems:

System

Leads

Organization
IBM 360/370Bob BradenUCLA
LSI/11 (Packet Radio)Jim MathisSRI
MulticsDave ClarkMIT
PCDave ClarkMIT
PDP-9Peter KirsteinUCL
PDP-11 / BCPL Richard KarpStanford
Tenex/TOPS20Bill Plummer
Ray Tomlinson
BBN
UNIX / CMike WingfieldBBN
Unix / Macro11Jack HavertyBBN

The TCP/IP protocol was deployed to the ARPANET network with some persuasion, as noted in the emails at the top of the page. Not all sites were preparing to convert over their protocols, so Cerf, Jon Postel, and the TCP/IP team turned off the NCP network channel numbers on the ARPANET IMP's for a full day in mid 1982, so that only sites using TCP/IP could still operate. To emphasize the point, they disabled NCP again for two days later that fall. The full switchover to TCP/IP was performed on January 1, 1983, without too many problems, although a few recalcitrant sites were down as long as three months while they retrofitted their systems.

In 1984, the US Department of Defense made TCP/IP the standard for all military computer networking, which gave it a high profile and stable funding. In 1985, Dan Lynch and the Internet Architecture Board held a three day workshop on TCP/IP for the computer industry, which was attended by about 50 researchers and 250 vendor representatives. This meeting helped popularize knowledge of TCP/IP in the computer industry, and triggered the development of several TCP/IP networking products by different companies, starting the protocol on its way to become a commercial standard.

In September, 1988, Lynch organized an Internet convention that later became the Interop trade show. Fifty companies were invited to the first show to demonstrate interoperation of their TCP/IP packages, and five thousand engineers attended. The interoperability demonstration was successful, validating TCP/IP's open design and showing that the network standard could become a multi-vendor product. The Interop show grew tremendously over the next twelve years, held annually in a new location around the world.

TCP/IP, originally inspired by low-reliability wireless packet radio networks, has now become now the most reliable and widely deployed network in the world, and the IPv4 version developed in the 1970's remains the standard protocol in use on the Internet today.

IPv6. With the rapid growth of the Internet through the 1990's, there was a rapid reduction in the number of free IP addresses available under IPv4, which was never designed to scale to planetary levels. To get more addresses you need more bits, which means a longer IP address, which means a new architecture, which means changes to all of the TCP/IP routing software. In other words, a major change on which everyone needs to agree, and does not come about quickly.

After examining a number of proposals, the IETF settled on IPv6, recommended in January 1995 in RFC 1752, and sometimes called the Next Generation Internet Protocol, or IPng, or TCP/IP v6. Since then a number of organizations such as the IPV6 Forum have been working towards its widespread implementation.

By 2004, IPv6 was widely available from industry as an integrated TCP/IP protocol and supported by most new Internet networking equipment. Practical feedback began to be being received from experience with integration with existing TCP/IP v4 networks, for example as discussed in RFC 3901, DNS IPv6 Transport Operational Guidelines, and RFC 3974, IPv6 SMTP Operational Experience in Mixed IPv4/v6 Environments.

Resources. Additional information about the TCP/IP protocol can be found below:

  • IEN 151; Vinton Cerf; Final Report of the Stanford University TCP Project; 1 April 1980.
  • RFC 759; Jon Postel; Internet Message Protocol; August 1980.
  • RFC 760; Jon Postel; DoD Standard Internet Protocol; Jon Postel; January 1980.
  • RFC 761; Jon Postel; DoD Standard Transmission Control Protocol; Jan, 1980.
  • RFC 801; Jon Postel; NCP/TCP Transition Plan; Nov 1981.
  • RFC 1180; T. Socolofsky, C. Kale; A TCP/IP Tutorial; Jan 1991.
  • RFC 3449; H. Balakrishnan, V. N. Padmanabhan, G. Fairhurst, M. Sooriyabandara; TCP Performance Implications of Network Path Asymmetry; December 2002; BCP 69.

IPV6. Some RFC's related to TCP/IP v6 are listed below:

  • RFC 1886; DNS Extensions to support IP version 6; S. Thomson, C. Huitema; December 1995.
  • RFC 2460; Internet Protocol, Version 6 (IPv6) Specification; S. Deering; December 1998.
  • RFC 2893; Transition Mechanisms for IPv6 Hosts and Routers; R. Gilligan; August 2000.
  • RFC 3316; Internet Protocol Version 6 (IPv6) for Some Second and Third Generation Cellular Hosts; J. Arkko, G. Kuijpers, H. Soliman, J. Loughney, J. Wiljakka; April 2003.
  • RFC 3513; Internet Protocol Version 6 (IPv6) Addressing Architecture; R. Hinden; S. Deering; April 2003.

___