Computer
networking as we know it today may be said to have gotten its start with the
Arpanet development in the late 1960s and early 1970s. Prior to that time there
were computer vendor “networks” designed primarily to connect terminals and
remote job entry stations to a mainframe. But the notion of networking between
computers viewing each other as equal peers to achieve “resource sharing” was
fundamental to the ARPAnet design. The other strong emphasis of the
Arpanet work was its reliance on the then novel technique of packet switching
to efficiently share communication resources among “bursty” users, instead of
the more traditional message or circuit switching.
Computer
networking as we know it today may be said to have gotten its start with the
ARPANET development in the late 1960s and early 1970s. Prior to that time there
were computer vendor" networks" designed primarily to connect
terminals and remote job entry stations to a mainframe. But the notion of networking
between computers viewing each other as equal peers to achieve "resource
sharing" was fundamental to the ARPANET design. The other strong emphasis
of the ARPANET work was its reliance on the then novel technique of packet
switching to efficiently share communication resources among" bursty"
users, instead of the more traditional message or circuit switching. Although
the term "network architecture" was not yet widely used, the initial
ARPANET design did have a definite structure and introduced another key concept:
protocol layering, or the idea that the total communications functions could be
divided into several layers, each building upon the services of the one below.
The original design had three major layers, a network layer, which included the
network access and switch-to-switch (IMP-to-IMP) protocols, a host-to-host
layer (the Network Control Protocol or NCP), and a "function-oriented
protocol" layer, where specific applications such as file transfer, mail,
speech, and remote terminal support were provided. Similar ideas were being
pursued in several other research projects around the world, including the
Cyclades network in France, the
National
Physical Laboratory Network in England, and the Ethernet system at
Xerox PARC in the USA. Some of these projects focused more heavily on the
potential for high-speed local networks such as the early 3-Mbps Ethernet.
Satellite and radio channels for mobile users were also a topic of growing
interest. By 1973 it was clear to the networking vanguard that another protocol
layer needed to be inserted into the protocol hierarchy to accommodate the
interconnection of diverse types of individual networks. Cerf and Kahn
published their seminal paper describing such a scheme, and development of
the new Internet Protocol (IP) and Transmission Control Protocol (TCP) to
jointly replace the NCP began. Similar work was being pursued by other groups
meeting in the newly formed IFIP WG 6.1, called the Internetwork Working Group. The basis for the network interconnection approach developing in this
community was to make use of a variety of individual networks each providing
only a simple "best effort" or "datagram" transmission
service. Reliable virtual circuit services would then be provided on an
end-to-end basis with the TCP (or similar protocol) in the hosts. During the
same time, public data networks (PDNs) were emerging under the auspices of
CCITT, aimed at providing more traditional virtual circuit types of network
service via the newly defined X.25 protocol. The middle and late 1970s saw
networking conferences dominated by heated debates over the relative merits of
circuit versus packet switching and datagrams versus X.25 virtual circuits.
The computer vendors continued to offer their proprietary networks, gradually
supporting the new X.25 service as links under their own protocols. Digital
Equipment Corporation (DEC) was the notable exception, adopting the research
community approach of peer-to-peer networking at an early date, and coming out with
its own new suite of protocols (DECNET). By the late 1970s, a new major
influence was emerging in the computer network community. The computer
manufacturers realized that multivendor systems could no longer be avoided and
began to take action to satisfy the growing user demand for interoperability.
Working through their traditional international body, the ISO, a new group
(SC16) was created to develop standards in the networking area. Their initial
charter was to define an explicit "architecture" for "Open
Systems Interconnection" (OSI). By the early 1980s there were three major
players in the networking game: the ARPANET-style research community, the
carriers with their PDNs in CCITT, and the manufacturers in ISO. The conference
circuit became more acrimonious, with the research community lambasting the
slow progress, ponderousness (7 layers!), lack of experimental support, and
all-inclusiveness (five classes of transport protocol) of the ISO workers,
while still taking occasional shots at the PDNs and X.25. The CCITT.
Comments
Post a Comment