![]() |
![]() |
||
| THINK Home | Digital Archive | Project Management | Feedback |
|
|
The THINK Protocols site is designed for networking students and researchers, who have an interest in understanding where the field of networking has been in order to more effectively understand the present and design networks of the future. (Currently, these web pages are in the prototype phase, and have been written by CS seniors as a semester-long research experience.) The materials range from mildly technical to quite technical. This project is very much a networking community effort, with contributions ranging:
Specific Network ArchitecturesAll of the following descriptions are prototypes of what will eventually be developed, plus the list will eventually be expanded to other network architectures.
1960-1964: While at RAND Corporation in California, Paul Baran and others designed a distributed digital data communications network. A far-reaching design goal was that the network should withstand heavy enemy attacks. While his solution is now known as a packet switched network, he named it "distributed adaptive message block switching".
1966-1970: At the National Physics Laboratory in the UK, Donald Davies and others designed and built a packet switched network, independent of the work by Baran. Davies' original design introduced the now familiar term: packet. The team then went on to build one of the first local area networks and to design the classic error control protocol: the alternating bit protocol.
1967-198?: Initiated by ARPA project managers Bob Taylor and Larry Roberts, and designed and built by a Boston firm BBN (Bolt Beranek and Newman) and a large community of software designers known as the NWG (Network working group), the ARPANET was definitively the first large-scale example of a packet switched network. A high point for the ARPANET was a 1972 public demonstration of the ARPANET that also proved to the networking community that packet switched networks (versus telephone style circuit switched networks) were truly viable. The ARPANET design drew upon some of the work by RAND and NPL, but the ARPANET differs in that its routers (called IMPs) were additionally responsible for reliable delivery between the original source IMP and the final destination IMP.
1972-197?: ...
|
![]() |
Written by the THINK Protocols
team, CS Dept.,
UT
Austin
Please direct comments to Chris Edmondson-Yurkanan. This document was last modified on Thursday, 26-Jul-2001 17:22:52 CDT. |