Difference between revisions of "Interface Message Processor"

From Computer History Wiki
Jump to: navigation, search
(Add some material)
(Add some links)
Line 3: Line 3:
 
A message from one Host to another was sent via the source Host's IMP, then routed through some number (possibly zero) intermediate IMPs, then through the destination Host's IMP (possibly the same IMP as the first one, in which case the message never went over a telephone line), on to the destination Host.
 
A message from one Host to another was sent via the source Host's IMP, then routed through some number (possibly zero) intermediate IMPs, then through the destination Host's IMP (possibly the same IMP as the first one, in which case the message never went over a telephone line), on to the destination Host.
  
The IMPs were Honeywell [[minicomputer]]s, with added special high-speed modem interfaces, and asynchronous bit-serial Host interfaces.
+
The IMPs were Honeywell DDP-516 [[minicomputer]]s, with added special high-speed modem interfaces, and asynchronous bit-serial Host interfaces. Later, other similar minicomputers were also used.
  
 
==Further reading==
 
==Further reading==
  
 
* F.E. Heart, R.E. Kahn, S.M. Ornstein, W.R. Crowther, and D.C. Walden, "[http://www.walden-family.com/public/1970-imp-afips.pdf The interface message processor for the ARPA computer network]", Proceedings AFIPS, 1970 SJCC, Vol. 36, pp. 551-567.
 
* F.E. Heart, R.E. Kahn, S.M. Ornstein, W.R. Crowther, and D.C. Walden, "[http://www.walden-family.com/public/1970-imp-afips.pdf The interface message processor for the ARPA computer network]", Proceedings AFIPS, 1970 SJCC, Vol. 36, pp. 551-567.
 
+
* J.M. McQuillan, W.R. Crowther, B.P. Cosell, D.C. Walden, and F.E. Heart, "[http://walden-family.com/impcode/1972-improvements-paper.pdf Improvements in the Design and Performance of the ARPA Network]", Proceedings AFIPS, 1972 FJCC, Vol. 40, pp. 741-754.
* J.M. McQuillan, W.R. Crowther, B.P. Cosell, D.C. Walden, and F.E. Heart, "Improvements in the Design and Performance of the ARPA Network", Proceedings AFIPS, 1972 FJCC, Vol. 40, pp. 741-754.
 
 
 
 
* John M. McQuillan, David C. Walden, "[http://www.walden-family.com/public/whole-paper.pdf The ARPA Network Design Decisions]", in "Computer Networks", Vol. 1, No. 5, August 1977, pp. 243-289.
 
* John M. McQuillan, David C. Walden, "[http://www.walden-family.com/public/whole-paper.pdf The ARPA Network Design Decisions]", in "Computer Networks", Vol. 1, No. 5, August 1977, pp. 243-289.
  
 
==External links==
 
==External links==
  
 +
* [http://walden-family.com/impcode/ Resurrection of the 1973 IMP code] - While focused on the retrieval, and running under a simulator, of the original IMP code, this site contains much other IMP-related material too (manuals, etc)
 
* [http://www.chiappa.net/~jnc/tech/arpanet.html ARPANET Technical Information]
 
* [http://www.chiappa.net/~jnc/tech/arpanet.html ARPANET Technical Information]
  
 
{{stub}}
 
{{stub}}

Revision as of 00:17, 16 March 2018

The Interface Message Processor (or IMP, as they were universally known) was the packet switch which the ARPANET was made out of; the IMPs were connected to each other via high-speed dedicated point-point telephone links, connected to the IMPs via a modem. All ARPANET hosts connected to an IMP, using the Host-to-IMP Protocol.

A message from one Host to another was sent via the source Host's IMP, then routed through some number (possibly zero) intermediate IMPs, then through the destination Host's IMP (possibly the same IMP as the first one, in which case the message never went over a telephone line), on to the destination Host.

The IMPs were Honeywell DDP-516 minicomputers, with added special high-speed modem interfaces, and asynchronous bit-serial Host interfaces. Later, other similar minicomputers were also used.

Further reading

External links