Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!ucsd!pacbell.com!att!cbnews!cbnews!military
From: budden@trout.nosc.mil (Rex A. Buddenberg)
Newsgroups: sci.military
Subject: Re: Embedded S/W in Desert-Storm weapons
Message-ID: <1991Jan24.034206.20618@cbnews.att.com>
Date: 24 Jan 91 03:42:06 GMT
References: <1991Jan22.015149.20141@cbnews.att.com>
Sender: military@cbnews.att.com (William B. Thacker)
Organization: Naval Ocean Systems Center, San Diego
Lines: 31
Approved: military@att.att.com



From: budden@trout.nosc.mil (Rex A. Buddenberg)

Can't speak for the sand pounders, but most embedded systems in
the Navy ships fall in a couple categories:

Combat Direction Systems (including the fire control for Tomahawk,
gun systems, Aegis, Phalanx (CIWS), and the like) are mostly built from
AN/UYK-43s and AN/UYK-44s.  These monsters are almost univerasally
coded in CMS-2.  
  The early versions of Joint Operations Tactical
System are hosted on commercial HP9020 computers and coded in
Rocky Mountain BASIC [!!].  While the system is 'real time' in
terms of tactical plot, it is not 'real-time' in terms of being
event driven.
  More current versions of JOTS and Tactical Flag Command Center
(same software for both) are hosted on a ruggedized Sun workstation
and mostly coded in C, although there is an incremental migration 
to Ada in progress.  This implementation is not 'embedded' in
terms of  'real time' event driven systems -- the computer keeps up
by sheer horsepower relative to the data loading.  So that may
not meet your definition of embedded, but it is quite adequate
to the mission requirements.

If you're interested in the future, an opinion from a very interested
observer.  There is a bit of controversy going on in NAVSEA right
now with the 'standardized UYK-43's' on one side and the workstations-
on-a-LAN gang on the other.  My money is on the latter.

Rex Buddenberg