Xref: utzoo comp.edu:2897 comp.lang.ada:3165 comp.lang.misc:3872
Path: utzoo!attcan!uunet!mcsun!ukc!inmos!mph
From: mph@oberon.inmos.co.uk (Mike Harrison)
Newsgroups: comp.edu,comp.lang.ada,comp.lang.misc
Subject: Re: Teaching Concurrency
Message-ID: <3569@ganymede.inmos.co.uk>
Date: 12 Jan 90 13:12:19 GMT
References: <7588@hubcap.clemson.edu> <602@agcsun.UUCP> <33100@watmath.waterloo.edu>
Sender: news@inmos.co.uk
Reply-To: mph@inmos.co.uk (Mike Harrison)
Organization: INMOS Limited, Bristol, UK.
Lines: 33

In article <33100@watmath.waterloo.edu> sccowan@watmsg.waterloo.edu (S. Crispin Cowan) writes:
>When to use tasking critically depends on the cost of the tasking
>operations.


Not necessarily so!

There are different possible goals in writing a program, run-time efficiency
is only one, another may be 'write-time' efficiency.

Ada tasking is, (in addition to permitting an implementation to allow the
use of multi-processor systems), an abstraction mechanism for the convenient
expression of concurrency. If a user can express his needs more easily and
effectively using tasking constructs, he may not be (seriously) concerned
about the running speed of the program.

For example, in a previous job, I had to write simulation of the behaviour 
of a multi-processor system (in executing Ada tasking as it happens), using
1, 2, 3 or 4 cpus.
It was much easier to represent the processors as separate tasks and to use
redzvous to simulate interactions than to write a single serial program with
explicit handling of the (simulated) paralellism.

However, I didn't care if the program took twice as long to run.


Mike.


Michael P. Harrison - Software Group - Inmos Ltd. UK.
-----------------------------------------------------------
UK : mph@inmos.co.uk             with STANDARD_DISCLAIMERS;
US : mph@inmos.com               use  STANDARD_DISCLAIMERS;