[mod.os] Classifying/defining dist. comp., parallelism, concurrency

darrell@sdcsvax.UUCP (02/13/87)

Re: terminology
Michael Scott posts:
>I find it useful to keep distinct the meanings of the words "concurrent",
>"parallel", and "distributed."  In the interests of broadening the discussion,
>I suggest the following definitions:
>
>    "Concurrent" implies the simultaneous existence of more than one
>    thread of control.
>
>    "Parallel" implies the simultaneous *execution* of more than one
>    thread of control.
>
>    "Distributed" implies interaction between threads of control on
>    processors that share no physical memory.
>
>Parallel and distributed both imply concurrent.
>Most distributed computations are parallel.
>Coroutines are concurrent but not parallel.

The problem is much worse that this. [Pull out my classic Indian
Blindmen describing Nude elephant slide: an elephant is like a wall,
no it's like a snake, no, you're both wrong, like a tree...]
Other people refer to real-time systems as concurrent systems, also note
Concurrent Computer [formerly Perkin-Elmer], and so forth.

The Cray and other worlds have problems with: multiprogrammed (easy),
multiprocess(or/ing [surprising hard]), multitasking [also surprisingly
hard, lots of name collisions here], micro-tasking, et. al.   And I set
Marty Fouts a little thing about Loosely-coupled versus tightly coupled
processing.  The use of terminology is not only abomitable, but
detrimental.

Exercise left to the reader: do some research akin to parallel or
whatever computing.  Stop using the words "Parallel Process" for a day.
Next day remove the obvious synonyms like "Multiprocess," one day at a
time: distributed process next, etc.  Describe the same concepts.  See
how much you hand wave.  This is akin to the exercise which the guys
(correction: humans) studying light did to understand the duelistic
nature of light (Light was a particle on MWF, a wave on TTS).  Have fun.

>From the Rock of Ages Home for Retired Hackers:

--eugene miya
  NASA Ames Research Center
  eugene@ames-aurora.ARPA
  "You trust the `reply' command with all those different mailers out there?"
  "Send mail, avoid follow-ups.  If enough, I'll summarize."
  {hplabs,hao,nike,ihnp4,decwrl,allegra,tektronix,menlo70}!ames!aurora!eugene