[comp.parallel] My Parallelism better than your Parallelism?

gkj@doc.imperial.ac.uk (Guido K Jouret) (02/02/90)

Considering all the current research on different avenues to parallel
processing, wouldn't it be nice if we could agree on some standard
terminology regarding parallelism?

In particular, I'm interested in collecting definitions/examples
of the different "kinds" of parallelism mentioned in the literature.
It seems that everyone has their own set of terms for describing
the kind of parallelism they are aiming to exploit.

Some examples:

Algorithmic Parallelism
Geometric Parallelism
Data Parallelism
Pipeline Parallelism
Horizontal Parallelism
Vertical Parallelism
Speculative Parallelism
Application-Level Parallelism
Outer-Loop Parallelism
Concurrency
Multi-processing

etc. etc. etc.

If you send me the name of your favourite "flavour" of parallelism (i.e.
one I've mentioned or perhaps one I haven't...) and a suitable definition
(in particular what makes it distinct from other sorts of parallelism)
coupled with a short example, I'll compile the answers and post a
report summarizing the results.

Thanks in advance,

gkj...

 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~+< Guido K. Jouret >+~~~~~~~~~~~~~~~~~~~~~~~~~~~
/ email:  gkj@uk.ac.ic.doc                         =  Humor is like a frog:
| rmail:  Functional Programming Section           =                           
|         Dept. of Computing                       =  It can be dissected, but
|         Imperial College                         =    usually dies in the
|         London SW7 2AZ                           =         process.
|         U.K.                                     = 
\   tel:  44-1-589-5111 xt: 7532                   =
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~