Re[2]: Comparison of high-level concurrency models?

On Tue, 27 May 2008, Gregg Irwin wrote:

> This being the prag list, and hoping to save my own fingers from as
> much crispiness as possible, what would be the "harsh, but simplifying
> restrictions" you would impose, or what models/lanugages worked for
> you (or didn't)? My context is more distributed workflow oriented than
> pure HPC, but any thoughts based on your experience would be most
> welcome.

Harsh but simplifying...

Basically choose one of the models like mapreduce XOR message passing
XOR ... and stick RELIGIOUSLY with it.

Attempting to kludge the odd thing with backdoors like passing data
via globals or shm segments, ad hoc synchronization techniques like
semaphores etc. etc as well is the sure route to crispy finger hell.

Some frameworks, especially by Big Vendors offer you The Works. Every
data passing technique, every synchronization technique in the
textbooks.

Madness lies that way.

Moores Law has been upgrading our CPU's for the last 20 years, but
it's done sweet nothing for our brains.

When we go concurrent, we need to go simple and consistent. Keep It
Simple Stupid. Very simple.

A basic rule of thumb is if a sequence of events happened in one
order, but can arrive on any subsystem out in another order, you're
doing it wrong and it's going to bite you in the bum when you aren't
looking.


John Carter                             Phone : (64)(3) 358 6639
Tait Electronics                        Fax   : (64)(3) 359 4632
PO Box 1645 Christchurch                Email : [email protected]
New Zealand