[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Fwd: Insecure parallelism with Java?]

     Sadly, Per Brinch Hansen is absolutely correct.
>     As I am writing an application that centres on Java concurrency, I'm interested
> in what experts of various types make of this, especially those involved in building
> JVMs.

I wish you luck - lots of it.  Unfortunately the things that go wrong tend
to do so in a very hard to pin down manner, and do so infrequently - it is
extremely hard to know your program is working correctly :-(

BUT there are ways of using concurrency, even in Java, that are much easier
to tame.

A good place to start with discovering the situation is a paper "Java
threads in the light of occam/CSP" by Peter Welch, published in
"Architectures, Languages and Patterns" Eds P H Welch and A W P Bakkers, IOS
Press, 1998.

There is a lot of work in this area reported to the WOTUG conference (next
meeting: September this year) - the above reference is from the proceedings
of the 1998 conference.

occam is a language that DOES implement all the safety checks that Brinch
Hansen advocates - and works very well in practice. I have a compiler that
takes occam and produces Java, I'm having no trouble at all with ~80
threads. (Note: the occam community is very familiar with programs having
~100 000 threads!)

Take a look at
for details of the conference and links to the large amount of work already
done in this area.


| Barry M Cook, BSc, PhD, CEng, MBCS                                         |
| Senior Lecturer,                           Department of Computer Science, |
| Chartered Information Systems Engineer.    Keele University,               |
|                                            Keele,                          |
| Phone: +44 1782 583411                     Staffordshire,                  |
| FAX:   +44 1782 713082                     ST5 5BG,                        |
| email: barry@xxxxxxxxxxxxxx                UK.                             |