[Top][All Lists]

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

Re: [Chicken-users] google summer of code

From: Ivan Raikov
Subject: Re: [Chicken-users] google summer of code
Date: Thu, 06 Mar 2008 11:08:39 +0900
User-agent: Gnus/5.11 (Gnus v5.11) Emacs/22.1 (gnu/linux)

  Well, I think modifying or rewriting the garbage collector would be
quite a lot of work for one summer, for someone who is presumably not
familiar with the Chicken internals. The rest I agree with. MPI is one
such multi-process model, and although it is often criticized for its
communication overhead, I would really like to see some examples
outside of scientific computing where this overhead really matters.


Alex Shinn <address@hidden> writes:

> I'm not so sure this really would be an intractable amount
> of work.  What I am pretty sure of is that it would be a bad
> idea though.
> As you said, the semaphore/mutex thread model is broken.
> Computer Science has advanced a lot since then, and there
> are new concurrency models which are just as fast and easier
> to program in.  Erlang is well known for its extremely high
> performance, yet this model can be implemented just as
> easily with multi-processes as with multi-threads (in fact,
> since there would be no need to sync the GC or other global
> state it may well be faster as processes).  And if
> implemented as processes you can easily extend to multiple
> machines, which is a *much* bigger win.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]