[erlang-questions] Message passing via shared memory

Hynek Vychodil <>
Wed Feb 4 15:03:04 CET 2009


I agree with you. I would like mentioned also GC issues which appear even in
one core. Current Erlang GC blocks only one process each time and provide
soft realtime behavior of Erlang system. IMHO for example SBCL
Lisp/Hunchentoot problems at Chart-6 in this blog
http://berlinbrowndev.blogspot.com/2008/08/application-server-benchmarks-including.htmlcomes
from global GC. See Erlang response time deviation in first table for
comparison. If share memory between proceses soft realtime characteristic of
GC will gone and/or GC complexity will rise.

On Wed, Feb 4, 2009 at 2:09 PM, Andras Georgy Bekes <>wrote:

> > I'm pretty interested in this method, because while memory copying is
> > OK for MPP or clustered environments, its is much slower than memory
> > sharing on SMP environments if you are deailing with certain amount
> > of data.
> Memory sharing is faster only if the number of processors (cores) is
> small. As the number of cores increases, the environment is getting
> similar to a cluster.
>
> On the long term, going into a shared memory direction is a bad idea I
> think.
>
>        Georgy
> _______________________________________________
> erlang-questions mailing list
> 
> http://www.erlang.org/mailman/listinfo/erlang-questions
>



-- 
--Hynek (Pichi) Vychodil

Analyze your data in minutes. Share your insights instantly. Thrill your
boss.  Be a data hero!
Try Good Data now for free: www.gooddata.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://erlang.org/pipermail/erlang-questions/attachments/20090204/467e22ff/attachment.html>


More information about the erlang-questions mailing list