From: Marcin Raczkowski <mailing.mr@...>
Date: 2008-01-05T16:17:27+09:00
Subject: Re: Manual Memory Management

sorry for responding to my own post.
I forgot to add that for now there are 2 solutions to "load huge object 
and forget" problem.

1) fork:

3.times{GC.start}; # to use as much COW on linux as we can
unless fork
   # do some woork with large object
   exit!
end

but you can use this if you don't need to have any value from object in 
main program (like hash for example)

2) drb

you start separate process and start DRb server.
This way you can do heavy processing in separate DRb process - which can 
be forced to GC more often, and you can ensure there's no clousures etc.
so you can safetelly free memory your way.
this way you can also return for example hash of data you processed