Re: limitations of forking on windows
From: Purl Gurl (purlgurl_at_purlgurl.net)
Date: Tue, 29 Jun 2004 08:21:44 -0700
> > It is not the number of forked processes. It is the
> > amount of system resources used by each forked process.
> right.. and resource usage is fine when I get these errors.
I am curious what technique you used to determine memory
usage on your NT5 machine while "forked" processes were
actively running. Would you mind sharing your methodolgy?
Your comments suggest you do not have a memory usage
problem but rather a limitation within perl core
not allowing sufficient memory usage.
> > Comparing Linux system resource usage and NT5 system
> > resource usage, is not a valid comparison. Linux is
> > very minimalist, NT5 is graphics and services rich.
> my gnu/linux machine running KDE is just as "graphics and services
> rich" as any windows box... I'm not quite sure what you are getting
KDE is an adjunct software. Still, Linux is not as graphics
rich, audio rich nor has as many background processes running
Redhat Personal has become vaporware mostly because of so many
difficulties writing support into core for drivers needed to
run various audiovisual devices wanted by common users.
Linux simply does not support what a majority of consumers
want; games, graphics, music and all that. This is not a
point of critique because Linux is not intended to be
competitive with Win machines. Linux and Windows simply
cannot be compared as they serve very different functions.
> > You are also comparing threading to forking.
> I am comparing the identical piece of code on 2 platforms... even
> though it is handled at the OS level differently (forked or emulated
> forking via threads).
That is correct. Same code but different processes invoked.
You cannot make a valid comparison between Win32 forking
and Linux forking. Those are simply not the same.
I am very curious what methodology you used to check your
NT5 memory usage during your forking processes. Please do
share your code for this. Questions about memory monitoring
appear here on a regular basis. Your sharing of your code
to monitor NT5 memory during program execution would be
a benefit for many readers.