nivtwig wrote:

Search using Google!
http://www.google.com/
(How-to: http://www.google.com/intl/en/help/basics.html )

Example, if you searched for "optimize pagefile usage in Windows XP" you
might come across:
http://www.petri.co.il/pagefile_optimization.htm

Enjoy!

You wrote : "Most things that get put in the pagefile are put there
because they are currently unused. "

That is not true !!!.
At first impression, and to the unsuspecting user, this answer of
Microsoft and others I have seen (on my searches on the internet)
sounds logical and right, but it isn't. Here's why: In theory and
according to what you said, pages that haven't been used for some time
(or are not currently used in your words) should be swapped to disk.
But : That is not what is happening in reality !!!!.
In practice, what happens in Windows XP is: When I launch Firefox (or
launch any other application with high mem usage) at the first time
with 60 tabs (Firefox restores them automatically), I see on Task
Manager that the process total memory usage grows slowly to 600MB, and
as it grows, about half of the memory  is swapped to disk (Growing
slowly to 300MB), although it was only seconds since I launched it.
How can windows determine that it is a "currently unused memory" if
only seconds have passed since it was last read/used ? and on top of
that it has a lot of other free available memory, so why should it
swap it ?
It seems that windows always swaps about half of the process total
memory size to disk on relatively large processes, regardless of how
much time the memory was unused. Always half, no matter what.

This is WRONG, the behavior of Windows XP on this issue is wrong, and
that is why I want to disable the page file. I can prove that it
causes unnecessary slowness :It causes many page faults that I can see
in the task manager, and are unnecessary since I know I have enough
memory for all my applications together.

As I said, before I asked the question, I searched a lot on this
newsgroup and on the internet. I have seen many answers and articles
from Microsoft MVPs, with the same answer as you gave, suggesting that
I should not turn it off for several reasons that sound logical, but
when you look at the practice and what really happens, the answers of
Microsoft MVPs on this issue in my opinion are wrong, it seems you are
all citing each other old answers as reference without really checking
the issue.

To summarize: Your answer didn't answer the question. It suggested why
I shouldn't turn off the page file, but didn't answer why it doesn't
get turned off.
I want to disable it because it is unneeded in my system according to
my usage patterns, and it slows down the system (The page faults I see
means it has to go and bring it from the disk back to memory, slowing
down the computer)

Well, to put it not bluntly: I think you are the one who doesn't
understand page file and VM.
But instead of arguing on definitions, just explain me the following:

Why do I see on the task manager the following info for the process of
Firefox, and what is the meaning of of it (after I disabled the Page
file on Windows XP):

Image Name | Mem Usage |  VM Size   | Page faults

Firefox.exe    | 344,840K      |  334,716K | 437,513

Notice the "VM Size" column shows a smaller number than Mem size. As I
understand it, In the task manager the "VM size" is the part of the
file that is swapped to the page file. It is not the total Virtual
Memory that is used by the process as usually defined by the articles
you read.

The total memory consumed by the process (which is the "Virtual
Memory" according to the usual defition used in several articles, but
not in Task Manager terminology) is the resident set (344M) plus the
"VM Size" (334M) = ~678M

Why isn't the "VM Size" in Task Manager (which is the the part of the
process Virtual Memory that is swapped to disk) equal to 0 ? That is
what I expected to see.
Why are there so many Page faults?