pagefile in two segments?

  • Thread starter Thread starter BobG
  • Start date Start date
B

BobG

Defrag and MyDefrag both show pagefile.sys as a big hunk and a smaller
hunk. Anyway to combine em?
 
If you are letting Windows manage the size of
your pagefile then trying to combine the two is
a waste of time.
 
BobG said:
Defrag and MyDefrag both show pagefile.sys as a big hunk and a smaller
hunk. Anyway to combine em?
Click to expand...

Turn off the page file.

Reboot.

Turn it back on and set the min and max to the same number (don't let
Windows mange it).
 
Useless advice if your not going to advise the OP on how to determine the
size to set it to, he's far better off setting it to system managed anyway.
 
David B. said:
Useless advice if your not going to advise the OP on how to determine the
size to set it to, he's far better off setting it to system managed anyway.

--
Click to expand...

I assumed the OP could read the recommended size. SOME people who ask
questions here aren't complete morons.

Set the way I recommended keeps it from becoming broken like described
and reduces the number of tasks Windows has to do by one.
 
Bennett

If the free disk space on the volume, where the pagefile is to be
placed, is less than 60% you can get problems achieving a single
contiguous file. The reason for this situation is that the system places
the pagefile in the centre of the drive. The figure of 40% - 60% is a
guide but much lower figures like 25% make achieving a single contiguous
file difficult. If the free space available is marginal you can try
removing temporarily a large file(s) (like system restore points) to
temporarily create extra space. To retain a contiguous file it is
necessary to have a fixed minimum = maximum setting.

Using the Windows Managed option on a system with less than 60% free
disk will ultimately mean that the pagefile becomes fragmented, which of
itself is not so important, but it causes difficulties when trying to
defragment large fragmented files.


--


Gerry
~~~~
FCA
Stourport, England
Enquire, plan and execute
~~~~~~~~~~~~~~~~~~~
 
BobG said:
Defrag and MyDefrag both show pagefile.sys as a big hunk and a smaller
hunk. Anyway to combine em?
Click to expand...

Yes there is but a 2 segmented pf is completely negligible. You'll never
notice the difference in seek/access times.

You might be able to do it by:
Turn OFF the pf.
Defrag the disk. Repeat until nothing more will defrag.
Restart
Turn ON the pf
Set the pf to finite min/max settings. (not really recommended; system
managed is better)

Or there are 3rd party programs that will do it, too.

But it's all a waste of time; you'll never see anything improve.

HTH,

Twayne`
 
Gerry said:
Bennett

If the free disk space on the volume, where the pagefile is to be
placed, is less than 60% you can get problems achieving a single
contiguous file. The reason for this situation is that the system
places the pagefile in the centre of the drive. The figure of 40% -
60% is a guide but much lower figures like 25% make achieving a
single contiguous file difficult. If the free space available is
marginal you can try removing temporarily a large file(s) (like
system restore points) to temporarily create extra space. To retain a
contiguous file it is necessary to have a fixed minimum = maximum
setting.
Using the Windows Managed option on a system with less than 60% free
disk will ultimately mean that the pagefile becomes fragmented, which
of itself is not so important, but it causes difficulties when trying
to defragment large fragmented files.
Click to expand...

As you said, it's not that important to do, but there are defraggers
that will open up the disk space for you, let you put some things at the
beginning, end, outer track, leave a gap, etc. etc. etc..

It isn't correct that XP MUST write the pf where you said though. If
you defrag and there is NO space left in the defragged drive, the pf
will go where it can find the room and will be continguous.

IMO it's a waste of time anyway; only the purists mess with the likes of
it. Each to their own, I guess.

HTH,

Twayne`
 
Twayne said:
IMO it's a waste of time anyway; only the purists mess with the likes of
it. Each to their own, I guess.
Click to expand...

I wonder what they'll get up to when solid-state drives become the norm?
 
Twayne

"If you defrag and there is NO space left in the defragged drive, the pf
will go where it can find the room and will be continguous."

Misleading and incorrect statement.! The pagefile will be written where
there is space. Yes. However, your statement assumes the free space is
contiguous, which is usually not so.

--


Gerry
~~~~
FCA
Stourport, England
Enquire, plan and execute
~~~~~~~~~~~~~~~~~~~
 
Gerry said:
Misleading and incorrect statement.! The pagefile will be written where
there is space. Yes. However, your statement assumes the free space is
contiguous, which is usually not so.
Click to expand...

JKDefreg, which is free, will make your free space contiguous, within
the limits imposed by items that are immovable.

I find its display quite mesmerising, especially when I maximise it.
 
Swifty

I was writing in the context of the Microsoft Disk Defragmenter. It's
true you can get third party defragmenters to make free space contiguous
but that does not necessarily result in a contiguous pagefile. You still
have to get the system to write a pagefile in the contiguous free space
before it is populated by other files. This can be difficult on a disk
with limited free disk space. It is infinitely easier on a disk with
plenty of free space.

To create either a minimum = maximum pagefile or to place the pagefile
in a dedicated partition is a better solution if you want to avoid the
downsides of a fragmented. pagefile. A windows managed pagefile, used by
a majority of users, will always be more susceptible to fragmentation.
Taking the pagefile out of the defragmentation process reduces the time
defragmentation takes.

If you are interested in third party defragmenters try Defraggler
(freeware).
http://www.defraggler.com/features


--


Gerry
~~~~
FCA
Stourport, England
Enquire, plan and execute
~~~~~~~~~~~~~~~~~~~
 
Swifty said:
I wonder what they'll get up to when solid-state drives become the
norm?
Click to expand...

Dunno; but not too sure I care. I just want to see SSDs come into
common, affordable use, to be honest! I suppose alpha migration will
become a new "problem" then<g>. There will always be purists around and
that's OK with me; they make life interesting if nothing else. Nothing
wrong with wanting things perfect as long as they realize actual
perfection is so elusive as to be impossible.

Twayne`
 
Gerry said:
Twayne

"If you defrag and there is NO space left in the defragged drive, the
pf will go where it can find the room and will be continguous."
Click to expand...

lol! I said that? Chalk that one up to a bit of an aberration I guess.
Misleading and incorrect statement.! The pagefile will be written
where there is space. Yes. However, your statement assumes the free
space is contiguous, which is usually not so.
Click to expand...

True: That's why there's usually a preference to do defrags beforehand
to minimize that. But you're correct, and that's what I meant to say.
XP has the "smarts" to not fill orphaned sectors or parts of sectors
with the page file so actually it is possible for the fragmentation to
create an unfragged or at least negligibly so if the defrags have been
carried out. There are so many dependencies though that there's no way
to reasonably make strong statements on it.

Twayne`
 
Gerry said:
Swifty

I was writing in the context of the Microsoft Disk Defragmenter. It's
true you can get third party defragmenters to make free space
contiguous but that does not necessarily result in a contiguous
pagefile. You still have to get the system to write a pagefile in the
contiguous free space before it is populated by other files. This can
be difficult on a disk with limited free disk space. It is
infinitely easier on a disk with plenty of free space.
Click to expand...

Correct. But a pagefile with a few sections, which is often like 2 or 3
in this kind of a case, is going to be entirely negligible and often
will actually be continguous, as imagined on a display at least; obvious
not by the levels of the platters/tracks it resices on, of course. XP
doesn't treat the pf like any other file; if it can be contiguous, it
usually will be. Too many intevening immovable files of course will
split it if the contigous space isn't available in all one piece with, I
forget the %, room to spare.
To create either a minimum = maximum pagefile or to place the pagefile
in a dedicated partition is a better solution if you want to avoid the
downsides of a fragmented. pagefile. A windows managed pagefile, used
by a majority of users, will always be more susceptible to
fragmentation. Taking the pagefile out of the defragmentation process
reduces the time defragmentation takes.
Click to expand...

Literally true, but not fragged to any degree of noticeable performance
hits is a good defrag is performed first. What really creates issues on
a nearly full drive is things like the Recycle bin, REstore points,
Table corruption and a bunch of other programs that create immovable
files that make getting enough continguous space together pretty
difficult. They all seem to want 10% or 12% of the drive to themselves
and that's no small amount of space. It's easy for nearly half the
drive space to become "dedicated" for use whether it's visible on the
maps or not.
There is also a huge disconnect between the "picture" that defraggers
show you and the reality of where things may be placed on the platters
of any disk drive. Some will call it contiguous if the head can move to
the track and be ready to read the sector before a different platter
comes around less than x times, whatever they decide to make it. I've
seen one, forget the name of it, that refused to call it contiguous
unless the data really was contingous on one platter of the drive. That
alone made it nearly impossible to have a pf that wasn't in at least
two fragments. Interestingly enough, the Vertical Data Storage methods
that were so popular never seem to have panned out as being such an
advantage after all. You don't even hear it mentioned very often these
days. I think Vertical storage had a chance but it was just too
expensive to compete with the slim profit margins of current drives.
Interesting subjects, actually, IMO.

Let's make it all moot though and pray for SS drives in the Gigabyte
area in the near future. The small ones work; now let's find a way to
produce big ones everyone can use and afford! Sorry; couldn't resist -
I've played with a couple of them tossed out but still usable and it's
pretty neat, to say the least. They were only 50 Meg or so, but still;
that's enough for some people.

HTH,

Twayne`
 
Back
Top