Move pagefile to faster drive ?

C

chocolatemint77581

I recently put a second drive on my computer.

My primary drive is 18.6 GB and the 2nd drive is 160 Gb.

Would it be beneficial to move my fixed pagefile to the new drive?
(I am assuming that the 2nd drive is probably a little faster.)


Thanks.
 
H

HeyBub

I recently put a second drive on my computer.

My primary drive is 18.6 GB and the 2nd drive is 160 Gb.

Would it be beneficial to move my fixed pagefile to the new drive?
(I am assuming that the 2nd drive is probably a little faster.)
If your computer USES the pagefile much, sure.
 
D

db

my suggestion is to try to reverse the scenerio.

its better to have your c drive relocated to the
160 gig and then use your 18 gig as a slave
and for the virtual memory if needed.


--
db·´¯`·...¸><)))º>
DatabaseBen, Retired Professional
- Systems Analyst
- Database Developer
- Accountancy
- Veteran of the Armed Forces
- @Hotmail.com

"share the nirvana mann" - dbZen

~~~~~~~~~~~~~~~
 
T

Twayne

I recently put a second drive on my computer.

My primary drive is 18.6 GB and the 2nd drive is 160 Gb.

Would it be beneficial to move my fixed pagefile to the new drive?
(I am assuming that the 2nd drive is probably a little faster.)


Thanks.

See db's comments about the drive usage: he has a good point. Also:

A drive won't be "faster" because it's new. It might seem that way at
first but as it begins to acquire files, become fragmented, etc. etc.,
it will be exactly the same as your old drive unless you KNOW it is
faster.
Drives that are affordable come in two spindle speeds: 5400 or 7200
rpm. If you have one of each, then one will be faster th an the other.
If they are the same, then there will be no difference in speed. Only
the OS and drive management/maintenance will make any difference;
,meaning you, the user.

Unless you have a specific, known reason for it, it's better to not use
fixed size page file too. Let the system manage the size. It's not
very often that there is a need for fixed sizes of the pf and your
apparent level of expertise here makes me suspect that's not the case
you're in. With a fixed size, you're either going to encounter out of
space problems with a fixed pf size, or you'll be wasting a lot of space
100% of the time for no good reason.
It often turns out people use a fixed size pg because they've heard
it minimizes the fragmentation of the pf. As long as the drive has
sufficient room on it and was defragged before setting up the page file,
it will not become excessively fragmented. And if you're low on disk
space, because of a fixed pf size, you'll likely start to have other
annoyances bugging you too. XP tries to be sure there will be minimum
fragmentation to the pf. And if it gets too large and does fragment,
that fragmentation goes away when it drops down to normal size again.
For best performance, a drive should be less than 80% occupied. That
counts the space for pf, restore points, etc. etc..

HTH,

Twayne`
 
B

BillW50

In Twayne typed on Sat, 22 Aug 2009 14:09:37 -0400:
See db's comments about the drive usage: he has a good point. Also:

A drive won't be "faster" because it's new. It might seem that way at
first but as it begins to acquire files, become fragmented, etc. etc.,
it will be exactly the same as your old drive unless you KNOW it is
faster.
Drives that are affordable come in two spindle speeds: 5400 or 7200
rpm. If you have one of each, then one will be faster th an the
other. If they are the same, then there will be no difference in
speed. Only the OS and drive management/maintenance will make any
difference; ,meaning you, the user.

Unless you have a specific, known reason for it, it's better to not
use fixed size page file too. Let the system manage the size. It's
not very often that there is a need for fixed sizes of the pf and your
apparent level of expertise here makes me suspect that's not the case
you're in. With a fixed size, you're either going to encounter out of
space problems with a fixed pf size, or you'll be wasting a lot of
space 100% of the time for no good reason.
It often turns out people use a fixed size pg because they've heard
it minimizes the fragmentation of the pf. As long as the drive has
sufficient room on it and was defragged before setting up the page
file, it will not become excessively fragmented. And if you're low
on disk space, because of a fixed pf size, you'll likely start to
have other annoyances bugging you too. XP tries to be sure there will
be minimum fragmentation to the pf. And if it gets too large and
does fragment, that fragmentation goes away when it drops down to
normal size again. For best performance, a drive should be less than
80% occupied. That counts the space for pf, restore points, etc.
etc..
HTH,

Twayne`

Oh nonsense! The correct size of your pagefile solely depends on how
much free RAM you have (if you always have 200MB or more free, you don't
need a pagefile). And on a XP machine with 1GB or more, few will need a
pagefile at all. And there are many reasons why not to run a pagefile or
a fixed one as well. One very good reason is if you use a SSD instead of
a HDD. As excessive and unnecessary writing to a SSD drive just isn't
very smart. As the longevity goes downhill from such a practice.

People who say it cannot be done should not interrupt the people who are
doing it. -- Anonymous
 
H

HeyBub

BillW50 said:
Oh nonsense! The correct size of your pagefile solely depends on how
much free RAM you have (if you always have 200MB or more free, you
don't need a pagefile).

Uh, no. The size of the pagefile depends on how many programs (and their
sizes) you have open at one time. 100 separate sessions of Internet Explorer
will burn up a significant amount of pagefile diskspace, even with 4GB of
RAM.
And on a XP machine with 1GB or more, few
will need a pagefile at all. And there are many reasons why not to
run a pagefile or a fixed one as well. One very good reason is if you
use a SSD instead of a HDD. As excessive and unnecessary writing to a
SSD drive just isn't very smart. As the longevity goes downhill from
such a practice.

Depends on the drive. A RAM-drive is a Solid State Drive. You can write to
it several times before it breaks.
 
D

db

respectfully, most third party
programs are encoded to
utilize paging, regardless
of how much memory
one has.

in otherwords, there are
many third party programs
that will not function solely
in core memory.

--
db·´¯`·...¸><)))º>
DatabaseBen, Retired Professional
- Systems Analyst
- Database Developer
- Accountancy
- Veteran of the Armed Forces
- @Hotmail.com

"share the nirvana mann" - dbZen

~~~~~~~~~~~~~~~
 
B

BillW50

In HeyBub typed on Sat, 22 Aug 2009 17:33:23 -0500:
Uh, no. The size of the pagefile depends on how many programs (and
their sizes) you have open at one time. 100 separate sessions of
Internet Explorer will burn up a significant amount of pagefile
diskspace, even with 4GB of RAM.

That is what I said. If you always have 200MB free with no pagefile, you
don't need a pagefile. Btw, FF is far worse for using more memory. It
always has been, despite claims of fixing all of the memory leaks.
Depends on the drive. A RAM-drive is a Solid State Drive. You can
write to it several times before it breaks.

I also use a RAMDrive and RAMDrives are one of the most reliable drives
I have ever used. That is until they lose power. ;-)
 
B

BillW50

In db typed on Sat, 22 Aug 2009 17:36:32 -0500:
respectfully, most third party
programs are encoded to
utilize paging, regardless
of how much memory
one has.

in otherwords, there are
many third party programs
that will not function solely
in core memory.

I haven't found one yet. Name one and we will talk.
 
B

Barry Schwarz

I recently put a second drive on my computer.

My primary drive is 18.6 GB and the 2nd drive is 160 Gb.

Would it be beneficial to move my fixed pagefile to the new drive?
(I am assuming that the 2nd drive is probably a little faster.)

Having the pagefile on a different drive than Windows and you
applications will usually be faster because the drive heads, on
average, have to do less seeking.

The performance of a particular drive is usually affected to a greater
degree by seek speed and rotation rate rather than capacity.
 
D

db

photo shop.

but don't feel like talking.

--
db·´¯`·...¸><)))º>
DatabaseBen, Retired Professional
- Systems Analyst
- Database Developer
- Accountancy
- Veteran of the Armed Forces
- @Hotmail.com

"share the nirvana mann" - dbZen

~~~~~~~~~~~~~~~
 
D

dennis

db said:
respectfully, most third party
programs are encoded to
utilize paging, regardless
of how much memory
one has.

in otherwords, there are
many third party programs
that will not function solely
in core memory.

That seems to be a commonly misunderstanding of how memory access works.
 
M

Mint

See db's comments about the drive usage: he has a good point.  Also:

A drive won't be "faster" because it's new.  It might seem that way at
first but as it begins to acquire files, become fragmented, etc. etc.,
it will be exactly the same as your old drive unless you KNOW it is
faster.
   Drives that are affordable come in two spindle speeds:  5400 or 7200
rpm.  If you have one of each, then one will be faster th an the other.
If they are the same, then there will be no difference in speed.  Only
the OS and drive management/maintenance will make any difference;
,meaning you, the user.

Unless you have a specific, known reason for it, it's better to not use
fixed size page file too.  Let the system manage the size.  It's not
very often that there is a need for fixed sizes of the pf and your
apparent level of expertise here makes me suspect that's not the case
you're in.  With a fixed size, you're either going to encounter out of
space problems with a fixed pf size, or you'll be wasting a lot of space
100% of the time for no good reason.
   It often turns out people use a fixed size pg because they've heard
it minimizes the fragmentation of the pf.  As long as the drive has
sufficient room on it and was defragged before setting up the page file,
it will not become excessively fragmented.  And if you're low on disk
space, because of a fixed pf size, you'll likely start to have other
annoyances bugging you too. XP tries to be sure there will be minimum
fragmentation to the pf.  And if it gets too large and does fragment,
that fragmentation goes away when it drops down to normal size again.
  For best performance, a drive should be less than 80% occupied.  That
counts the space for pf, restore points, etc. etc..

HTH,

Twayne`

In my case, the older drive is probably at least 5 years older than
the new drive.

So it's gonna be faster.

How many files are on it is a matter of housekeeping.

If a fixed pagefile size is based on past page file usage, it is a
good choice.
And since it is fixed, it can be placed in a better position by
defragmenters.

Take care,
Andy
 
M

Mint

Having the pagefile on a different drive than Windows and you
applications will usually be faster because the drive heads, on
average, have to do less seeking.

The performance of a particular drive is usually affected to a greater
degree by seek speed and rotation rate rather than capacity.

After putting my pagefile on a another drive, I noticed an increase in
performance.

I put a 200 Mb pagefile on my primary drive( in case a log file gets
big) and a 200 Mb on my secondary drive.

Andy
 
D

db

windows paging and virtual mem
if I recall are distinctly different.

if you are confident about your
assumption then turn off you
virtual memory.

also, photo shop requires a
scratch disk, which is its
own distinct requirement.


--
db·´¯`·...¸><)))º>
DatabaseBen, Retired Professional
- Systems Analyst
- Database Developer
- Accountancy
- Veteran of the Armed Forces
- @Hotmail.com

"share the nirvana mann" - dbZen

~~~~~~~~~~~~~~~
 
D

dennis

db said:
windows paging and virtual mem
if I recall are distinctly different.

Applications only deal with virtual addressing. They don't care about
where data is actually stored - in physical memory or the paging file.
That is all internally to Windows. You cannot explicit code your
application to write to the paging file.
also, photo shop requires a
scratch disk, which is its
own distinct requirement.

The scratch disk is just some internal data storage to photoshop, and is
not related to the paging file.
 
D

db

perhaps, you should argue your
professional opinion with the
software engineers:

http://support.microsoft.com/kb/315270



--
db·´¯`·...¸><)))º>
DatabaseBen, Retired Professional
- Systems Analyst
- Database Developer
- Accountancy
- Veteran of the Armed Forces
- @Hotmail.com

"share the nirvana mann" - dbZen

~~~~~~~~~~~~~~~
 
T

Twayne

BillW50 said:
In Twayne typed on Sat, 22 Aug 2009 14:09:37 -0400:

Oh nonsense! The correct size of your pagefile solely depends on how
much free RAM you have (if you always have 200MB or more free, you
don't need a pagefile).

Nonsense. 1, you cannot run XP without a pagefile, even if you turn it
off. There will still be one created for you. 2. 200 M or more free
doesn't help much when it's a Gig of data you have to push out there; in
which case windows will still save you by extending the pf farther, but
will annoy you with messages about doing so. 3. You obviously don't do
any serious RAM-instensive work periodically. Windows will just "fix
it" for you rather than crash, which would be the alternative.
The pf not being used most of the time is great! And should always be
the target. Thus, the pf will occupy only a Meg or so of disck space -
unless you're dumb enough to give it a set size and waste that space
100% of th e time.

And on a XP machine with 1GB or more, few
will need a pagefile at all.

See above. Same response. I have 1.5 Gig, and DO need a pagefile for
when I'm working with images and video. Especially with video, the pg
can become very large, filled with buffer after buffer of still needed
but not now data.

And there are many reasons why not to
run a pagefile or a fixed one as well. One very good reason is if you
use a SSD instead of a HDD. As excessive and unnecessary writing to a
SSD drive just isn't very smart. As the longevity goes downhill from
such a practice.

You're talking about a new fledgling industry of which you obviously
know very little or you wouldn't have picked that as the major "issue"
at this point in time. As a past designer with SSD design experince,
the only real longevity problem is alpha migration - and eventually SSD
drives will achieve that too. You're not talking about PROMs with their
finite number of write cycles here; it's an entirely different
tachnology.

So you've been attempting to redirect this post to other purposes you
can troll better with, but it doesn't change the facts. You're simply
working to overcome your own feeling of being powerless here, IMO.


Twayne
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top