Sign in with
Sign up | Sign in
Your question

Performance after your hard drive reaches 50% capacity

Last response: in Storage
Share
Anonymous
a b G Storage
August 31, 2004 12:50:13 PM

Archived from groups: comp.sys.ibm.pc.hardware.storage,microsoft.public.windowsxp.perform_maintain,comp.sys.laptops (More info?)

hi,

Last week i was reading the magazine "laptop" (it was an act of
desperation, stuck in an airport :-) ) and it had a section on tips to
improve performance. One of the tips it had was to keep your hard
drive clean; to me that made sense, but then it said something that
seemed kind of odd: "once you get over 50 percent capacity, hard
drives are much slower because they can't create a mirror image". Is
there any truth to this statement? I can understand once you have less
than 6gb (if you use a tweak) you start running out of room and the
pagefile can no longer reach maximum size, but if you have a 60 gb
hard drive do you really need 30gb free for max performance? And what
is this mirror image thing?
Anonymous
a b G Storage
August 31, 2004 6:56:16 PM

Archived from groups: comp.sys.ibm.pc.hardware.storage,microsoft.public.windowsxp.perform_maintain,comp.sys.laptops (More info?)

Dan Irwin wrote:

> hi,
>
> Last week i was reading the magazine "laptop" (it was an act of
> desperation, stuck in an airport :-) ) and it had a section on tips to
> improve performance. One of the tips it had was to keep your hard
> drive clean; to me that made sense, but then it said something that
> seemed kind of odd: "once you get over 50 percent capacity, hard
> drives are much slower because they can't create a mirror image". Is
> there any truth to this statement? I can understand once you have less
> than 6gb (if you use a tweak) you start running out of room and the
> pagefile can no longer reach maximum size, but if you have a 60 gb
> hard drive do you really need 30gb free for max performance? And what
> is this mirror image thing?

Find out what the guy was smoking, it might be pretty good stuff.

As the drive fills, performance declines simply because (a) the average seek
is longer and (b) the storage density and thus the data transfer rate is
highest on the outermost tracks, which are filled first.

--
--John
Reply to jclarke at ae tee tee global dot net
(was jclarke at eye bee em dot net)
August 31, 2004 8:20:59 PM

Archived from groups: comp.sys.ibm.pc.hardware.storage,microsoft.public.windowsxp.perform_maintain,comp.sys.laptops (More info?)

Dan Irwin wrote:

> hi,
>
> Last week i was reading the magazine "laptop" (it was an act of
> desperation, stuck in an airport :-) ) and it had a section on tips to
> improve performance. One of the tips it had was to keep your hard
> drive clean; to me that made sense, but then it said something that
> seemed kind of odd: "once you get over 50 percent capacity, hard
> drives are much slower because they can't create a mirror image". Is
> there any truth to this statement? I can understand once you have less
> than 6gb (if you use a tweak) you start running out of room and the
> pagefile can no longer reach maximum size, but if you have a 60 gb
> hard drive do you really need 30gb free for max performance? And what
> is this mirror image thing?

That sounds bogus to me.

--
The e-mail address in our reply-to line is reversed in an attempt to
minimize spam. Our true address is of the form che...@prodigy.net.
Related resources
Anonymous
a b G Storage
August 31, 2004 11:12:48 PM

Archived from groups: comp.sys.ibm.pc.hardware.storage,microsoft.public.windowsxp.perform_maintain,comp.sys.laptops (More info?)

And to me! I reckon you need 20%+ free space. You can run on less but you
will see problem like the ability to defrag ( 15% needed ). Are you sure you
read the Article right?

--

~~~~~~


Hope this helps.

Gerry
~~~~~~~~~~~~~~~~~~~~~~~~
FCA

Stourport, Worcs, England
Enquire, plan and execute.
~~~~~~~~~~~~~~~~~~~~~~~~
Please tell the newsgroup how any
suggested solution worked for you.

http://dts-l.org/goodpost.htm

~~~~~~~~~~~~~~~~~~~~~~~~


"CJT" <abujlehc@prodigy.net> wrote in message
news:4134A56A.20204@prodigy.net...
> Dan Irwin wrote:
>
>> hi,
>>
>> Last week i was reading the magazine "laptop" (it was an act of
>> desperation, stuck in an airport :-) ) and it had a section on tips to
>> improve performance. One of the tips it had was to keep your hard
>> drive clean; to me that made sense, but then it said something that
>> seemed kind of odd: "once you get over 50 percent capacity, hard
>> drives are much slower because they can't create a mirror image". Is
>> there any truth to this statement? I can understand once you have less
>> than 6gb (if you use a tweak) you start running out of room and the
>> pagefile can no longer reach maximum size, but if you have a 60 gb
>> hard drive do you really need 30gb free for max performance? And what
>> is this mirror image thing?
>
> That sounds bogus to me.
>
> --
> The e-mail address in our reply-to line is reversed in an attempt to
> minimize spam. Our true address is of the form che...@prodigy.net.
Anonymous
a b G Storage
September 1, 2004 12:02:08 AM

Archived from groups: comp.sys.ibm.pc.hardware.storage,microsoft.public.windowsxp.perform_maintain,comp.sys.laptops (More info?)

when i get back home i will scan in the page

"J. Clarke" <jclarke@nospam.invalid> wrote in message news:<ch2hs101v9j@news4.newsguy.com>...
> Dan Irwin wrote:
>
> > hi,
> >
> > Last week i was reading the magazine "laptop" (it was an act of
> > desperation, stuck in an airport :-) ) and it had a section on tips to
> > improve performance. One of the tips it had was to keep your hard
> > drive clean; to me that made sense, but then it said something that
> > seemed kind of odd: "once you get over 50 percent capacity, hard
> > drives are much slower because they can't create a mirror image". Is
> > there any truth to this statement? I can understand once you have less
> > than 6gb (if you use a tweak) you start running out of room and the
> > pagefile can no longer reach maximum size, but if you have a 60 gb
> > hard drive do you really need 30gb free for max performance? And what
> > is this mirror image thing?
>
> Find out what the guy was smoking, it might be pretty good stuff.
>
> As the drive fills, performance declines simply because (a) the average seek
> is longer and (b) the storage density and thus the data transfer rate is
> highest on the outermost tracks, which are filled first.
Anonymous
a b G Storage
October 1, 2004 2:09:36 AM

Archived from groups: comp.sys.ibm.pc.hardware.storage,microsoft.public.windowsxp.perform_maintain,comp.sys.laptops (More info?)

"Dan Irwin" <harryguy082589@aol.com> wrote in message
news:2a779348.0408310750.418fd6d4@posting.google.com...
> hi,
>
> Last week i was reading the magazine "laptop" (it was an act of
> desperation, stuck in an airport :-) ) and it had a section on tips to
> improve performance. One of the tips it had was to keep your hard
> drive clean; to me that made sense, but then it said something that
> seemed kind of odd: "once you get over 50 percent capacity, hard
> drives are much slower because they can't create a mirror image". Is
> there any truth to this statement?

No truth to that.

> I can understand once you have less
> than 6gb (if you use a tweak) you start running out of room and the
> pagefile can no longer reach maximum size, but if you have a 60 gb
> hard drive do you really need 30gb free for max performance? And what
> is this mirror image thing?

Something from lala land.
Anonymous
a b G Storage
October 1, 2004 1:53:48 PM

Archived from groups: comp.sys.ibm.pc.hardware.storage,microsoft.public.windowsxp.perform_maintain,comp.sys.laptops (More info?)

"Ron Reaugh" <rondashreaugh@att.net> wrote in message
news:Au%6d.462308$OB3.71488@bgtnsc05-news.ops.worldnet.att.net...
>
> "Dan Irwin" <harryguy082589@aol.com> wrote in message
> news:2a779348.0408310750.418fd6d4@posting.google.com...
>> hi,
>>
>> Last week i was reading the magazine "laptop" (it was an act of
>> desperation, stuck in an airport :-) ) and it had a section on tips to
>> improve performance. One of the tips it had was to keep your hard
>> drive clean; to me that made sense, but then it said something that
>> seemed kind of odd: "once you get over 50 percent capacity, hard
>> drives are much slower because they can't create a mirror image". Is
>> there any truth to this statement?
>
> No truth to that.

Actually, it is kind of true.

Modern hard disks have an increasing range of disk transfer rates from the
inner diameter to the outer diameter of the disk. This is called a "zoned"
recording technique. Since hard disks are filled from the outside in, the
fastest data transfer occurs when the drive is first used (empty).

This is why when the computer is new and the HDD almost empty, a benchmark
will show higher performace than when the HDD is almost full

i do not know what is meant with the mirror image though.
--

Fleur de Coin
- http://www.fleur-de-coin.com/
Åëëçíéêüò óôñáôüò
- http://skopia.digitalrice.com/
Anonymous
a b G Storage
October 2, 2004 4:27:34 PM

Archived from groups: comp.sys.ibm.pc.hardware.storage,microsoft.public.windowsxp.perform_maintain,comp.sys.laptops (More info?)

In article <cjiuv2$452$1@newsmaster.public.dc.hol.net>,
vikhr@hotmail.com says...
> Actually, it is kind of true.
>
> Modern hard disks have an increasing range of disk transfer rates from the
> inner diameter to the outer diameter of the disk. This is called a "zoned"
> recording technique. Since hard disks are filled from the outside in, the
> fastest data transfer occurs when the drive is first used (empty).

The transfer rate for the inner-tracks is usually only
25% (or thereabouts) higher then the outer-track speeds.
Sometimes not even that much of a difference (e.g.
25MB/s vs 30MB/s)

> This is why when the computer is new and the HDD almost empty, a benchmark
> will show higher performace than when the HDD is almost full
>
> i do not know what is meant with the mirror image though.

What can happen is that NTFS (assuming that's what the
OP is using) starts to degrade in performance once the
drive hits 90-95% full. It will start to fragment
faster and other general slow-downs (dunno exact
reasons).

I try to keep my drives at the 15-25% empty mark (as
defrag won't run if you have less then 15% free).
Anonymous
a b G Storage
October 2, 2004 4:58:45 PM

Archived from groups: comp.sys.ibm.pc.hardware.storage,microsoft.public.windowsxp.perform_maintain,comp.sys.laptops (More info?)

Toshi1873 wrote:

> In article <cjiuv2$452$1@newsmaster.public.dc.hol.net>,
> vikhr@hotmail.com says...
>> Actually, it is kind of true.
>>
>> Modern hard disks have an increasing range of disk transfer rates from
>> the inner diameter to the outer diameter of the disk. This is called a
>> "zoned" recording technique. Since hard disks are filled from the outside
>> in, the fastest data transfer occurs when the drive is first used
>> (empty).
>
> The transfer rate for the inner-tracks is usually only
> 25% (or thereabouts) higher then the outer-track speeds.
> Sometimes not even that much of a difference (e.g.
> 25MB/s vs 30MB/s)

According to Hitachi the outermost zone on the 7K250 has a STR of 61.4
MB/sec vs 29.7 on the innermost.

>> This is why when the computer is new and the HDD almost empty, a
>> benchmark will show higher performace than when the HDD is almost full
>>
>> i do not know what is meant with the mirror image though.
>
> What can happen is that NTFS (assuming that's what the
> OP is using) starts to degrade in performance once the
> drive hits 90-95% full. It will start to fragment
> faster and other general slow-downs (dunno exact
> reasons).
>
> I try to keep my drives at the 15-25% empty mark (as
> defrag won't run if you have less then 15% free).

--
--John
Reply to jclarke at ae tee tee global dot net
(was jclarke at eye bee em dot net)
Anonymous
a b G Storage
October 4, 2004 4:42:58 AM

Archived from groups: comp.sys.ibm.pc.hardware.storage,microsoft.public.windowsxp.perform_maintain,comp.sys.laptops (More info?)

"Toshi1873" <toshi1873@nowhere.com> wrote in message news:MPG.1bc8a106574114eb9899b3@news.giganews.com
> In article <cjiuv2$452$1@newsmaster.public.dc.hol.net>, vikhr@hotmail.com says...
> > Actually, it is kind of true.
> >
> > Modern hard disks have an increasing range of disk transfer rates from the
> > inner diameter to the outer diameter of the disk. This is called a "zoned"
> > recording technique. Since hard disks are filled from the outside in, the
> > fastest data transfer occurs when the drive is first used (empty).
>
> The transfer rate for the inner-tracks is usually only
> 25% (or thereabouts) higher then the outer-track speeds.

Utterly Clueless.

> Sometimes not even that much of a difference (e.g. 25MB/s vs 30MB/s)

That would be an ATA66 drive run on an ATA33 channel or used with
a 40-wire cable.

Most 7200 rpm drives manage a 1:1.75 (75%) difference between inner
and outer STR. Even 15krpm drives manage a 1:1.5 (50%) difference.

The only drive that I could find to deviate was a Barracuda 36ES2
that had a 1:1.3 (30%) difference. I think this drive was suspected to
be an IDE drive on a SCSI interface and it probably got it's low
(lower than IDE) average access time by shortstroking the capacity
of an originally bigger drive to say half or even less.

>
> > This is why when the computer is new and the HDD almost empty, a
> > benchmark will show higher performace than when the HDD is almost full
> >
> > i do not know what is meant with the mirror image though.
>
> What can happen is that NTFS (assuming that's what the
> OP is using) starts to degrade in performance once the
> drive hits 90-95% full. It will start to fragment
> faster and other general slow-downs (dunno exact reasons).
>
> I try to keep my drives at the 15-25% empty mark (as
> defrag won't run if you have less then 15% free).
Anonymous
a b G Storage
October 4, 2004 7:13:47 PM

Archived from groups: comp.sys.ibm.pc.hardware.storage,microsoft.public.windowsxp.perform_maintain,comp.sys.laptops (More info?)

On 2004-10-02 18:27, Toshi1873 wrote:
-snip-

> What can happen is that NTFS (assuming that's what the
> OP is using) starts to degrade in performance once the
> drive hits 90-95% full. It will start to fragment
> faster and other general slow-downs (dunno exact
> reasons).
>
> I try to keep my drives at the 15-25% empty mark (as
> defrag won't run if you have less then 15% free).

Well, I filled my disk above 85% and got a complaint popup
from defrag, but it did still defragment the drive ok.
(I had kept it well defragmented prior to reaching the limit.)

I figure if the fragmented files fit in the free
space, they can still be moved around.

If you had an imaginary disk of immesureable storage capacity, you'd
still need 15% of it free, to defrag?


/Rolf



.... . . . . . . . . . .
Universe too fragmented.
Please defrag before installing new galaxies.
defrag y/n?
>y
password> gibgnab
done
..
Anonymous
a b G Storage
October 6, 2004 5:02:32 PM

Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

In article <g0c8d.86$Q%1.158228@uab.ericsson.se>,
allspam@round.bin says...
> Well, I filled my disk above 85% and got a complaint popup
> from defrag, but it did still defragment the drive ok.
> (I had kept it well defragmented prior to reaching the limit.)

True, I forgot that you can override it.

> I figure if the fragmented files fit in the free
> space, they can still be moved around.
>
> If you had an imaginary disk of immesureable storage capacity, you'd
> still need 15% of it free, to defrag?

Blame Microsoft! Figure they wrote the defrag tool back
when a 8GB disk was big. The defrag tool is hard coded
to complain at 15%.

(Then there's the issue that the max Recycle Bin size is
still 4GB, even though drive sizes are 10x what they
used to be.)

I have had issues with NTFS getting slow once you hit
200-300k files on the drive. Could've been something
else wrong too.
!