80% maximum capacity on a hard drive?

  • Thread starter Thread starter James
  • Start date Start date
J

James

Some years back, in a networking class. I was told that the maximum
percentage that a hard drive or a volume could be filled is 80 percent,
because otherwise you would end up getting excessive wear on the hard drive.
Does that limit still apply? Or did that only apply to the drive containing
the system volume? What about a storage hard drive soley used for storing
data?

Thanks.
 
James said:
Some years back, in a networking class. I was told that the maximum percentage that a
hard drive or a volume could be filled is 80 percent, because otherwise you would end up
getting excessive wear on the hard drive.
Click to expand...

That was always complete pig ignorant drivel.
Does that limit still apply?
Click to expand...

It never did and so obviously doesnt now either.
Or did that only apply to the drive containing the system volume?
Click to expand...

Nope, it never did and so obviously doesnt now either.
What about a storage hard drive soley used for storing data?
Click to expand...

Nope, it never did and so obviously doesnt now either.
 
James said:
Some years back, in a networking class. I was told that the maximum
percentage that a hard drive or a volume could be filled is 80 percent,
because otherwise you would end up getting excessive wear on the hard drive.
Does that limit still apply? Or did that only apply to the drive containing
the system volume? What about a storage hard drive soley used for storing
data?

Thanks.
Click to expand...

Total BS and always was. I have filled every sector on FAT16 and FAT32
HDs under W9x and under XP, and NTFS HDs under XP, with no surprises.
Clearly, defrag won't run; but all files remained readable and nothing
crashed unexpectedly.
 
McSpreader said:
Never heard that one. Whatever was the reasoning for it?
Click to expand...

This was thrown at me by a teacher of an NT class as one of a zillion
factoids that we had to learn. At the time nobody in the class questioned
that.
 
James said:
This was thrown at me by a teacher of an NT class as one of a zillion
factoids that we had to learn. At the time nobody in the class questioned
that.
Click to expand...


It was probably based upon the effect of a fragmented drive.
 
This was thrown at me by a teacher of an NT class as one of a
zillion factoids that we had to learn. At the time nobody in the
class questioned that.
Click to expand...

IMO the factoid is OK-ish, the reasonoid is BS.

It makes good sense to allow headroom in the space utilisation of
a drive to cope with transient peaks of demand. This is
particularly true of the system drive of a heavily used system e.g.
a server. The reasoning is that it's wise to avoid exhausting all
free space, which could be expected as a minimum to impact system
performance - more likely to crash apps, services, the o/s, and
risk data integrity. Housekeeping activities, e.g. backups and
defragging, usually require a fair amount of free space in which to
work efficiently.
 
James said:
Some years back, in a networking class. I was told that the maximum
percentage that a hard drive or a volume could be filled is 80
percent, because otherwise you would end up getting excessive wear on
the hard drive.
Click to expand...

Utter drivel.
Or did that only apply to the drive containing the system volume?
Click to expand...

The only grain of truth is that you need a certain amount of free space on
a drive to defragment it.

Possibly that myth started by telling poeple to always keep some space
free on their C: drive since windows won't start up or will crash if it has
too small a swap file, or no space for temporary files when printing etc.

As far as excessive wear is concerned, all you have to worry about is
heat and physical shock.

I remember once being told heavy usage would wear a track out
on the hard disk ! I think these people think of hdds like a vinyl record ;)
 
McSpreader said:
IMO the factoid is OK-ish, the reasonoid is BS.

It makes good sense to allow headroom in the space utilisation of
a drive to cope with transient peaks of demand. This is particularly
true of the system drive of a heavily used system e.g. a server.
Click to expand...
The reasoning is that it's wise to avoid exhausting all free space,
which could be expected as a minimum to impact system performance
- more likely to crash apps, services, the o/s, and risk data integrity.
Housekeeping activities, e.g. backups and defragging, usually require
a fair amount of free space in which to work efficiently.
Click to expand...

And this has what exactly to do with "a heavily used system e.g. a server"?

And why would 2 GB free space be enough for a 10GB drive but not for a
100GB drive?
 
Mike Redrobe said:
Utter drivel.


The only grain of truth is that you need a certain amount of free space on
a drive to defragment it.
Click to expand...

But it's not 20%.
Possibly that myth started by telling poeple to always keep some space
free on their C: drive since windows won't start up or will crash if it has
too small a swap file, or no space for temporary files when printing etc.

As far as excessive wear is concerned, all you have to worry about is

and physical shock.
Click to expand...
I remember once being told heavy usage would wear a track out
on the hard disk !
Click to expand...
I think these people think of hdds like a vinyl record ;)
Click to expand...

The comparison isn't that far off.

Both readers produce heat in the media.
If nothing is done about it, continuous use of a track can overheat that
track. Harddisk drives will continously move the heads to avoid that a
head stays over the same track too long. Depending on how smart the OS
software or disk firmware is it is conceivable that application software
can keep the heads at the same track for far too long and overheat it.
 
Previously James said:
Some years back, in a networking class. I was told that the maximum
percentage that a hard drive or a volume could be filled is 80 percent,
because otherwise you would end up getting excessive wear on the hard drive.
Does that limit still apply? Or did that only apply to the drive containing
the system volume? What about a storage hard drive soley used for storing
data?
Click to expand...

Depends on your access pattern and filesystem. However today HDD head
movements are not a primary wear concern. Depending on filesystem,
access pattern and file sizes you may experience slow access
(defragmenting helps) and increased drive heat production. If the
drive is cooled as per its specifications, this is not a problem.

From personal experience, large numbers of small files may be
problematic, regardless of filesystem and drive fill degree
on FAT filesystems. I have had no issues with ext2/ext3 even
when filled 100%. I have no idea how NTFS does.

Arno
 
Both readers produce heat in the media.
If nothing is done about it, continuous use of a track can overheat that
track. Harddisk drives will continously move the heads to avoid that a
head stays over the same track too long. Depending on how smart the OS
software or disk firmware is it is conceivable that application software
can keep the heads at the same track for far too long and overheat it.
Click to expand...


Folkert,

I would *love* to quote your diatribe on my new website-to-be.

May I? I will, of course, provide microscopic photographic evidence
detailing this phenomenon of air-borne heads overheating the platters to
the extent that damage may occur. Infra-red photography and possibly
micro-photon photography will naturally pick up all these nuances your
remit details. And, naturally, I will quote you as the source of such
genius. (May I quote the company of which you are an employee? I'm
sure it will enhance their reputation as a leader in their field.)


Odie (Sorry - *Duncan*)
 
Fascinating! I guess we will have to move to liquid-cooled heads
to avoid this. Also I am quite worried that HDDs misaligned to
galactic rotation will also suffer from track overheating as a
consequence. That effect may be as large as the one you describe!

Arno

Folkert,
Click to expand...
 
Arno said:
Fascinating! I guess we will have to move to liquid-cooled heads
to avoid this. Also I am quite worried that HDDs misaligned to
galactic rotation will also suffer from track overheating as a
consequence. That effect may be as large as the one you describe!
Click to expand...


Arno, don't be silly. It will have to be liquid-cooled platters.

I would have thought you'd have known better!!!!

Perhaps Folkrent would like to see platters and heads totally submerged
in some exotic cooling liquid.

Don't suppose it's entirely impossible, though...


Odie
 
Odie said:
Arno, don't be silly. It will have to be liquid-cooled platters.

I would have thought you'd have known better!!!!

Perhaps Folkrent would like to see platters and heads totally submerged
in some exotic cooling liquid.

Don't suppose it's entirely impossible, though...


Odie
Click to expand...

Uh, it's been done. In the late 60's and into the 70's we used HDs which
were sealed, with the heads and disks in helium; pretty good at removing
the heat, but rather hard to stop the leaks.
 
Bob Willard said:
Uh, it's been done. In the late 60's and into the 70's we used HDs which
were sealed, with the heads and disks in helium; pretty good at removing
the heat, but rather hard to stop the leaks.
Click to expand...

Bob, would you please not piss on smelly's and babblehead's party.
It takes all the fun out of it. Thank you.
 
Folkert said:
Bob, would you please not piss on smelly's and babblehead's party.
It takes all the fun out of it. Thank you.
Click to expand...

Coming from a man who's the life and soul of any party!

Hi, Folkert. Nice break?


Duncan
 
Odie said:
Coming from a man who's the life and soul of any party!
Click to expand...

I don't know... Maybe after a couple beers he stops scrutinizing
everyone else for imperfections... 8)
 
chrisv said:
I don't know... Maybe after a couple beers he stops scrutinizing
everyone else for imperfections... 8)
Click to expand...

Funny you say that when it was Bob who actually made them look poor.
(I'll guess that Bob just joined babblehead's crowded killfile now too)

I left them both in their obvious delusions that they think they spotted
a wrong in what I wrote.

Obviously babblehead is so utterly stupid that when smellie the troll
" I think I'm a harddrive recovery specialist, therefor I am"
just 'thinks' I was talking utter BS he then could safely join the party.

Isn't it always nice to see people who obviously can't
control themselves make a fool out of themselves.
 
Back
Top