Sign in with
Sign up | Sign in
Your question

Read Speed Faster on Edge or Middle of Platter

Last response: in Storage
Share
a b G Storage
August 23, 2009 6:58:33 PM

I have researched this and many websites say that a Hard Drive's read speed is faster on the edge of the platter rather then in the middle and others say vice versa. Well Can anyone confirm this information becuase it seems that data in the center of the hard drive platter will be read faster since the spins required is shorter then on the edge. However many websites say data is read much faster if its located on the very edge of the platter. Please tell me which one is true.
-Thank you for any help :) 
a b G Storage
August 23, 2009 7:47:22 PM

Edge. Modern hard drives do not have the same number of sectors on each track. Because the outer tracks are longer than inner tracks, they have more sectors. That means an outer track contains more data than an inner track.
m
0
l
a b G Storage
August 23, 2009 8:39:36 PM

Oh! I see, thank you for the explanation.
m
0
l
a c 353 G Storage
August 23, 2009 10:22:35 PM

jsc is correct. Also the angular velocity is much higher at the outer edge.
(Values rounded off). If you have a 3" platter the outer edge will make a complete revoluction in 1/7600 Sec or about 9 inches per 0.132 mSec (14.6 inches per 14. uSec).
The center (1.5 inches) thakes the same time. but only moves about 4.7 inches per 0.132 mSec. this would equate to 28 microSec per inch.
m
0
l
August 24, 2009 12:56:09 AM

do a little research on short stroking your hard drive...it can double the speed of your HDD.
m
0
l
a b G Storage
August 24, 2009 1:11:48 AM

RetiredChief said:
jsc is correct. Also the angular velocity is much higher at the outer edge.
(Values rounded off). If you have a 3" platter the outer edge will make a complete revoluction in 1/7600 Sec or about 9 inches per 0.132 mSec (14.6 inches per 14. uSec).
The center (1.5 inches) thakes the same time. but only moves about 4.7 inches per 0.132 mSec. this would equate to 28 microSec per inch.

The linear velocity is higher at the outside than the middle, but the angular velocity (7200rpm in most cases) is the same. The disk will make a complete rotation every 8.33 milliseconds (not 1/7200 second, since it is 7200 rotations per minute). At the outer edge, this corresponds to 14.37 meters per second, while at the inner edge, this is about 7.18 meters per second. Since the linear data density is similar at both locations, the data rate will be quite a bit higher at the outer edge.
m
0
l
a c 353 G Storage
August 24, 2009 1:56:10 AM

cji - You are right - RPMs is minutes not sec and yes should have use linear instead of angular. But my principle is correct in that velocity is grater at the edge and bit density is the same on edge as in middle.
m
0
l
a b G Storage
August 24, 2009 6:39:37 AM

True.
m
0
l
a b G Storage
August 24, 2009 3:02:31 PM

Ah, thanks for the help guys, I actually found a software that I tried and it quite literally shows you a drive map of all your sectors and moves all your data and software to the edge of your hard drive, its cool but it pushes it out onto the edge. It calls this Consolidation :) 

-Anyway I will i try it out, but thanks for all the help.
m
0
l
a b G Storage
August 24, 2009 3:10:46 PM

Wait, i just looked up short stroking, the website gave me a good understanding then its basically moving all your software on the edge of your hard drive...right? Well a freeware called Ultimate Defrag does just that, it moves everything on the edge and there is no need for short stroking.
m
0
l
August 24, 2009 4:53:42 PM

short stroking will keep anything from being written to any other part of the disc...you determine how much of that you want to use. If the software does the same thing it is short stroking, but just calling it another name. If it allows writing to the other part of the disc, though, it will slow down your write times quite a bit. Plus you waste a lot of time moving the data around on the disc...of course if it forces the writes to the specified portion of the disc you are good to go. I will check it out, since I didn't know there was such a software.
m
0
l
a b G Storage
August 24, 2009 5:21:39 PM

The key variable is "recording density":

In order to maintain the same or similar recording density,
the armature's write heads must write binary digits ("bits")
at a frequency that is faster at outer tracks and
slower at inner tracks.

Thus, there is a step function in the recording logic,
which gradually reduces the bit recording frequency
as the armature moves from outer to inner tracks.

You can illustrate this yourself by drawing a series
of concentric circles, and placing equally spaced dots
along those circles: the inner tracks will have fewer
dots if all dots are equally spaced, relative to
adjacent dots.

You can also see a graphic illustration with the HDTune
benchmark software: http://www.hdtune.com


Western Digital have an excellent animation here:
search for "Intelliseek":

http://www.wdc.com/en/products/products.asp?driveid=338

then "View demo":

http://www.wdc.com/en/flash/index.asp?family=intellisee...

The inner tracks obviously have a shorter circumference:
in geometry, circumference of a circle = Pi x diameter
or 2 x PI x radius, where "Pi" = ~3.14159265


MRFS

m
0
l
a b G Storage
August 24, 2009 5:30:14 PM

In general, it has become a recommended policy
to size the C: system partition at 20-50GB
instead of formatting the primary HDD
as a single XP drive letter.

This policy produces short strokes for all OS I/O,
and effects a separation between user data files
and OS software files.

This policy also permits rapid restoration of
drive image files, written by software like
Symantec's GHOST, without causing a loss
of the user's data files e.g. if those data files
were added AFTER the last drive image file
was written.


Another excellent application of this knowledge
is to create a small first partition on a secondary HDD,
then to create a contiguous pagefile.sys using the
freeware CONTIG software.

This contiguous pagefile.sys will be assigned to
the lowest sector addresses in that first partition,
again resulting in "short strokes" when the OS
must "swap" programs to and from disk
e.g. when a Windows XP minimize button is clicked.


MRFS
m
0
l
a b G Storage
August 24, 2009 7:26:56 PM

Well...the software i was talking to you about, data is still wrote on random parts of the disc as usual but 1once a week, i use this software and pushes all the sectors with data and everything on my hard drive to the very possible edge :)  And either way i have a another beastly software called diskeeper which keeps my computer running at light speed since it has realtime defragmentation, all a bunch of stuff like frag shield, hyperfast, invisitasking technology, and it even defragments my paging file and master file table (MFT zone) during boot up.
m
0
l
!