|
How A Hard Drive Is
Physically Organized To Hold Data
Hard drives used in
today’s microcomputers have their origin in the hard drives of early
mainframe computers of the 1970s.
These drives consisted
of large platters or disks that were much larger and thicker than
phonograph records. Several platters were stacked together with enough
room to allow read/write heads to move back and forth between the
platters. A drive requires a controller board filled with ROM programming
to instruct the heads how to move across the platters and write and read
data. All heads moved in unison while the platters spun at a fast speed.
Application programmers of the 1970s wrote their programs so that data was
spaced evenly over the disks, so that the heads moved as little as
possible while reading or writing a file. In today’s systems, there are
several layers of software between data stored on a drive or disk and the
application software that might be reading its data from or writing its
data to the drive. Thus, application programmers do not need to concern
themselves with how data is stored on a hard drive.
Hard drive structure
and function have not changed, however. Modern hard drives have two or
more platters that are stacked together and spin in unison. Read/write
heads are controlled by an actuator and move in unison back and forth
across the disk surfaces as the disks rotate on a spindle. There are
several types of hard drives for PCs, all using a magnetic medium; the
data on all of them is stored in tracks and sectors. Just as with disks,
data files are addressed on the hard drive in clusters made up of one or
more sectors. The figure above shows a hard drive with four platters. All
eight sides of these four platters are used to store data, although on
some hard drives the top side of the first
platter just holds
information used to track the data and manage the disk.
Each side or surface of
one platter has a head, the
electromagnetic device used to read data from or write data to the
surface. The drive in the above figure has eight heads.
Each side of each
platter is divided into tracks and sectors. A cylinder is comprised of a
stack of identical tracks from each surface. For example, if you took the
outermost track from each surface and stacked them one on top of each
other, that would be a cylinder.
The number of tracks in
a cylinder depends on the number of platters on the drive. If a disk has
300 tracks per surface, then it also has that same number of cylinders.
Data is written to the drive beginning at the outermost track, just as
with disks. The entire first cylinder is filled before the read/write
heads move inward and begin filling the second cylinder.
For older hard drives,
the tracks closer to the center of a platter are smaller, but have to
store the same amount of data as the larger tracks toward the outside of a
platter. At some point as the heads move toward the center of the drive
and the tracks get smaller and smaller, the read/write heads have to
adjust the way they write data so that sectors store a consistent number
of bytes, even if they are different physical sizes.
Two methods can be used
to adjust for the smaller tracks: write pre-compensation and reduced write
current.
Write pre-compensation
speeds up the writing
of data to the drive as the tracks become smaller near the center of the
platters. If a hard drive uses write pre-compensation, it indicates at
what track or cylinder the pre-compensation begins. Some drives don’t use
this method. Some tables list the write pre-compensation as the total
number of cylinders for the drive type. Interpret this to mean that
pre-compensation
is not used.
<Previous>
<Home>
<Next>
|
