asked on

Cylinder Skewing vs Defragmentation

"Cylinder skewing" technique was surpose to improve Hdd performance by reducing the disk time lost during normal head steps , which identify and read the desired information from a track within one disk rotation. So why do a "defragmentation" to bring the scattered data around the cylinder to a contiguous position i.e. data on same track / sector when "cylinder skewing" make data saved on different track / sector.

ASKER CERTIFIED SOLUTION

jhance

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

Ryan Rowley

"Cylinder skewing" Do you mean the disk Interleave Ratio
Or simular techniques to reduce the amount of time the
disk has to wait for the data to come under the R/W heads?

Those are Hardware level processes that have little to do
with the Operating System Software Level Defragmentation.

Defrag trys to move all the data of a single file into
contiguous series of addresses in the fat,vtoc or supernode of your High level disk structure.

You might say your "Cylinder skewing" is to your
low level formated disk structure, what defrag is to
your high level formated disk structure.

You use both, not one or the other.

Many modern Harddrives are now fast enough that they
can be effecient with a 1:1 ratio. Changing the lowlevel
characteristics makes no change in performance of the drive. This is do to the slowness of the overhead of the
drivers, O/S and applications being run.

A 15000 RPM SCSI fast/wide is less likely to need adjusting than a 5400 RPM old IDE drive.

jhance

>>effecient with a 1:1 ratio.

Let me point out that the term "1:1 ratio" refers to SECTOR INTERLEAVE and NOT cylinder skew. These are DIFFERENT things but somewhat related.

hhheng

ASKER

Agree with jhance about interleave which applies to Sector while Cylinder skewing applies to Track.

For example, if a drive reads a sector and skips a sector to reach the next sequential sector, interleave factor would be 1:3 and so on.

As I understand it , defragmentation is gathering of data scatter over different tracks/sectors so that data are placed in interleave ration of 1:1 and on contiguous track.

Thanks and cheers...