Each cylinder contains a number of tracks, each
accessible by one of the heads on the drive (one head per surface). To improve efficiency,
the drive will normally use all of the tracks in a cylinder before going to the next
cylinder when doing a sequential read or write; this saves the time required to physically
move the heads to a new cylinder. Switching between heads is a purely electronic process
instead of a mechanical one. However, switching between heads within a cylinder still
requires a certain amount of time, called the head switch time. This is usually
less than the track switch time, and is usually on the order of 1 to 2 milliseconds.
(Seems kind of slow for an electronic process, doesn't it? The reason is that this time
includes all of the overhead of the switch as well; it is all of the time that passes
between when the read stops on one head and when it actually starts again on the next
one.)
Head switch time is not commonly discussed, but it is an important component of sustained transfer rate, since STR measures transfer
rate over larger reads or writes that encompass more than one track. See the discussion of
sustained transfer rate for more details. You may
also want to read about head and cylinder skew
here.
Head switch time is primarily influenced by the characteristics of the hard disk's controller. It does not vary greatly from drive model
to model or between manufacturers.
Tip: Even though it is
typically smaller than cylinder switch time, head switch time is more important to
transfer performance because head switches occur more often than cylinder switches (unless
you are using a single-platter disk). See the discussion of calculating of sustained transfer rate for more on this subject.
Next: Cylinder Switch Time
