🚀 go-pugleaf

In article <1136@paris.ics.uci.edu> Andrew Kwan <kwan@io.ics.uci.edu> writes:
>rotation speed and the number of sectors per track.  It is not
>dependent on the number of turns it has to make - sector accesses do
>not require having to wait until the track makes a full turn to get to
>the next sector to be accessed, only for the next one or two sectors
>to pass (to get to the one you want).

  Assuming that you want to read all the sectors in a track, which happens
quite often when reading a contiguous {sp??} sequence of sectors then the  
interleave does make a difference!!  The SE can read a whole track of
information in two disk rotations {latency time}, while it would take the
plus at a 3:1 interleave 3 disk rotations to read the same information.
ie. approx. 50% longer on the plus for the same about of information.

  I think your basic premise is correct if you want to only read one sector,
but this rarely happens.  I believe if you think disk accessing through just
a little farther than you did in your previous posting.  When accessing
several sectors in a row, latency can make a significant diffence is access
time and effective throughput.

  If interleave didn't make any difference then what point would there be in
having it. :-)
-- 
David M. O'Rourke                                  dorourke@polyslo.calpoly.edu

"If it doesn't do Windows, then it's not a computer!!!"
Disclaimer: I don't represent the school.  All opinions are mine!