Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!wuarchive!waikato.ac.nz!canterbury!phys169
Newsgroups: comp.sys.ibm.pc.misc
Subject: Disk seek optimisation
Message-ID: <1991Jun5.111750.951@csc.canterbury.ac.nz>
From: phys169@csc.canterbury.ac.nz
Date: 5 Jun 91 11:17:49 +1200
Organization: University of Canterbury, Christchurch, New Zealand
Lines: 14

Here's a question somebody may be able to answer:

If software on a PC was to order read requests in such a way as to reduce head
movement (I think some file servers use this, called "elevator optimisation" or
similar), with the fast disks in use today, how much improvement is possible?
Or is the time required to calculate it significant compared to the head
movement time. I could look up disk specs and write some sample programs, but
what I need is some experience from people who have used it - is it generally a
worthwhile exercise? I suspect that the ideal is for the controller cards to
have the intelligence to do this, since some operating systems manage to get 
the CPU to do useful stuff while waiting for the disk, not DOS, of course! :-(

Thanks in advance,
Mark Aitchison, Physics, University of Canterbury, New Zealand.