🚀 go-pugleaf

In article <41c9ut$g73@usenet.ucs.indiana.edu>,
scott goehring <sgoehrin@copper.ucs.indiana.edu> wrote:
> preemptive multitasking is _less_ efficient than nonpreemptive
> multitasking, because of preemption overhead.

Nonsense.

The overhead has to be paid one way or the other, whether it's by explict
delay calls in the application or clock interrupts. It's *more* efficient
to use interrupts, because they will occur at the same rate regardless of
CPU performance (so you don't have to recompile your software when you go
to a faster CPU), and because they avoid costly redundant calls to the
scheduler from the application.

> many supercomputers run
> either a nonpreemptive OS or a an OS with limited preemption (the only
> time a job can be preempted is when a job with higher priority than
> the running job enters the run queue),

For the job to enter the run queue something must have preempted the job
currently running. What this is is "realtime scheduling". Scheduling policy
is a separate issue from the class of multitasking. For example, Polyforth
uses cooperative multitasking but realtime scheduling, while the Amiga uses
preemptive multitasking and realtime scheduling, and Windows uses cooperative
multitasking and fair-share scheduling, and UNIX traditionally uses preemptive
multitasking and fair-share scheduling.

> because this maximizes CPU utilization.

Even on my Amiga the "preemption overhead" is too small to measure.

> however, maximizing CPU utilization is not generally the goal of most
> systems; balancing CPU utilization with user responsiveness is a more
> common goal.

And since a user interface is a real-time problem, the best way of doing
this is to provide a preemptive real-time scheduler, like the Amigas, which
provides *excellent* user responsiveness even on a 512K 16-bit machine
running at 7.14 MHz.

> i guess this all goes to prove that "those who do not learn from
> history are doomed to repeat it."

No kidding. Ironic, eh?

> all of this nonsense has been done
> before, just on different computers by different companies.

Yep.
-- 
Peter da Silva    (NIC: PJD2)      `-_-'         "Har du kramat din varg idag?"
Bailey Network Management           'U`
1601 Industrial Blvd.        Sugar Land, TX  77478  USA         +1 713 274 5180
Bailey pays for my technical expertise.         My opinions probably scare 'em.