> Richard Erlacher wrote:
> >
> > Years ago, the distinction between minicomputers and microcomputers was
> > based on the notion that microcomputers had a synchronous bus, while mini's
> > had an asynchronous bus. Now that we all seem to own at least one
> > microcomputer not only with an asynchronous bus but with computing power far
> > in excess of the best that IBM et. al. could muster back in the '80's, I'm
> > not so sure I can make a simple distinction like that.
> >
> > As for mainframes, well, since you can buy a more powerful computer for what
> > it would cost to operate a mainframe of yesteryear for just one day, I guess
> > nobody will miss 'em.
Last time this subject came up, it was pointed out to me that mainframes are
much more batch oriented than micros or minis. They're designed to be fed
their job, then left alone to plow through the data, then cough up a report -
they're not optimised for direct user interaction. And there's still a place
in the world for modern mainframes - IBM still makes them - because some jobs
require more computing horsepower and reliability than modern pcs can deliver,
and hooking one clydsdale to a big job instead of 400 squirrels is alot easier
and cost effective to manage.
And as for reliability, let me draw my line in the sand and say that I think
using PCs for mission critical servers is not a good idea. Even if you put
a decent OS on them - Linux, Solaris, FreeBSD, etc, they're not designed or
supported the way a real mini or mainframe is. 'course the flip side is you
can keep another machine around for a spare for a reasonable cost...
--
Jim Strickland
jim_at_DIESPAMMERSCUMcalico.litterbox.com
-----------------------------------------------------------------------
BeOS Powered!
-----------------------------------------------------------------------
Received on Fri Dec 15 2000 - 13:58:06 GMT