HP3000-L Archives

December 2001, Week 2

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Reply To:
Date:
Tue, 11 Dec 2001 18:06:45 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (42 lines)
X-no-Archive:yes
In two years, we will have new processors, with more than double the number
of transistors, and who knows at what speed they will run? In five years, we
will be on a third generation of processors, eight times as many
transistors, and presumably exponentially faster speeds. Even if a company
were to buy the very last model that hp sells, by EOL, they should be able
to buy a system with no less than four times the transistors. And that is
being quite optimistic about the processors that hp would select for its
last model. Also, it is not unreasonable to expect Linux to be able to
handle more processors than MPE does now, although it remains to be seen
what this would mean for emulation.

As I understand it, products like Amisys only run on MPE, and the market
offers nothing else like it. For at least one organization, there is no
reason on their horizon to leave MPE. It does its job better and at a better
price than those systems which are being considered for migration. For
several companies, there is really very little choice but to continue with
MPE if at all possible, because a migration is either too costly (possibly
exceeding the annual net profit of the company), or simply too risky. A
reasonable course of action for such a company would be to take an emulated
MPE, and run their current batch processing in parallel. Most problems would
rapidly become apparent, as would a perfect parallel, with appropriate
audits.

I cannot say I care at all for the hand that has been dealt to us. I've even
wondered if it is time to consider returning to teaching, and getting away
from IT as a profession. I would love to seen MPE native on IA64, and even
offered my thoughts on how this could still happen (by CSY being sold to
another proprietary computer manufacturer).

So, I think that the two proposals, for an emulated MPE and for an OpenMPE,
are the best thing on the table for those who choose to or who want to stay
on it, to be able to do so. I have my own questions about both. But, like
the joke about democracies, they seem to be the second worst alternative,
with the worst alternative being all the others.

Greg Stigers
http://www.cgiusa.com

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

ATOM RSS1 RSS2