"Toy" computers http://www.conmicro.cx/hercules

From: Christopher Smith <csmith_at_amdocs.com>
Date: Mon Apr 29 15:35:36 2002

> -----Original Message-----
> From: Raymond Moyers [mailto:rmoyers_at_nop.org]

> They don't pay those costly License fees and support contracts
> out of ignorance. todays mainframes are exteamly muscular,
> having taken advantage of the same technical advancements

Indeed. I would probably buy one regardless of the cost that the
power bill would cause me. :)

> Well, i had assumed most would know that a box that typically
> served 5000 seats or more was powerfull even as most might
> not know what it is that really sets these things apart.

That's exactly where the misunderstanding came in, I think. It's
taken as obvious, for the most part, but your statement that the
entire campus full of hardware would be emulated on a single peesee
seemed to say differently.

At any rate, it makes much more sense now.

> Channel I/O for example, translating to terms and concepts
> more familiar to those without OPER console time, imagine a
> "PC" where every orifice was pumped by its own dedicated
> DMA controller, where you can have 65535 of these devices.
> and fill em all up with no load on the CPU.

Not having much time on mainframe systems, myself, I still imagined
something like this, and you can pretty much infer it from talk of
the things.

> Remember when the mickysoft press release parroting nattering
> nabob computer press was declaring the death of the mainframe ?

They're not still doing it? That surprises me.

> and very humorous events like when the idiot press would read a
> product release about NT being ported to an FSIOP card, and run
> to print "NT Ported to the mainframe ! "

I would like to see that, actually -- the article, I mean.

VM/NT -- Heh. I wonder what the "error number" is for all of the
NT IOPs in the machine simultaneously jumbling up their RAM.

> A FSIOP card, File Server I/O Processor, is a PC on a card that
> plugs into the mainframe so that it can share mainframe DASD
> (disk) or have a faster channel for I/O to PC based middleware,
> it certainly isn't NT running on the mainframe in the manner
> the hapless readers of these sorry articles was led to believe.

I suppose such a product would be good if you need it. It could do
better than to run NT. Maybe they should "port CP/M" to the mainframe.

> As for CPU power, PC's are certainly as fast per CPU in a general
> sense as a mainframe, but without the I/O capacity could never
> hope to replace a mainframe anytime soon.

I don't know exact numbers, but honestly, the CPU in a modern peesee isn't
the weak spot at all. Generally there's some kind of bottleneck (or five)
that needs repaired in the design.

> The PC running something decent does rival the power of a
> mainframe or a supercomputer of yesterday however,

I wouldn't doubt that it might compare for certain (probably single-user)
non-io-bound applications. I'm not sure I can make that conclusion for
a supercomputer at all. When is "yesterday" in this context? :)

> utter garbage that at last time i checked w2k needed 64megs
> of ram for the installer to run ? how absurd !

Actually, I'd have expected it to need more than that.

> Compare with the size of bsd/linux/unix that will still run on
> a machine with 4megs or compare with a mainframe nucleus
> and you see that they on the other hand, have stayed small
> tight and fast. and assembler is still very mainstream
> on the mainframe where thruput of massive loads is still
> the focus.

Your comment about mainframes having "stayed small" is oddly amusing,
but perhaps it's because I got no sleep yesterday.

> mips or so emulating 360/70/90 instructions is testament
> that the lowly PC has become very muscular in its own right.

Indeed. I'm sure there are some tasks to which a peesee is well suited,
but the problem is -- aside from the common operating environment --
the baggage in the design, still hanging around from the beginning.
(probably not too well-thought-out back then ;) They probably should have
done something ground-up by now to take advantage of newer cores, bus
technology, etc.

Chris


Christopher Smith, Perl Developer
Amdocs - Champaign, IL

/usr/bin/perl -e '
print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");
'
 
Received on Mon Apr 29 2002 - 15:35:36 BST