home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
ftp.cs.arizona.edu
/
ftp.cs.arizona.edu.tar
/
ftp.cs.arizona.edu
/
icon
/
newsgrp
/
group01a.txt
/
000031_icon-group-sender _Tue May 23 16:34:23 2000.msg
< prev
next >
Wrap
Internet Message Format
|
2002-01-03
|
3KB
Return-Path: <icon-group-sender>
Received: (from root@localhost)
by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA27759
for icon-group-addresses; Tue, 23 May 2000 16:33:28 -0700 (MST)
Message-Id: <200005232333.QAA27759@baskerville.CS.Arizona.EDU>
From: gep2@terabites.com
Date: Tue, 23 May 2000 14:53:24 -0500
Subject: Hardware for HLLs
To: Icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 2949
>> ...and a better reference I have lost suggest Sun are on to this and
> have a java non-virtual machine.
> I checked that out and it looks like you're right. Oh, well, I guess there's
nothing new under the sun.
>> "Richard A. O'Keefe" <ok@atlas.otago.ac.nz> wrote:
>> And the market share for hardware that doesn't support good virus hosts is
fairly limited.
> 2 + 2 = ... Awright! Hardware virus acceleration! We're gonna get RICH!!!
Some years ago a company called Proximity Technology developed a marvelous LSI
chip to do proximity ranking of a series of [near-]matches. The idea was
wonderful, although within just a few years the software version was faster when
running on widely available standard PCs.
The reasons you don't see more of these specialized HLL processors are simple.
The market is limited; few people need the additional performance badly enough
to put up with the cost and configuration hassles (need an available slot and of
the right kind etc etc). Given the limited money available to support the
development, you'll probably never be able to keep up in the throughput
technology curve with the general-purpose, mass-market processors. Eventually
(and probably SOON) your balls-out high-speed specialized custom processor will
be blown away by a simple interpreter running on cheap generic iron.
If you go back about 7-8 years there was a lot of press about how Sun's
revolutionary SPARC RISC processor was going to put the Intel-family CISC
processors out of business, and redefine the PC. It hasn't really happened that
way, has it?
The HP9000 at one of my consulting customers (a $50K+ box) uses their
fancy-schmancy HP/PA microprocessor. (You will remember that the whole raison
d'etre of RISC was that by designing a brain-damaged processor with a crippled
primitive instruction set, they could turn the crank faster...) Anyhow, the
HP/PA "we can turn the crank faster" microprocessors runs at... wait for it...
100MHz. (!!)
Anyhow, after designing a specialized, custom-built processor (which only works
for SOME of the workload you have, and probably isn't even DRAMATICALLY faster
for that segment) you then have to figure out how to find the money to keep
REdesigning it to compete with the constantly raised bar coming from companies
like AMD and Intel (which probably have a LOT more money to play the high-stakes
game than you do).
So anyhow it's scarcely surprising that there aren't a lot of such specialized
HLL-based processors out there. If anybody was going to do it for JAVA, it
would probably be Sun... and even there, they haven't exactly been hugely
successful selling even their SPARC chips.
Gordon Peterson
http://web2.airmail.net/gep2/
Support the Anti-SPAM Amendment! Join at http://www.cauce.org/
12/19/98: the day the Conservatives demonstrated their scorn for their
fraudulent sham of representative government. Voters, remember it!