Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site brl-tgr.ARPA
Path: utzoo!watmath!clyde!burl!ulysses!gamma!epsilon!zeta!sabre!petrus!
bellcore!decvax!genrad!panda!talcott!harvard!seismo!brl-tgr!tgr!hpcea!
hpcehfe!michael%hplabs.cs...@CSNET-RELAY.ARPA
From: michael%hplabs.cs...@CSNET-RELAY.ARPA
Newsgroups: net.unix-wizards
Subject: Dhrystone scores for HP9000-500
Message-ID: <1247@brl-tgr.ARPA>
Date: Thu, 5-Sep-85 08:16:29 EDT
Article-I.D.: brl-tgr.1247
Posted: Thu Sep 5 08:16:29 1985
Date-Received: Sat, 7-Sep-85 05:26:29 EDT
Sender: n...@brl-tgr.ARPA
Lines: 86
In answer to a recent question seen here in UNIX-WIZARDS: yes the HP9000
can interface to other systems. It supports IEEE 802.3 with an option for
Ethernet connection and supporting firmware, RJE, (I have heard X.25 but
don't find it in the price list), as well as HP's own 500 and 200 or 300
networking system.
Also I ran the Dhrystone benchmark program recently posted to the net on
a series 500 machine with one B series CPU card. I have added the results
to the table of results for other machines that came with the program
source. The 500 is a stack machine and has no registers in the sense that
c requires for its 'register' designation. Thus the -DREG=register option
had no effect on performance. The times were obtained with no other users
on the system and only the usually transparent background processes running
(getty ect.) most (if not all) of which would be blocked. I include this
comment because I found the benchmark to be somewhat variable giving 3
different results on about 15 runs.
>/*
> * "DHRYSTONE" Benchmark Program
> *
> * Version: C/1
> * Date: 12/01/84
> * Author: Reinhold P. Weicker, CACM Vol 27,No 10, 10/84 pg. 1013
> * Translated from ADA by Rick Richardson, vaximile!rer
> * Every method to preserve ADA-likeness has been used,
> * at the expense of C-ness.
> * Compile: cc -O dry.c -o drynr : No registers
> * cc -O -DREG=register dry.c -o dryr : Registers
> * Defines: Defines are provided for old C compiler's
> * which don't have enums, and can't assign structures.
> * The time(2) function is library dependant; One is
> * provided for CI-C86. Your compiler may be different.
> * Run: drynr; dryr
> * 50000 Runs are made and the time is printed
> *
> * MACHINE MICROPROCESSOR OPERATING COMPILER DHRYSTONES/SEC.
> * TYPE SYSTEM NO REG REGS
> * -------------------------- ------------ ----------- ---------------
> * IBM PC/XT 8088-4.77Mhz PCDOS 2.1 CI-C86 2.1 ???? ????
> * IBM PC/XT 8088-4.77Mhz PC/IX cc ???? ????
> * IBM PC/XT 8088-4.77Mhz VENIX/86 2.0 cc 297 324
> * IBM PC/AT 80286-6Mhz PCDOS 3.0 CI-C86 2.1 666 684
> * ATT 3B2/300 MAC32-?Mhz UNIX 5.0.2 cc 735 806
> * IBM PC/AT 80286-6Mhz VENIX/86 2.1 cc 961 1000
> * ATT PC7300 68010-10Mhz UNIX 5.2 cc 1041 1111
> * Sun2/120 68010-10Mhz Sun 4.2BSD cc 1136 1219
> * PDP 11/70 - UNIX 5.2 cc 1162 1250
> * IBM PC/AT 80286-7.5Mhz VENIX/86 2.1 cc 1190 1315 *
> * VAX 11/780 - UNIX 5.2 cc 1515 1562
> * ATT 3B20 - UNIX 5.2 cc 1515 1724
> * Macintosh 68000-??Mhz ???? ?? ???? ????
> * Lisa 68000-??Mhz ???? ?? ???? ????
> * HP9000-500 B series CPU HP-UX 4.02 cc 1724
> * (Note: that this benchmark can vary alot, 1724 was highest 1666 was
> * the mode. This is for a single CPU system. Multiply by 1.9 or 2.8
> * for 2 or 3 cpu systems respectively. The 9000-500 is a stack
> * machine and consequently ignores the c register statement.
> *
> * * 15Mhz crystal substituted for original 12Mhz;
> *
> **************************************************************************
> *
> * The following program contains statements of a high-level programming
> * language (C) in a distribution considered representative:
> *
> * assignments 53%
> * control statements 32%
> * procedure, function calls 15%
> *
> * 100 statements are dynamically executed. The program is balanced with
> * respect to the three aspects:
> * - statement type
> * - operand type (for simple data types)
> * - operand access
> * operand global, local, parameter, or constant.
> *
> * The combination of these three aspects is balanced only approximately.
> *
> * The program does not compute anything meaningfull, but it is
> * syntactically and semantically correct.
> *
> */
Michael A. Moran
...!hplabs!hpcea!hpcehfe!michael
Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site im4u.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!mit-eddie!think!harvard!seismo!
ut-sally!im4u!jsq
From: j...@im4u.UUCP (John Quarterman)
Newsgroups: net.unix-wizards
Subject: Re: Dhrystone scores for HP9000-500 (really networking)
Message-ID: <527@im4u.UUCP>
Date: Wed, 18-Sep-85 10:22:16 EDT
Article-I.D.: im4u.527
Posted: Wed Sep 18 10:22:16 1985
Date-Received: Fri, 20-Sep-85 06:30:23 EDT
References: <1247@brl-tgr.ARPA>
Reply-To: j...@im4u.UUCP (John Quarterman)
Organization: U. Texas CS Dept., Austin, Texas
Lines: 43
In article <1...@brl-tgr.ARPA> michael%hplabs.cs...@CSNET-RELAY.ARPA writes:
>In answer to a recent question seen here in UNIX-WIZARDS: yes the HP9000
>can interface to other systems. It supports IEEE 802.3 with an option for
>Ethernet connection and supporting firmware, RJE, (I have heard X.25 but
>don't find it in the price list), as well as HP's own 500 and 200 or 300
>networking system.
It's good that someone at HP responded. Unfortunately, that's not really
an answer. IEEE 802.3 gets you up to the network layer, ISO layer 3,
or at least the bottom half of it, since 802.3 doesn't provide internetwork
addressing. You still need layers 4-7, especially layer 4, transport.
RJE is an application package whose transport support is not especially
usable for other applications. X.25 is a network layer protocol. HP's
500, 200, or 300 networking systems may provide layers 3.5-7, but which
protocols? If not TCP/IP, XNS, or the developing ISO protocols, they're
not terribly useful.
Perspective: At UT we run UNIX on machines made by more than half a
dozen different manufacturers, not to mention the TOPS-20, VMS, LISPM,
and other systems. *All* of them, from small workstations to large
mainframes, can internetwork over our Ethernets and to the ARPA
Internet using the TCP/IP protocol suite. Remote login, file transfer,
mail, and, on the UNIX machines, remote execution, remote procedure call,
remote interactive conversations, etc. With the exception of some
of the XEROX machines, which use XNS (we can deal with that), the AT&T
3B machines, which only speak 3BNET (i.e., which can't talk to any
other manufacturer's machines) and the HP machines, which don't even
handle ethernet at the moment. This kind of diverse networking
situation is quite common at large corporations and universities.
Don't get me wrong: HP makes what appears to be really nifty hardware.
But HP can't really expect us (or other organizations with similarly
diverse environments) to throw away all our other machines and replace
them all with HP hardware just so we can talk to the existing HP machines.
This means our HP machines sit unused. Fortunately, they were donated.
One suspects HP would sell more of them if they internetworked.
AT&T has announced that they are putting TCP/IP on their 3Bs.
What is HP doing?
--
John Quarterman, UUCP: {ihnp4,seismo,harvard,gatech}!ut-sally!jsq
ARPA Internet and CSNET: j...@sally.UTEXAS.EDU, formerly j...@ut-sally.ARPA