thompson@PAN.SSEC.HONEYWELL.COM (John Thompson) (09/23/90)
One tiny 'gotcha' with the new series 400t nodes running in a token rung environment. It's not anything major, but it might confuse someone else like it confused me for a while. We run TCP/IP services, and, since we have very few nodes (3) with more than one controller, we used the 'net0' device on ifconfig. Then we got the 400t nodes in, and they didn't want to start TCP/IP service up automagically. GOTCHA! These nodes have 2 controllers! 'net0' is now an ambiguous device reference, and ifconfig does nothing. Changing to 'dr0' made it work. Question for HP/Apollo people who are reading this: Could you make 'net0' be the primary boot device? Failing that, a mention of this 'feature' would be nice if included in the upgrade literature. John Thompson (jt) Honeywell, SSEC Plymouth, MN 55441 thompson@pan.ssec.honeywell.com As ever, my opinions do not necessarily agree with Honeywell's or reality's. (Honeywell's do not necessarily agree with mine or reality's, either)
ranous@hplabsb.HP.COM (Alex Ranous) (09/25/90)
In article <9009222118.AA24683@pan.ssec.honeywell.com> thompson@PAN.SSEC.HONEYWELL.COM (John Thompson) writes: >GOTCHA! These nodes have 2 controllers! 'net0' is now an ambiguous >device reference, and ifconfig does nothing. Changing to 'dr0' >made it work. > >Question for HP/Apollo people who are reading this: >Could you make 'net0' be the primary boot device? Failing that, a >mention of this 'feature' would be nice if included in the upgrade >literature. > > >John Thompson (jt) >Honeywell, SSEC >Plymouth, MN 55441 >thompson@pan.ssec.honeywell.com I'm not sure what literature was shipped w/ your node, but in the first item, 4.1.1, in the Bugs, Limitations, and APR's section of the 10.3 release notes, this limitation is mentioned. This limitaion is true on any Apollo w/ more than one interface, not just the new 400 series machines. The limitation is also noted in the comments right before the ifconfig command in the rc.local file, for 10.2 and 10.3. Alex Ranous Information Architecture Group Hewlett-Packard
chen@digital.sps.mot.com (Jinfu Chen) (09/27/90)
Disclaimer: There are lies, lies, and benchmarks.
Having said that, here are some benchmark results ran on Apollo DN series
and the new 400t. Our 400t upgrade hasn't shown up yet (P.O. still on its
way :-) and I only have 30 minutes access to one that in another Motorola
facility. Codes that run on 400t are exactly the same as the ones use in
other 68020/30 nodes.
System O.S. Compiler Dhrystone Dhrystone Whetstone
1.1 2.1 (reg/noreg)
----------------------------------------------------------------------------
DN300 Aegis SR 9.7 cc 4.89 962 813.0/813.0 377059
DN320(PEB) Aegis SR 9.7 cc 4.89 950 773.2/771.2 173410
DN660 Aegis SR 9.6 cc 4.80 1787 379747
DN330 Aegis SR 9.7 cc 4.89 2106 1705.5/1704.5 540540
DN560 Aegis SR 9.6 cc 4.80 2887 674157
DSP90 Domain/OS 10.1.0.11 cc 6.7(316) 3282 2582.2/2579.1 759494
DN3000 Domain/OS 10.2 cc 6.7(316) 4380 3440.0/3438.8 789474
DN570-T Aegis SR 9.7 cc 4.89 4749 4267.4/3856.0 895522
DN570-T(FPX)Aegis SR 9.7 cc 4.89 4622 3934.9/3969.3 1463414
DN2500 Domain/OS 10.2 cc 6.7(316) 7271 5543.2/5499.5 1764705
DN4000 Domain/OS 10.2 cc 6.7(316) 10056 7579.6/7612.3 1764706
DN4000(FPA1)Domain/OS 10.2 cc 6.7(316) 9971 7722.0/7575.8 2500000
DN3500 Domain/OS 10.1.1.2 cc 6.7(316) 9312 7208.1/7180.5 2307692
DN4500 Domain/OS 10.1.1.2 cc 6.7(316) 16291 12881.1/12776.8 3000000
400t Domain/OS 10.03.a cc 6.7(316) 22388 12024.0/12140.8 4285714
DN10000 Domain/OS 10.1 cc 6.7(325) 43859 35714.3/35756.9 20000000
----------------------------------------------------------------------------
System configurations (as best I could find):
System CPU FPU RAM
--------------------------------------------
DSP90 68020 - ? 68881 3.0Mb
DN300 68010 - ? ? 1.5Mb
DN320 68010 - 12 Mhz ? 1.5Mb
DN320(PEB) 68010 - 12 Mhz ? 1.5Mb
DN330 68020 - 12 Mhz 68881 2.0Mb
DN560 2901 - 20 Mhz ? 3.0Mb
DN570T 68020 - ?? Mhz 68881 8.0Mb
DN570T(FPX) 68020 - ?? Mhz FPX 8.0Mb
DN660 2901 - 20 Mhz ? 4.0Mb
DN2500 68030 - 20 Mhz 68882 16.0Mb
DN3000 68020 - 15 Mhz 68881 8.0Mb
DN3500 68030 - 25 Mhz 68882 16.0Mb
DN4000 68020 - 25 Mhz 68881 8.0Mb
DN4000(FPA) 68020 - 25 Mhz Weitek 8.0Mb
DN4500 68030 - 33 Mhz 68882 16.0Mb
400t 68030 - 50 Mhz 68882 16.0Mb
DN100000 Prism - 16 Mhz Buitlin 128.0Mb
--------------------------------------------
Note:
Different compiler optimization switch and inline functions are used.
I no longer have access to some of those older hardware, such as hi-tech
heater DN660. Therefore, it's hard to compare performance under similar
conditions (OS, compiler version) with newer hardware.
--
Jinfu Chen (602)898-5338
Motorola, Inc. SPS Mesa, AZ
...uunet!motsps!digital!chen
chen@digital.sps.mot.com
CMS: RXFR30 at MESAVM
----------jtevik@gisplot1.nrri.umn.edu (John Tevik) (09/27/90)
In article <4d0bef1c.12c9a@digital.sps.mot.com> chen@digital.sps.mot.com (Jinfu Chen) writes: >System configurations (as best I could find): > > System CPU FPU RAM > -------------------------------------------- . . . > DN560 2901 - 20 Mhz ? 3.0Mb ^^^^^^^^^^^^^^^^^^^^^^^^ Mine's got a 68020 at 12 Mhz with an '881 ---------- John Tevik jtevik@gisplot1.nrri.umn.edu (128.101.45.101) Programmer University of Minnesota - Duluth Natural Resources Research Institute
krowitz@RICHTER.MIT.EDU (David Krowitz) (09/27/90)
To the best of my knowledge, all DN560's were built with 16 Mhz 68020/68881 CPU's, as were non-Turbo DN570/580/590's. Turbo models of the 570/580/590 had 20 Mhz 68020/68881 CPU's and could include an optional FPX floating point accelerator built around one of the Weitek FPU chip sets. Our DN560 is most definitely a 16 Mhz 68020/68881. -- David Krowitz krowitz@richter.mit.edu (18.83.0.109) krowitz%richter.mit.edu@eddie.mit.edu krowitz%richter.mit.edu@mitvma.bitnet (in order of decreasing preference)