15 May
2015
15 May
'15
4:14 p.m.
A correspondent just pointed me to a nice collection of photos of a (the?) 11/74 MP
system:
http://oboguev.livejournal.com/2696291.html
I assume this is the one at DEC used by the RSX-11M+ group for their multiprocessor
support work. As far as I know these didn?t become a product. I remember the name of the
machine as ?Cerberus?, not sure what that label ?daemon? refers to. Maybe a DECnet node
name?
paul
16 May
16 May
6:59 p.m.
On 15 May 2015 at 16:14, <Paul_Koning at dell.com> wrote:
A correspondent just pointed me to a nice collection
of photos of a (the?) 11/74 MP system:
http://oboguev.livejournal.com/2696291.html
I heard it was supposedly called Phoneix (PHEANX as DECnet node name).
And yes, that is one of the 11/74 systems, and that is the RSX-11/M+
group's machine, from what I recall of previous discussions of it.
It's interesting to see the different names that the 11/74 could have
had as a product painted on the machine(s). I also recall somewhere
that one of the other names it could have had was 11/72. (I think for
a dual processor configuration?)
It's neat that the newest version of RSX-11/M+ still supports the
11/74 (try it on E-11).
Cheers,
Christian
--
Christian M. Gauger-Cosgrove
STCKON08DS0
Contact information available upon request.
17 May
17 May
7:18 a.m.
On 2015-05-17 00:59, Christian Gauger-Cosgrove wrote:
On 15 May 2015 at 16:14, <Paul_Koning at
dell.com> wrote:
Sortof. It's from Dave Carroll, and the 11/74 at Colorado Springs. This
was PHEANX. It was after the RSX group didn't exist anymore, and whay
gathered what remained of 11/74 systems into the Colorado Springs site,
since that is where all PDP-11 stuff went finally at DEC, before it
moved over to Mentec.
A correspondent just pointed me to a nice
collection of photos of a (the?) 11/74 MP system:
http://oboguev.livejournal.com/2696291.html
I heard it was supposedly called Phoneix (PHEANX as DECnet node name).
And yes, that is one of the 11/74 systems, and that is the RSX-11/M+
group's machine, from what I recall of previous discussions of it. It's interesting to see the different names that
the 11/74 could have
had as a product painted on the machine(s). I also recall somewhere
that one of the other names it could have had was 11/72. (I think for
a dual processor configuration?)
I've seen/heard some reference to 11/72, but all the papers from DEC on
PDP-11 MP systems called them 11/74 that I can remember. No matter how
many CPUs was in the system.
It's neat that the newest version of RSX-11/M+
still supports the
11/74 (try it on E-11).
It does indeed work. MIM is available for public access. Telnet to
mim.update.uu.se, login as guest with password guest.
Run RMS, and notice the CPA, CPB, CPC and CPD on the memory page.
However, E11 have some bugs, so it don't run stable with more than one
CPU online, so I normally only have one CPU running.
Johnny
1:01 p.m.
On Sun, May 17, 2015 at 01:18:51PM +0200, Johnny Billquist wrote:
It does indeed work. MIM is available for public
access. Telnet to
mim.update.uu.se, login as guest with password guest.
Run RMS, and notice the CPA, CPB, CPC and CPD on the memory page.
However, E11 have some bugs, so it don't run stable with more than one CPU
online, so I normally only have one CPU running.
I'd love to attack this ... I've never been able to reproduce the trouble,
probably because my home machines don't get the load that MIM does. Even
if E11's mP feature is officially unsupported, it was a *crazy* amount of
work and I'd really like to put a tougher hide on it.
John Wilson
D Bit
1:31 p.m.
On 2015-05-17 19:01, John Wilson wrote:
On Sun, May 17, 2015 at 01:18:51PM +0200, Johnny
Billquist wrote:
I'd be happy to. How do you want to go about it?
If I enable a second CPU, that second CPU will hang eventually. Want
access to the E11 console to poke around? Or you want to try something else?
Johnny
It does indeed work. MIM is available for public
access. Telnet to
mim.update.uu.se, login as guest with password guest.
Run RMS, and notice the CPA, CPB, CPC and CPD on the memory page.
However, E11 have some bugs, so it don't run stable with more than one CPU
online, so I normally only have one CPU running.
I'd love to attack this ... I've never been able to reproduce the trouble,
probably because my home machines don't get the load that MIM does. Even
if E11's mP feature is officially unsupported, it was a *crazy* amount of
work and I'd really like to put a tougher hide on it. 
4:16 a.m.
That system blows my mind. I wonder how it would have been to maintain.
Did any bits survive? I think I've seen a front panel on CHMs web site.
/P
On Fri, May 15, 2015 at 08:14:13PM +0000, Paul_Koning at Dell.com wrote:
A correspondent just pointed me to a nice collection
of photos of a (the?) 11/74 MP system:
http://oboguev.livejournal.com/2696291.html
I assume this is the one at DEC used by the RSX-11M+ group for their multiprocessor
support work. As far as I know these didn?t become a product. I remember the name of the
machine as ?Cerberus?, not sure what that label ?daemon? refers to. Maybe a DECnet node
name?
paul
5:54 a.m.
On 2015-05-15 22:14, Paul_Koning at Dell.com wrote:
A correspondent just pointed me to a nice collection
of photos of a (the?) 11/74 MP system:
http://oboguev.livejournal.com/2696291.html
Ah, yes. Those are the photos from Dave Carroll. I've send the original
link several times in the past. They are on Google, and there are links
on that page to them.
I assume this is the one at DEC used by the RSX-11M+
group for their multiprocessor support work. As far as I know these didn?t become a
product. I remember the name of the machine as ?Cerberus?, not sure what that label
?daemon? refers to. Maybe a DECnet node name?
Sortof. The M+ group had a system called CASTOR, which was in MA. This
was a 4-cpu system. The DECnet group also had an 11/74, but a 2-CPU
system. I don't remember for sure, but I think that was POLLUX (And yes,
these were the DECnet node names on EASYnet.) Finally, I *think* that
field service also had an 11/74. I don't know how many CPUs. This
machine might have had a name based on Cerberus.
At some point in the early 90s, the existing/remaining 11/74 components
were all shipped to Colorado Springs, where a new 4-CPU system was put
together. This system was then used by Dave Carroll for further RSX
development, and it was called PHEANX (the only free spelling of Phoenix
still available, and the reason for Phoenix should be somewhat obvious).
It remained in service until 2000, when it developed a hardware problem.
It was dismantled some time after that, and Dave took these pictures
shortly before the dismantling.
(It I remember/understood things right.)
Johnny
3388
days inactive
3390
days old
6 comments
5 participants
participants (5)
-
bqt＠update.uu.se -
captainkirk359＠gmail.com -
Paul_Koning＠Dell.com -
pontus＠Update.UU.SE -
wilson＠dbit.com