A List of S/360/370/390 Operating Systems?

34 views
Skip to first unread message

Metz, Seymour

unread,
Dec 8, 1997, 3:00:00 AM12/8/97
to

The 360/44 was not standard and was not a variant of the 360/40. The 360/40
had microcode in TROS to simulate the S/360 architecture, while the 360/44
was hardwired (and missing the SS instructions).

TSS and CP/67 were not TSO predecessor systems; they were unrelated. TSS,
IMHO, was better than MVS with TSO.

OS/360 with HASP was older than TSS.

ASP ran on machines much less expensive than a 360/91, e.g., 360/65. In
fact, the ASP support for the 7090 emulator was only relevant to the 360/65.
Is it true that HASP stands for Half ASP <g, d & r>?

Shmuel (Seymour J.) Metz
----------
From: HollyWiz
To: smetz; IBM-MAIN
Subject: Re: A List of S/360/370/390 Operating Systems?
Date: Saturday, December 06, 1997 1:20AM

I don't think the 360/44 was standard. Although it id does contain the 360
part, I believe the 44 was a particular version of the MOD 40. It probable
did run on other 360s. I know I worked on a RAX/44. Of course I also
worked
on a 360/67 that ran TSS and/or CP/67. Another TSO predicesor on a 360/65
look-alike with a DAT box under the covers. TSS quickly faded away when it
could sustain too many users. Enter OS/MVT with HASP. Its nearest next-of-
kin was a 360/91 running OS/MVT and ASP (yes, it did have that wonderful
space-age Operator Console). Connecting back to older 360s, the 91 used a
360/20 to run card-readers for Programmer Self-Service Input. The real
high-
tech 2540s were for production!

So there's a little bit more of old history. I can talk about one of the
first VM systems, before it had a real (pardon the pun) name and Product
number...

Edward(Ed) J. Finnell,III

unread,
Dec 8, 1997, 3:00:00 AM12/8/97
to

Don't know the complete lineage, but my Cousin(the scientist) works at
NASA and his story was that when NASA brought in ASP it was just way
too big to let them do what they needed to do. So having plenty of
resources they re-engineered it to do what they needed to do. From
the beginning it was called Half-ASP, but when it started to become
something "industry" might need it was sanitized to Houston Automatic
Spooling Program.

EDWARD J. FINNELL,III(EFIN...@UA1VM.UA.EDU)
MVS/Proj. Mgr.
http://www.ua.edu

Metz, Seymour

unread,
Dec 8, 1997, 3:00:00 AM12/8/97
to

Half-ASP was meant as a joke <g>. The name HASP stood for Houston Automatic
Spooling Priority. I know that "Program" would have made more sense, but
it's not part of the name.

Shmuel (Seymour J.) Metz
----------

From: Edward(Ed) J. Finnell,III


To: smetz; IBM-MAIN
Subject: Re: A List of S/360/370/390 Operating Systems?

Date: Monday, December 08, 1997 9:31AM

Bob Halpern

unread,
Dec 8, 1997, 3:00:00 AM12/8/97
to

The 44 had the performance of the 65 for scientific calculations. Our
physics department had one, and wished they didn't. Non mainstream
operations system, and all the tools developed around the campus were
unusable. When they decided to get rid of it, found 0$ value.

TSS was developed for IBM by Computer Sciences Corporation. For many
years after the full TSS users all met around a large table at SHARE.
All 12 of them.

CP67 latest incarnation is VM (large parts in hardware to support PR/SM)
and CMS (was Cambridge Monitor System, now Conversational - The IBM
scientific center in Cambridge MA developed it).

HASP was developed by Crabtree and Simpson for NASA Houston. That
effort followed their work on Direct Couple for the same facility
to drive a 709x with a 704x. HASP is now JES2.

ASP, Attached Support Processor, started its life at UCLA/IBM to
drive the 360/65 and to switch into 7094 emulator mode. It grew
and later became JES3. The groups background was taking the Direct
Couple by Crabtree/Simpson and making it into the commercial product.
So using another machine was a logical step. Originally a 360/40 was
driving the 360/65. Hah!

JES2 was being worked on back east (I believe Bethesda). JES3 was at
the Los Angeles Scientific Center, moved to Thousand Oaks (just
north of LA). From there it was moved back to the same place as JES2.

The JES in VS1 also impacted other JES development. It was planned
to absorb all the VS1 stuff into JES2. It got funny because there
were a couple of unique things that were so unique we could not
find anyone using them. When Crabtree found out he had just gone
thru hell getting a couple of the features implemented he had a tizzy
fit at SHARE. (I was the SHARE manager of the VS1 group for a while).
The rest of the differences were accounted for.

As you may have noticed, there are large changes in JES3 going on. The
SYSPLEX is a good reason to make these changes and to provide a
unified JES. Seems to me it is finally coming.


Metz, Seymour wrote:
>
> The 360/44 was not standard and was not a variant of the 360/40. The 360/40
> had microcode in TROS to simulate the S/360 architecture, while the 360/44
> was hardwired (and missing the SS instructions).
>
> TSS and CP/67 were not TSO predecessor systems; they were unrelated. TSS,
> IMHO, was better than MVS with TSO.
>
> OS/360 with HASP was older than TSS.
>
> ASP ran on machines much less expensive than a 360/91, e.g., 360/65. In
> fact, the ASP support for the 7090 emulator was only relevant to the 360/65.
> Is it true that HASP stands for Half ASP <g, d & r>?
>

> Shmuel (Seymour J.) Metz
> ----------

> From: HollyWiz


> To: smetz; IBM-MAIN
> Subject: Re: A List of S/360/370/390 Operating Systems?

Bob Halpern

unread,
Dec 8, 1997, 3:00:00 AM12/8/97
to

From my other post I gave some information. When IBM announced a
shared file system 7094-1405-1410 for spool handling, a lady at the
Los Angeles Scientific Center (partially housed at UCLA, the rest
in Westwood (adjacent community)) did some modeling. Found that the
1405 was so slow (500pound hydraulic head), things would run worse.

Simpson/Crabtree had gotten a 704x-709x Direct Couple system working
for NASA. It was brought to the UCLA facility for the making of a
commercial product. A CE, Dave Bottles, further modified the
hardware connection so that the transfers could be interrupt driven.
Things went much faster if you used the upgraded IBSYS on both
machines.

The Direct Couple group developed ASP. In early tests they sent a
copy to their fellow IBMers in Houston. It did not get a good
reception.

When HASP "leaked out" all over the place, the conclusion by the
ASP people was that users did not want to get a second machine just
to keep the first busy (necessary for 7094 emulator with 36 bit
registers internal to the emulator). So they designed a "Local ASP",
LASP. It was shot down during design review.

Tom Bishop

unread,
Dec 8, 1997, 3:00:00 AM12/8/97
to

Cute, but having worked on ASP in 1969 and HASP in 1972, I can tell you they
have very different parents. ASP was to be an operator, scheduling and
processing work on other machines. HASP was a spooler, overlapping and
improving the card reader and printer functions for the operators. ASP also
did spooling.

I had heard that the 360/44 was used (built) as the local ASP processor and that
they had HASP running on OS/PCP.

Thomas J. Bishop
Information Technology Agency
City of Los Angeles, CA. 90012, USA
+1 213 847 3920
tbi...@ita.ci.la.ca.us

>>> "Edward(Ed) J. Finnell,III" <EFIN...@UA1VM.UA.EDU> 12/08/97 07:31AM >>>


Don't know the complete lineage, but my Cousin(the scientist) works at
NASA and his story was that when NASA brought in ASP it was just way
too big to let them do what they needed to do. So having plenty of
resources they re-engineered it to do what they needed to do. From
the beginning it was called Half-ASP, but when it started to become
something "industry" might need it was sanitized to Houston Automatic
Spooling Program.

EDWARD J. FINNELL,III(EFIN...@UA1VM.UA.EDU)
MVS/Proj. Mgr.
http://www.ua.edu


!

Reply all
Reply to author
Forward
0 new messages