IBM 9020D? 
Author Message
 IBM 9020D?

I read in the paper today that Chicago airport got stuck because
of a failure in an IBM 9020D mainframe doing air traffic control.
They said that this machine dated from the 1960s.  Is this just
a misprint, and if not, does anyone know more about then and why
they are still being used?


Mon, 12 Jan 1998 03:00:00 GMT  
 IBM 9020D?
The 9020 series of computers were a unique application of System/360
architecture.  I worked on a project that replaced the 9020's in 1983
(don't hold me to the exact dates, it was too long ago ).  It
was called the HOST MODERNIZATION PROJECT.  But that is another story...

The 9020D was, I believe, multiple 360 Model 50's loosely coupled.  The
9020E was 360/65's.  One of the interesting facets of the design was that
several processors were delegated as I/O processors.  You could start an
I/O and store a program in the I/O processor that would be executed when
the I/O interrupted.

I find it incredible that any of these dinosaurs could still exist
anywhere except in a museum setting.  Considering that Chicago is one of
the busiest airports in the nation, I would be willing to bet that they
have the latest and greatest the FAA has to offer.  Therefore, I don't
KNOW but I would guess that the author of the article used some out of
date reference material....

 A.


: of a failure in an IBM 9020D mainframe doing air traffic control.
: They said that this machine dated from the 1960s.  Is this just
: a misprint, and if not, does anyone know more about then and why
: they are still being used?
--

Running MVS, VM and VSE on a Pentium 90 under Linux and Unixware!



Mon, 12 Jan 1998 03:00:00 GMT  
 IBM 9020D?

Quote:
>I find it incredible that any of these dinosaurs could still exist
>anywhere except in a museum setting.  Considering that Chicago is one of
>the busiest airports in the nation, I would be willing to bet that they
>have the latest and greatest the FAA has to offer.  Therefore, I don't
>KNOW but I would guess that the author of the article used some out of
>date reference material....

When I toured the regional air traffic control center in Indianapolis during
the mid 70s, they still used 360 machines, and had even taken to{*filter*}
memory frames from the ceiling due to cable length restrictions. When asked
why 370 systems were not in use, the reply was "If what we have does what we
want it to do, why do we need to change to newer, untested equipment?"

The air traffic control centers are very conservative when it comes to
trying something new. It makes sense, since so many lives depend on the
results. It's not inconceivable that thy would still be running old
equipment, right up to the time that they couldn't find any more parts for
them.



Tue, 13 Jan 1998 03:00:00 GMT  
 IBM 9020D?

Quote:

>The air traffic control centers are very conservative when it comes to
>trying something new. It makes sense, since so many lives depend on the
>results. It's not inconceivable that thy would still be running old
>equipment, right up to the time that they couldn't find any more parts for
>them.

Yet the System/370 machines were an order of magnitude more reliable than
the System/360s. Upgrading to the newer technology might actually save
lives, if one of the 360s crashes due to a memory error or other problem.

+--------------------------------------------+
|              Michael Quinlan               |

|       http://www.primenet.com/~mikeq       |
+--------------------------------------------+



Wed, 14 Jan 1998 03:00:00 GMT  
 IBM 9020D?
WOW, I guess I'd better clarify what the HOST MODERNIZATION PROJECT was
<supposed> to accomplish.  If there really are 9020's in use then I'll be
thinking twice about flying anywhere.  

I worked on what was called the design competition for the REHOST project.
We were to take the existing code and 'rehost' it on current systems.  I
seem to remember that there were about 2.2 million lines of Assembler and
Jovial source code.

There were two design groups, one consisting of Sperry, Amdahl, and TRW.  (I
was a consultant to TRW).  The other group was IBM.  Each group was given a
LARGE amount of money.  The rumor mill at TRW had the figure at 300 million for
each design group!  It was later INCREASED and the rumor mill pegged that
figure at a total of 360 million (720 million dollars for a design
competition!).  I can't speak for the accuracy of the numbers, but I do
know that money never seemed to be a problem and that while doing some cost
estimates for TRW, I was told to use a factor of $15.00 per line of
code.  Which seems a wee bit on the high side to me.  They built up an
entire datacenter in Santa Clara (It's the building across the street
from Fry's.  I think it's now called 'Weird Stuff'.)

The reason for the modernization, was that the number of flights was
increasing to a point where the existing systems could not process the
data quickly enough.  The Sperry-Amdahl-TRW (SAT) team was using a pair
of 5860's (if memory serves).  To make a long story short, IBM won the
competition (due in part to some crafty politics down in Washington DC).
I was under the impression that the first phase of the project, the
REHOST project, was complete.  If there still are 9020's running
ANYWHERE, then this is not true.  Which brings me back to my opening
statement.  WOW, what a boon-doggle!

--

Running MVS, VM and VSE on a Pentium 90 under Linux and Unixware!



Thu, 15 Jan 1998 03:00:00 GMT  
 IBM 9020D?

Quote:

>WOW, I guess I'd better clarify what the HOST MODERNIZATION PROJECT was
><supposed> to accomplish.  If there really are 9020's in use then I'll be
>thinking twice about flying anywhere.  

You need to go to the nearest telephone book and look up Amtrack's number.

Five ATRCC facilities are still running 9020E systems, including Chicago
and Washington.

This morning's Washington Post had an article on the systems, but pointed
out that the physical plant at some of the Centers isn't in good shape
either.  At ZDC the controller who was running the sector that includes
Camp David took an unexpected bath last January when a pipe broke over
the console; the HVAC is so poor that one end of the room is cold while
the other end is hot; there are masks handy for times when asbestos
might be in the air, etc...

Joe Morris / MITRE



Fri, 16 Jan 1998 03:00:00 GMT  
 IBM 9020D?
Joe, you are absolutely correct!

I spoke this morning with a friend who is STILL working (contracting) the
FAA modernization project after 12 years!  Not a bad contract!  He
asserts that Chicago was scheduled to have the Display Channel hardware
upgraded in 1993.  Apparently the portion of the contract that I worked
on was completed (the Central processors were upgraded). The Display
Channel consists of redundant I/O Compute Elements and some channel elements.
According to my (un-named) source, Chicago is up on a single IOCE and CE.
(I am not sure that my CE term is the correct terminology, but my feeble
brain couldn't remember what it was really called...  A thousand pardons).
Should Chicago lose the remaining Display Channel, I think they can still
do mission critical stuff on something called DARC (what an ominous acronym),
Direct Access Radar Channel (I think).   If it come to this, I would expect
to see air traffic in the Chicago sector impacted...

Oh, the reason that Chicago wasn't upgraded?  IBM was behind on the
follow on project (worth 1.2 Billion) and decided they didn't have the
man power for a project that was only $40 million.  Gee, I wonder if they
gave any of the money back to the FAA?

Thanks to Mr. Bonomi ( I hope I spelt that right) for bringing all of this
up.  It gave me an excuse to call my friend and rehash old times!

Regards,

Terry



: >WOW, I guess I'd better clarify what the HOST MODERNIZATION PROJECT was
: ><supposed> to accomplish.  If there really are 9020's in use then I'll be
: >thinking twice about flying anywhere.  

: You need to go to the nearest telephone book and look up Amtrack's number.

: Five ATRCC facilities are still running 9020E systems, including Chicago
: and Washington.

: This morning's Washington Post had an article on the systems, but pointed
: out that the physical plant at some of the Centers isn't in good shape
: either.  At ZDC the controller who was running the sector that includes
: Camp David took an unexpected bath last January when a pipe broke over
: the console; the HVAC is so poor that one end of the room is cold while
: the other end is hot; there are masks handy for times when asbestos
: might be in the air, etc...

: Joe Morris / MITRE
--

Running MVS, VM and VSE on a Pentium 90 under Linux and Unixware!



Sat, 17 Jan 1998 03:00:00 GMT  
 IBM 9020D?
So Terry, what's the deal?  Is ATC still running 9020s?  If Amtrack is also
still on 9020s, will you avoid both air and rail?

Is this an example of beltway banditry?



Sun, 18 Jan 1998 03:00:00 GMT  
 
 [ 8 post ] 

 Relevant Pages 

1. DALLAS Application Developer - IBM Visual Age/SmallTalk, Informix, Perl and IBM AIX

2. Looking for IBM VisualAge Java/IBM VisualAge Smalltalk teachers

3. ISE/IBM Announcement for ISE Eiffel on IBM Mainframe

4. IBM Mainframe to IBM Unix Cobol conversion?

5. Your own IBM AS400 9406-300 with IBM 4234 Printer Cheap

6. Withdrawal notice of IBM Ada products Withdrawal notice and transfer of IBM Ada products

7. IBM Professional/Microsoft/Ryan-MacFarland FORTRAN for IBM/PC

8. APL glyphs, Yterm's emulation and IBM's APL2 . IBM's TryAPL2

9. reposted from ibm-main Re: any problem using QSAM locate mode with IBM Hardware Compression?

10. comp.sys.ibm.pc, alt.msdos.programmer, comp.os.msdos.programmer, comp.sys.ibm.pc.misc, news.announce.newgroups, news.groups

11. Freebie IBM Dead-Tree Zine:IBM COBOL and PL/I Newsletter News /Updates from Santa Teresa Laboratory

12. IBM withdraws (from marketing) IBM COBOL for OS/390 & VM

 

 
Powered by phpBB® Forum Software