[ih] internet-history Digest, Vol 62, Issue 19

[John Day]

Yes, Retz was at UCSB.

ANTS (I and II) and ELF had the same basic OS architecture.  ANTS was 
written in a high level language and ELF in PDP-11 assembler.  They 
got basically the same throughput, (which was considered inadequate 
at the time).  (300 messages/sec sticks in my mind but don't quote 
me.)

There was a paper at conference about 1975 in which a group at 
Waterloo tried to solve the same problem with the same OS structure 
and got basically the same throughput.  The Waterloo group had no 
contact with the ARPANET efforts, so I concluded that what we were 
seeing were the limits of the architecture, not the programming.

In 1975, we got a copy of Unix at Illinois and by that summer had it 
up on the 'Net.  The NCP was squeezed into the kernel with everything 
else running as a user-app.   This was the first Unix on the 'Net. 
Some pretty ugly things had to be done to get it all to work since 
Unix had such lousy IPC (and mostly still does).  Once it was up on 
the Net, the next thing to do was to build better IPC for it.

But since it had the same architecture as ANTS, ELF and the Waterloo 
system, we expected it would get roughly the same throughput and this 
was indeed the case.  However, Unix was also done in a high-level 
language and had a lot more tools, so it quickly became the system of 
choice for everyone including us.

During 1975-76, our group at Illinois stripped down Unix so it would 
run on an LSI-11 with a plasma screen and a touch panel to create an 
"intelligent terminal."  It was used with a land use management 
database system for the 6 counties around Chicago that used databases 
we had on other systems on both coasts.  While it had a keyboard for 
data entry, one could do most things: querying the database, 
displaying maps with the data, etc. without using the keyboard.  A 
version of this intelligent terminal was also done for the DoD and 
installed on a few bases.

Today's OSs mainly use the same architecture we had for ANTS, ELF, 
and UNIX.  They have solved performance issues by using Moore's Law. 
IOW, they didn't. ;-)

The problem as we saw it was to get speed while maintaining the 
properties of modularity and good system design.  Most of the 
overhead was in context switching and data copying.  It was clear 
that just optimizing code, i.e. do the whole thing in assembler was 
not the answer.

Gary Grossman came up with a brilliant solution (as he usually did) 
that solved all of these problems very elegantly in an executive 
called the Hub and that small enough to be provably secure.  The 
kernel was about 6K bytes including buffers.  A later DoD project 
built a proven secure OS based on the Hub.  The group at Illinois 
wrote the code and SDC did proofs.  The Hub model continues to be 
used widely for embedded applications mostly.  Mike O'Dell wrote a 
paper on it for a European Unix Conference back in 1987.  It is small 
enough and simple enough that there isn't just one, it is just as 
easy to create it from scratch as it is to get a copy and port it. 
;-)

Take care,
John Day


At 10:32 -0700 2012/08/29, Dave Crocker wrote:
>On 8/28/2012 3:48 PM, Elizabeth Feinler wrote:
>>  There are several people looking for things that may be at the
>>  Computer History Museum (CHM) in Mountain View, CA. I have indicated
>>  below what these are:
>>
>>  Jake ==== On May 13, 2012, at 3:11 AM,
>>  internet-history-request at postel.org wrote:
>>
>>>  Send internet-history mailing list submissions to
>>>  internet-history at postel.org
>>>
>>>  To subscribe or unsubscribe via the World Wide Web, visit
>>>  http://mailman.postel.org/mailman/listinfo/internet-history or, via
>>>  email, send a message with subject or body 'help' to
>>>  internet-history-request at postel.org
>>>
>>>  You can reach the person managing the list at
>>>  internet-history-owner at postel.org
>>>
>>>  When replying, please edit your Subject line so it is more
>>>  specific than "Re: Contents of internet-history digest..."
>  >>
>>>
>>>  Today's Topics:
>>>
>>>  1. Re: The UCLA 360/91 on the ARPAnet/Internet (Dave Crocker) 2.
>>>  Re: Historical fiction (Vint Cerf) 3. Re: The UCLA 360/91 on the
>>>  ARPAnet/Internet (John Day) 4. Re: The UCLA 360/91 on the
>>>  ARPAnet/Internet (Sytel) 5. Re: Historical fiction (Noel Chiappa)
>>>  6. Re: Historical fiction (John Day) 7. Re: The UCLA 360/91 on the
>>>  ARPAnet/Internet (dave.walden.family at gmail.com) 8. Re: Historical
>>>  fiction (dave.walden.family at gmail.com) 9. Re: The UCLA 360/91 on
>>>  the ARPAnet/Internet (Vint Cerf)
>>>
>>>
>>>  ----------------------------------------------------------------------
>>>
>>>
>>>
>Message: 1
>>>  Date: Sat, 12 May 2012 17:58:48 -0700 From: Dave Crocker
>>>  <dhc2 at dcrocker.net> Subject: Re: [ih] The UCLA 360/91 on the
>>>  ARPAnet/Internet To: Vint Cerf <vint at google.com> Cc:
>>>  internet-history at postel.org Message-ID:
>>>  <4FAF0748.8080504 at dcrocker.net> Content-Type: text/plain;
>>>  charset=ISO-8859-1; format=flowed
>>>
>>>
>>>
>>>  On 5/12/2012 2:42 PM, Vint Cerf wrote:
>>>>  Steve Crocker led the group that developed the Sigma-7 Experiment
>>>>  Timesharing system. We called it SEX and the most popular
>>>>  document among the geeks was the SEX Users Manual....
>>
>>  I think it was called "The Joy of Sex" and then there was "More Joy"
>>  a parody on a very popular book at the time.  Both are at CHM
>
>I recalled the name SEX User's Manual, not the two-volume bawdy name.  Hmmm.
>
>
>>>
>>>  Cycling back to the topic of Arpa's wanting to share resources,
>>>  when we tried to get funding for 32K more memory for the system,
>>>  Arpa instead said we should get a PDP-11 and use one of the
>>>  terminal concentrator systems (ANTS from Illinois or ELF from Santa
>>>  Barbara) and do remote computing.
>>
>>  I believe that ELF was done by Dave Retz while he was at SRI.
>
>I believe Dave was at UCSB.
>
>
>>  Most DEC PDP-11s used it until it was superseded by UNIX.
>
>There was also the ANTS-I system at Illinois.  People varied on which
>they used as terminal concentrators for net access.
>
>Neither of the next-generation versions of ELF or ANTS were successful,
>however.  And, yes, Unix became the choice.
>
>
>d/
>
>--
>  Dave Crocker
>  Brandenburg InternetWorking
>  bbiw.net