[ih] Interprocess Communication

[Jack Haverty]

I have a vague memory that the IBM 360 architecture circa later 60s 
involved the mainframe CPU and processors in peripheral equipment like 
disk controllers, communicating over some kind of bus.  So even a single 
mainframe was a multiprocessor.

Also my recollection is that some form of message-oriented interaction 
protocol was used to manage disk I/O, since I/O operations often didn't 
complete in the same order as they were initiated.   But I never did any 
OS programming on the 360 so I don't remember any details.   Maybe that 
was QTAM.

I've always wondered how much of IBM's techniques influenced the early 
ARPANET mechanisms, e.g., the use of DO/DONT/WILL/WONT negotiations.

Jack

On 5/12/22 14:07, Brian E Carpenter wrote:
> On 13-May-22 06:58, Jack Haverty via Internet-history wrote:
>> MQTT has been around for a while.  I think it was originated at IBM.
>
> MQ Series originated at IBM Hursley in the UK in 1993. If this list 
> allowed
> attachments, I'd send a photo of my $1,000,000,000 mug - every employee
> at Hursley received one when MQ Series total revenue hit $1B in about
> 1998. See https://en.wikipedia.org/wiki/IBM_MQ
>
> I am no expert, but the article traces the concept back to "BTAM and QTAM
> (Basic and Queued Telecommunications Access Methods)" on OS/360 in 1964.
> RFC 62 itself cites references to formative work on multiprocessing 
> operating
> systems. The issues arose in multiprocessors before they arose in 
> networks.
>
> The formal equivalence between message-oriented systems and 
> procedure-oriented
> systems wasn't noted until 1978 by Lauer and Needham [1], but it was 
> there all
> along. There was a lot of Zeitgeist in those days. There still is.
>
>    Brian
>
> [1] Lauer, H.C., Needham, R.M., "On the Duality of Operating Systems 
> Structures,"
> in Proc. Second International Symposium on Operating Systems, IRIA, 
> Oct. 1978,
> reprinted in Operating Systems Review, 13,2 April 1979, pp. 3-19.
>
>> It's now widely used for communications between processes in various
>> "Internet Of Things" scenarios.   E.g., my model railroad uses it, and
>> my home automation system as well.   It is message-based rather than
>> using a connection or circuit, can use formats such as JSON to organize
>> the data, and runs over the Internet.
>>
>> See https://mqtt.org/
>>
>> Jack
>>
>> On 5/12/22 11:45, Alex McKenzie via Internet-history wrote:
>>> In August 1970, Dave Walden issued RFC #62, describing a system for 
>>> interprocess communication that did not include the concept of 
>>> "connections" or "circuits". (I believe it was also published in the 
>>> Communications of the ACM.) It was a generalization of interprocess 
>>> communication within a single computer.  I believe that at the time 
>>> his proposal was too radical for the ARPAnet Network Working Group 
>>> to consider seriously, and so far as I know it has never been 
>>> implemented.  That is my question: has that concept, or something 
>>> close, been implemented in the Internet or elsewhere?
>>> Thanks,Alex McKenzie
>>
>