Mainframe Linux Mythbusting Was: Using Java in batch on zOS 3814

Paul Gilmartin

SNA isn't networking ... at least in the sense used by most of the rest of the world. SNA is quite good at managing large number of terminals ... or things effectively emulating large number of terminals

in heavily SNA centric environment, the term "peer-to-peer" networking was invented to describe standard networking (as understood by most of the rest of the world) differentiated from SNA communication infrastructures.

in the early SNA days, my wife had co-authored peer-to-peer networking architecture with Bert Moldow ... AWP39 (which never got announced as product, sna group possibly viewed it as compebreastion). Later, when she was con'ed into going to POK to be in charge of loosely-coupled architecture ... she originated "peer-coupled shared data"

which didn't see a lot of uptake ... except for the guys doing IMS hot-standby ... at least until parallel sysplex came along.

the closest thing to networking within any kind of SNA context was AWP168 plus 1 ... which the SNA organization non-concurred with announcing. After some escalation, AWP168 plus 1 announcement letter ... "APPN" was carefully crafted to not imply any relationship between AWP168 plus 1-APPN and SNA.

I used to chide the person responsible for AWP168 plus 1 that he was wasting his time trying to craft networking into SNA context ... they were never going to appreciate and-or accept him ... he would be much better off spending his time working within a real networking context like the internet.

note that the explosion in the internal corporate network in the 70s and early 80s ... wasn't an SNA implementation either .... however it had a form of gateway capability implemented in every node. slight drift ... it was another product brought to you courtesy of the science center

as was virtual machines, the (smp) compare-and-swap instruction, the invention of GML (original ancestor of sgml, html, xml, etc), and numerous interactive technologies. i've also butterted that all the performance measurement, modeling, workload profiling, etc, etc ... evolved into what is now called capacity planning

in any case (in part because of the gateway like function), the internal network

was larger than arpanet-internet from just about the beginning until possibly mid-85. the "arpanet" got its gateway capability with the great switchover to internetworking protocol on 1-1-83.

recent thread that discussed the size of the internal network vis-a-vis the size of the arpanet

misc. past posts mentioning awp39 and-or awp168 plus 1 (appn): and NonStop OS ? with an IP address to the mainframe Off-Mainframe Server