Sujet : Re: The joy of FORTRAN
De : lynn (at) *nospam* garlic.com (Lynn Wheeler)
Groupes : alt.folklore.computers comp.os.linux.miscDate : 02. Oct 2024, 00:30:58
Autres entêtes
Organisation : Wheeler&Wheeler
Message-ID : <871q0zv1sd.fsf@localhost>
References : 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
User-Agent : Gnus/5.13 (Gnus v5.13)
Lynn Wheeler <
lynn@garlic.com> writes:
somewhat like IBM's failed "Future System" effort (replacing all 370s)
http://www.jfsowa.com/computer/memo125.htm
https://people.computing.clemson.edu/~mark/fs.html
https://en.wikipedia.org/wiki/IBM_Future_Systems_project
>
except FS was enormous amounts of microcode. i432 gave talk at early 80s
ACM SIGOPS at asilomar ... one of their issues was really complex stuff
in silicon and nearly every fix required new silicon.
I continued to work on 360/370 all during FS, even periodically
ridiculing what they were doing (which wasn't exactly career enhancing)
... during FS, 370 stuff was being killed off and claims lack of new 370
during FS gave the clone 370 system makers their market foothold. when
FS finally imploded (one of the final nails was analysis that if 370/195
applications were redone for FS machine made out of the fastest hardware
available, they would have throughput of 370/145 ... about 30times
slowdown) there was mad rush to get stuff back into the 370 product
pipelines, including kicking off the Q&D 3033&3081 efforts in parallel.
I have periodically claimed that John did 801/RISC to go the extreme
opposite of Future System (mid-70s there was internal adtech conference
where we presented 370 16-cpu multiprocessor and the 801/RISC group
presented RISC).
I got dragged into helping with a 370 16-cpu multiprocessor and we con
the 3033 processor engineers into working on it in their spare time (a
lot more interesting than remapping 370/168 logic to 20% faster chips).
Everybody thought it was great until somebody tells the head of POK that
it could be decades before the POK favorite son operating system ("MVS")
had effective 16-cpu support (at the time MVS docs had 2-cpu system
support with only 1.2-1.5 times the throughput of a 1-cpu system (I had
number of 2-cpu systems that had twice the throughput of single cpu
system) and head of POK invites some of us to never visit POK again
... and the 3033 processor engineers to keep their heads down and no
distractions. Note: POK doesn't ship a 16-cpu system until after the
turn of the century (more than two decades later).
-- virtualization experience starting Jan1968, online at home since Mar1970