🚀 go-pugleaf

found on facebook, 3390 teardown
https://www.youtube.com/watch?v=CBjoWMA5d84&feature=share

3390
https://www-03.ibm.com/ibm/history/exhibits/storage/storage_3390.html

3380->3390 3mbytes/sec to 4.2mbytes/sec higher bit density also smaller
disk can spin faster, 3390mod1, 3.78gbytes to 11.35gbytes, mod2 double
capacity

note: original 3380 had 20 track spacings between datatracks, then
spacing was cut in half and doubled number of tracks, for "K" spacing
was cut to 1/3rd original 3380 (and tripled number of tracks).
https://www-03.ibm.com/ibm/history/exhibits/storage/storage_3380c.html

trivia: father of 801/risc ropes me in helping him with idea for
"wide-head" ... read/writes 16 (closely-spaced) tracks
simultaneously. Every 17th track is servo track ... and wide-head
follows servo-tracks on both sides of the 16 data tracks (18 tracks
total). The problem is that it worked out to 16 times data transfer of
normal 3380 (i.e. 48mbyte/sec) ... which IBM channels couldn't handle
... even the later ESCON only handled 17mbytes/sec.

1980, I was roped into do channel-extender support for STL. They were
bursting at the seams on moving 300 people from the IMS group to offsite
bldg with dataprocessing support back into the STL datacenter. The group
had tried "remote" 3270 terminal support but found the human factors
totally unacceptable. Channel-extender support allowed placing channel
attached 3270 controllers at the offsite bldg. with no apparent
different in human factors (compared to inside STL). The hardware vendor
then tried to IBM to approve releasing my support, but there was group
in POK playing with some serial stuff, that got it vetoed (they were
afraid that if my support was in the market, it would make it harder to
get their stuff approved).

In 1988, I was asked to help LLNL standardize some serial stuff they
were playing with, which quickly becomes fibre channel standard
(including some stuff I had done in 1980). The POK people finally get
their stuff released in 1990 as ESCON when it is already obsolete. ESCON
is 200mbits/sec half-duplex ... only around 17mbytes/sec effective. FCS
is dual-simplex, concurrent 1gbit/sec in both directions, 2gbit/sec
aggregate (around 200mbytes/sec aggregate).

channel-extender posts
http://www.garlic.com/~lynn/submisc.html#channel.extender

Later some of POK channel engineers get involved in fibre channel
standard and define a heavy-weight protocol that drastically cuts the
native throughput ... which is eventually released as FICON. Latest
published stats is peak I/O throughput on z196 that gets 2M IOPS using
104 FICON (running over 104 fibre channel). At the same time a fibre
channel was announced for E5-2600 blade claiming over million IOPS
... two such fibre channel getting more throughput than 104 FICON
(running over 104 fibre channel)

FICON posts
http://www.garlic.com/~lynn/submisc.html#ficon


-- 
virtualization experience starting Jan1968, online at home since Mar1970