re: http://www.garlic.com/~lynn/2014d.html#16 [OT ] Mainframe memories http://www.garlic.com/~lynn/2014d.html#22 [OT ] Mainframe memories http://www.garlic.com/~lynn/2014d.html#23 [OT ] Mainframe memorie http://www.garlic.com/~lynn/2014d.html#25 [OT ] Mainframe memorie
univ ran fortran student jobs on 709 ibsys tape-to-tape in around second per job ... tapes were moved between 709 drive and 1401 drive where the 1401 handled input/output unit record. in transition from 709 to 360/67 & tss/360 ... it got a 360/30 to replace the 1401 ... could run in 1401 hardare emulation ... but also be used to get some familiarity with 360. as noted, tss/360 never got to production level, so when 360/67 came in, it ran mostly as 360/65 with os/360. initially student fortran ran over a minute elapsed time (almost 100 times longer than on 709). getting hasp cut the elapsed time in half to little over 30secs. I then did a lot of work on os/360 starting with mft11 to carefully order the placement of files and pds members to optimize PDS member search and arm seek distance ... which got almost additional three times improvement in elapsed time. This is old post with part of presentation I gave at fall1968 share meeting ... mostly about major rewrite I did for cp67 kernel and associated pathlength and throughput improvements ... but also discusses a little about the very careful rework of stage2 sysgen to optimize disk throughput. http://www.garlic.com/~lynn/94.html#18 CP/67 & OS MFT14 http://www.garlic.com/~lynn/94.html#20 CP/67 & OS MFT14 -- virtualization experience starting Jan1968, online at home since Mar1970 ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
