[email protected] (Tony Harminc) writes: > In my circles the term "core" survived for quite a long time after the > introduction of the first 370 models with non magnetic-core storage > (the 158 and 168, followed closely by the lower end 138, 148 and so > on). And amusingly the UNIX people still use "core" in some contexts, > much as they use "print" in a way long dissociated from ink on paper.
folklore I was told was IBM's move to the term "virtual storage" was because of patent on "virtual memory". there was possibly ancillary reason ... supposedly tss/360 was move to "single level store" .... programmer/application not seeing difference between memory access and file accesss ... all appeared the same. in the early 70s, the future system effort ... which was going to totally replace 360/370 ... was heavily "single level store" and overlapped os/360 movement to "virtual memory". however, part of tss/360 was very poor optimization to access to data on disk in its "single level store". I took a lot of what I saw tss/360 did wrong when I did my cms paged-mapped filesystem on cp67 at the scientific center in the early 70s. The FS effort wasn't any better than what tss/360 had been doing ... and one of the reasons I would periodically ridicule what they had been doing (which probably wasn't the most career enhancing activity). In any case, that was just one of the things that contributed to the failure of FS ... misc. past posts http://www.garlic.com/~lynn/submain.html#futuresys I suspect that over the years a lot of the stuff I was doing internally would ship in products ... the reputation that page-mapped filesystem got from the tss/360 and Future System implementations contributed to not shipping my cms paged mapped filesystem (although i could show factor of three times throughput improvement for moderately filesystem intensive applications). http://www.garlic.com/~lynn/submain.html#mmap -- 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
