IBM-Experience Started Mar 2, 2009 - in process
In 1966, I worked at IBM's Mohansic Systems Laboratory in Yorktown Heights, New York in an IBM software development (360 TSS - Time Sharing System) They had 8 or 10 1401s grinding out their lives printing dump tapes - I presume mostly crash dumps :-(( They went click-bang-click-bang for hours on end. I hated walking near them, thinking what a waste of machinery and time. They should write block tapes and the 1401s should go click-pututututututututututut-click pututut... So I bugged my boss. I said I could put a tapeblock back end on the 360 dump and with no 1401 experience figured I could increase the productivity of both the 360 (Mod 67) and the 1401s by a factor of maybe 5. But he figured I had more important fish to fry, * and certainly the 1401s were cheeeep (and he was not a boat rocker.) Ed (tales from the crypt) Thelen * It is almost embarrasing - (The Mod 67 had a core memory cycle time of 2 microseconds, and the swap "drum" wasn't all that swift either.) It was taking relatively long to get the 360 assembler going on that relatively early paged memory machine. They had trace dumps of the assembler in action. I wrote a Fortran program to input those traces and identify modules that got called in a time related manner. I got these placed together in the assembly .exe to increase the likelihood that satisfying one memory paging fault would load as many useful modules as practical - I got the assembler to start effectively with fewer memory paging faults :-)) then the wife wanted to go back to Minnesota, :-(( (I had been dragging the family all over the country. One kid now calls me an "itinerant programmer").
If you have comments or suggestions, Send e-mail to Ed Thelen