The following message is a courtesy copy of an article that has been posted to bit.listserv.ibm-main as well.
re: http://www.garlic.com/~lynn/2009g.html#66 Mainframe articles http://www.garlic.com/~lynn/2009g.html#67 Mainframe articles http://www.garlic.com/~lynn/2009g.html#68 IT Infrastructure Slideshow: The IBM Mainframe: 50 Years of Big Iron Innovation the pre-occupation with future system (which was going to replace all 370 ... in much the way 360 replaced all the stuff before it) ... resulted in the 370 software & hardware pipeline to drain. http://www.garlic.com/~lynn/submain.html#futuresys when future system was killed, there was mad rush to get stuff back in 370 product pipeline ... and basically a 308x & 370-xa effort was kicked off (expected to take 6-8 yrs) ... in parallel with crash 303x, Q&D stop-gap effort until 308x. 303x channel director was basically 158-3 processor engine with just the integrated channel microcode and the 370 microcode removed 3031 was 158-3 with the integrated channel microcode removed (only 370 microcode) and reconfigured to work with 303x channel director (i.e. 158-3 bascially multiplexed integrated channel microcode on 370 microcode on single engine, 3031 had two processor engines, one dedicated to integrated channel microcode and one dedicated to 370 microcode) http://www-03.ibm.com/ibm/history/exhibits/mainframe/mainframe_2423PH3031.html 3032 was 168-3 reconfigured to work with 303x channel director(s) http://www-03.ibm.com/ibm/history/exhibits/mainframe/mainframe_2423PH3032.html 3033 started out as 168-3 wiring diagram mapped to faster chip technology ... originally only going to abe 20% faster than 168-3. the chips were 20% faster ... the chips also had about ten times the circuits per chip ... but using the 168-3 wiring diagrams would have left all the additional circuits unused. during the 3033 development, there were some critical path redesign that took advantage of the higher onchip density resulting in 3033 being closer to 50% faster than 168-3. http://www-03.ibm.com/ibm/history/exhibits/3033/3033_album.html as soon as the 3033 was out the door ... that group started on 3090 (overlapped with 3081 activity). http://www-03.ibm.com/ibm/history/exhibits/mainframe/mainframe_2423PH3090.html initial 3081 ... was 3081D where each processor was about five mips ... not a whole lot faster than 3033 two-processor. fairly quickly after that, 3081K shipped with each processor about seven mips (14mips aggregate). 3083 was bascially single 3081k processor with x-cache slowdown removed so it ran about 15% faster or approx. 8mips http://www-03.ibm.com/ibm/history/exhibits/mainframe/mainframe_2423PH3083.html -- 40+yrs virtualization experience (since Jan68), online at home since Mar1970 ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
