[email protected] (Mike Schwab) writes: > And the web site is the server and the web page is the application. > > The cloud just reassigns different servers to serve the web page > (application).
re: http://www.garlic.com/~lynn/2017.html#21 History of Mainframe Cloud http://www.garlic.com/~lynn/2017.html#22 History of Mainframe Cloud note, I've previously referenced, large public clouds have also done enormous work on the efficiency and economics so that they can serve elastic on-demand, possibly having 80%-90% idle (zero electricity & cooling, instant on) ... just so they can have capacity for peak demand. we were doing commercial/DBMS and sicentific/technical cluster scaleup as part of our (ibm) HA/CMP product. reference to Jan1992 meeting in Ellison's conference room on commercial/DBMS cluster scaleup http://www.garlic.com/~lynn/95.html#13 within a few weeks, cluster scaleup was transferred, announced as supercomputer for technical/scientific *ONLY* and we were told we couldn't work on anything with more than four processors. some old email (mainframe DB2 group was also complaining that if I was allowed to go ahead, it would be at least 5yrs ahead of them) http://www.garlic.com/~lynn/lhwemail.html#medusa and posts http://www.garlic.com/~lynn/subtopic.html#hacmp ... which contributes to our decision to leave. Later, two of the other people in the Ellison meeting have also left and are at a small client/server startup responsible for something called "commerce server". We are brought in because they want to do payment transactions on the server. The startup had also invented this technology they called "SSL" they wanted to use; the result is now frequently called "electronic commerce". At the time, the payment industry was primarily circuit-based with end-to-end diagnostics and trouble desk was expected to do 1st level problem determination within 5mins. Early "electronic commerce" pilot was large sporting goods vendor that advertised during NFL half-time. An early trouble call was closed as "NTF" (no trouble found) after 3hrs of investigation. I had to document a lot of processes and craft software to try and bring a packet-oriented infrastructure up to the level of what the payment industry expected from circuit-based infrastructure (I also had complete authority over the server to payment network gateway operation ... but could only make recommendations over the client/server operation ... some of which were almost immediately violated and continue to account for some number of exploits). A couple of other observations 1) the platforms used for servers had come from an interactive environments ... which grewup assuming everything was done interacting with user ... and problems just frequently "punted" back to the user. Mainframes tended to assume there was no responsible human interaction and for decades developed hueristics to (automagically) try and handle issues (at the time, much better suited for dark room server operation). 2) I would claim that it takes 4-10 times the effort to take a well designed, implemented and tested application and turn it into a service (in part develop adhoc hueristics to handle lots of situations). In this period there was lots of work on object-oriented operating systems ... for instance Apple had "pink" and Sun had "spring" (at one point I was asked if I would consider heading up an effort to turn "spring" out as commercial product). "pink" somewhat morphs into application development environment which Apple spins off as Taligent. I had one week JAD with Taligent to look at what it would take to enhance their application development environment for services ... after a week the estimate was it would be 1/3rd hit to all their existing object Frameworks and a couple new Frameworks (never went further than that). trivia: 17Feb1992 press about scientific & technical *ONLY* http://www.garlic.com/~lynn/2001n.html#6000clusters1 and then 11May1992 press cought by surprise in cluster scaleup http://www.garlic.com/~lynn/2001n.html#6000clusters2 we did eventually find a object-oriented application development company that did fixed-priced ommercial services ... it was founded by the person that formally was head of the 60s IBM FAA air traffic control project. -- 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
