The following message is a courtesy copy of an article that has been posted to bit.listserv.ibm-main,alt.folklore.computers as well.
Anne & Lynn Wheeler <[email protected]> writes: > we claim that the payment gateway was the original SOA ... some posts > http://www.garlic.com/~lynn/subnetwork.html#gateway re: http://www.garlic.com/~lynn/2009s.html#7 Union Pacific Railroad ditches its mainframe for SOA the gateway started out as rs/6000 ha/cmp configuration with multiple (diverse) internet connections from different ISPs ... and some number of misc. boxes around the perimeter serving various integrity and security purposes. i had originally started out planning on advertising alternate routes on the internet backbone ... but during the payment gateway they announced that the internet backbone was converting to hierarchical routing. As a result I had to fall back to multiple-A record ... for alternate path. sort of standard SLA for high-volume merchant is trouble desk, 5-minute first-level problem determination, ... very early pilot had trouble call that was closed as NTF after 3hrs. I specified recovery and diagnostic critera that had to go into webserver talking to the payment gateway (something like done previously for mainframe stuff as well as ha/cmp) ... inventing a bunch of compensating procedures and writing a trouble shooting guide. I put together matrix of 20-30 failure modes and 5-6 states and the webserver/payment gateway interaction had to demonstrate recovery &/or diagnoses for all possible conditions ... as part of my final signoff. one of the issues was that I didn't have final signoff was the browser/client code. Early major commerce server was sports product that did advertising on sunday football games. got them to put in multiple ISP connections ... but one of their ISPs had regularly scheduled maintenance all day sunday on rotating cities across the country. there was some guarantee that whole class of users wouldn't be able to reach the website during half-time (anticipated high traffic) for at least one sunday. browser people said that the client multiple-A record support was too complicated (i.e. wasn't part of college example programs) ... even after I provided them with example client multiple-A record support from tahoe 4.3 distribution. It took another year to get multiple-A record support into their client. now tcp/ip is the technology basis for the modern internet ... but nsfnet backbone can be considered the operational basis for the modern internet and cix the business basis for the modern internet ... some old email regarding nsfnet backbone activity http://www.garlic.com/~lynn/lhwemail.html#nsfnet and past posts mentioning nsfnet backbone http://www.garlic.com/~lynn/subnetwork.html#nsfnet -- 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
