Anne & Lynn Wheeler <[email protected]> writes: > "overnight batch window". The problem was that the parallel distribution > object-oriented technologies introduced a factor of 100 times (two > orders of magnitude) increase in processing overhead ... totally > swamping any anticipated throughput benefits of large number of killer > micros.
re: http://www.garlic.com/~lynn/2009h.html#1 z/Journal Does it Again a lot of the "object-oriented" re-engineering failure was whole generation that had no concept of scale-up and speeds&feeds ... being accustomed to dedicated servers that operated at 5-10% avg. utilization. some of the (failed) "straight-through processing" re-engineering efforts didn't even realize the magnitude of problem until they attempted scaled-up deployment. that whole genre has somewhat come home to roost being able to use large number of rack-mounted "blades" (reduced physical footprint) and virtualization, to obtain 10:1 reduction (or more) in number of (physical) servers (relatively easy to leverage virtualization to coalesce 10-20 such servers onto a single machine) ... and for some operations an order of magnitude reduction in number of such datacenters. there is some overlap with these activities and "cload computing" initiatives. -- 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
