On 04/04/2011 10:31 AM, McKown, John wrote: > We own too much historical data and processing. The cost to convert our VSAM > files to MS SQL files, convert the current CICS transactions and batch > programs to something runnable on Windows, and the cost to convert the batch > processes (JCL) has a return of investment of about 4 to 5 years. NYC > management says that is too long and has a large up-front capital outlay. It > is, short term, cheaper to stay on z/OS. What is happening is our company is > transitioning from a normal Health Insurance company to a supplemental > insurance company (kind of like Aflac, but no duck). As we do this, our > policy base on the mainframe shrinks and the new stuff is all on the > distributed side (lots of apps on agent's laptops, tablets, and smartphones). > So it is simply less expensive to just let the z die a gradual death. I > believe they sincerely want the z dead and gone. But the assassin costs too > much. I don't know about all companies. But around here "long term" seems to > be "next month" <grin>.! S! > eriously, most things around here seem to be on a quarterly basis. I'm > trying to figure out the difference between "quick and agile" and > "indecisive". But the new business is off to a strong start. Not that it will > help the z.
that was the case with our management too ... but it's taken far too long to let it die a gradual death ... for roughly 22 years they've been starving the mainframe (no personnel resources, no budget support, etc.). at every opportunity, a piece of it was moved to the distributed environment. we're close, but still not there ... of the original team of 5 who supported every aspect of the mainframe from OS to ADABAS to CICS to SNA to TCP/IP to DFSMS to AFP to ... I am the last one (and have been for the last few years) ... and I'm eligible for retirement in 3 years. so now they're going to throw even more money at the project to migrate us off of the mainframe ... because it has to be done before I retire (don't they realize that if they got me a new zEnterprise I wouldn't WANT to retire!) ... I think they'd get far more value by investing in the mainframe (upgrade our aging z890, bring on some IFLs, get z/VM, etc.) and spend a few bucks training some young folks to manage it. but that's not their vision. keep trying to point out that a system that can be run by a single guy has to have some advantage over a bunch (about 200) of servers that takes 22 sysadmins and DBAs (and we're hiring 2 more) to manage. and all those servers and ancillary switches, routers, loadbalancers, etc. must cost more than our single z890. and the cooling and power costs of those 200 servers must dwarf the environmentals of our little z890. and what about those per-core Oracle licenses? anyway, my management has decided that it's time to bite the bullet and pay the assassin. assassin says they can get the job done by year-end 2012 ... on the "up" side ... I get to work with VMware clusters when our mainframe goes away ... way cool and fun stuff to play with ... not as cool as z/VM ... but cool enough to keep me happy ... ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
