Just remember you asked, Access scripts, specialized backup scripts (especially when flat files and databases must stay synchronized), archive management. When managing several large systems I had an issue with n-way synchronization of particular directories across systems, so that turned into a large synchronization process that was totally automated.
When my employer had a home-brew office system and multiple mainframes over very disparate locations, we had to automatically update mail distribution lists on 'private' file systems, we wrote a single threaded system that would take weeks to run. I took the tools the 'central tools group' provided, and determined how to multi-thread them without interfering with I/O of the sister processes (which was the big bottle neck). It was a iterative process, but either alone or with a helper we could update about 4000 users each over a weekend (where single threaded we could get possibly 1 per hour done). At one place the head admin was demoted after a dumb move on his part. I helped him make lemons out of lemonade. He was assigned to the durge work, I/O balancing and backups. Together we designed a way to fully automate backups and make it easy to monitor. But the big thing we did was to analyze the I/O on the system, and re-balance I/O so the I/O was balanced between I/O processors (channels in mainframe talk), devices hung off of each channel, removing 'daytime I/O' (office hours) from disks used for systems files, but allow them to be used for 'online backups' or images that were seldom accessed, but only do those in 'off prime' time. -- Doing this deferred big systems upgrades by over a year, and we were getting performance out of systems that the vendor kindly said 'you can't do that much with the equipment you have'. Eventually I did have to tell the bosses, that we can always do more with less, until there is no more less we can do with. We finally got all the good out we could without reducing end user function. The good thing is we delayed that news by a couple of years! (Normal upgrade cycle was 9 to 12 months, we deferred to 24 months without measured reductions). There were lots of other things, but these were the big systems. ... Most of this was before we had ethernet and 'networking' was all T1's or better between big systems. "You don't manage people; you manage things. You lead people." - Admiral Grace Hopper, USN "Tell me and I forget. Teach me and I remember. Involve me and I learn." - Ben Franklin _______________________________________________ Tech mailing list Tech@lists.lopsa.org https://lists.lopsa.org/cgi-bin/mailman/listinfo/tech This list provided by the League of Professional System Administrators http://lopsa.org/
