Bri P writes: >I also think that, for something as crucial as DR, >why put yourself through the risk that comes from >doing these swaps?
Yes, as I recall that was a major area of conversation. But that's also why there's a certain school of thought that, at least for some cases, the definition of "production" ought to include flipping workloads back and forth between data centers as mere periodic standard operating procedure. You might call this "nomadic production" -- just moving every so often as a normal state of existence. To use another (bad?) analogy, nomadic existences (migratory birds, human populations, etc.) seem to favor survival. It was pretty easy to figure out that nomadic production is technically feasible with mainframes. (It's just DR flipping by another name, and you can get those all the way down to RTO=RPO=0 with the appropriate design.) Actually the minor mystery was financial. But financially it seems to be at least much less expensive than it was in the past. To pick one budgetary element, if you wanted to do this in the 1990s you had to pay very roughly double for software as full capacity licensing ruled the day. But now if you're on Variable Workload License Charge (VWLC) and schedule the production flips appropriately (perhaps midnight early morning on the 2nd day of the month? :-)), the difference in software cost seems to be something very much like zero, at least for the IBM stuff. That policy is unique to mainframes as far as I know. NOTE: Consult your IBM rep, blah blah. "Tim said it was zero" gets you zero credit. :-) Patrick Falcone writes: >A recent interview I had with a large financial institution, >east coast, came from the fact that they were going to a >GDPS, *geographically dispersed personnel support* structure. >The support team was somewhat duplicated at a different site. >located about 300 hundred miles away. The people are always the weak link. I think most everybody here is familiar with the fact that you have to plan DR as if the IT people didn't exist -- or at least most of them didn't exist. Regardless of what you plan, when push comes to shove employees are going to want to save their spouses, parents, children, grandchildren, and other relatives before they want to bother saving any business. And the cases where business interests and family interests precisely align are rare. And employees may not survive, may not get to the office, may not connect, etc. Best case you'll keep the employees who are currently in the office on shift, and even then it might be a good idea to have some orphans, only children, widows, and widowers on staff. :-) It's also very, very difficult to rehearse staffing outcomes. So things need to be automatic and automated. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] ---------------------------------------------------------------------- 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
