In a message dated 6/15/2010 11:58:54 A.M. Central Daylight Time, [email protected] writes:
guess the great irony is how will we ever get it right if we scramble the data, when we can't even get it right when it is unscrambled. Of course I suppose, one could always argue that the production data is already scrambled. :-) >> google ===> regression testing of production data It's a rich and well researched science. You also want to test for 'load' both batch and online. Depending on requirements might get by with statistical sampling but back when it counted had to do same amount of transactions within a few percentage points of current to be able to roll into production. ---------------------------------------------------------------------- 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
