Tamar,
We were working on a project for Microsoft and it needed to have extremely high performance. We needed to make sure the site would still hit the target after it had been running for a month so we needed a valid spread of data values to make sure we had the SQL database optimized and non-realistic data would skew the test results.
>I haven't worried too much about statistical validity, but actually the model I'm using could handle it without much tweaking. My goal is to create realistic volumes of realistic data, instead of always testing with three recods that contain John Smith of 123 N. Main St. <g>