Generation using M is not good for generating massive amount of data

Currently we use a global rowcounter M as the parameter to the generators.

This is affecting massive data generation in a bad way.

Given this example:
Generate daily transactions, all transactions are connected to a "Day"

- Insert Day (using dayly series)
- Insert 100 transactions for that day (using minute series to spread them out 
during the day )

Everything is fine on the first execution.
For the second execution the Day will be generated with DateTime.AddDays(102). 
(The global row counter M is now 102)

Meaning the second iteration will be created 101 days after the first iteration.

Change generation back to N or something else.
N had it's own problems, for every Execution, all the dates where the same.

Original issue reported on code.google.com by a03pe...@gmail.com on 27 Jun 2012 at 3:09

tbayart / sql-server-data-producer

Generation using M is not good for generating massive amount of data #55