I need to regularly check many (100.000's) rows and check if their current state is the same as the latest stored version in another database. Is there a way to get some sort of unique value for a row to match them, or would I have to manually check the rows column by column?
The source database is a SQL Server 2005 database and the table doesn't have a timestamp mechanism for create, update and/or delete action. I've looked around to check if there is row information available but the only thing available is a pseudo column %%lockres%% and the row information, but that doesn't provide date and or time information.
I'm limited in my tools, but I have a webserver running Apache and PHP and direct access to the source and destination databases. I only have read permissions on the source database.
What would be the most efficient way to compare the data and maintain performance on the source database.