We are at the begining of the project I am working on testing CRC32 and are about to establish the ETL DataStage strategy, we want to use CRC32 in development of Server jobs to detect changes.
I have searched the Forum about CRC32 topic and there were enough information and comments on how it can be used.
I tried to use CRC32 function for the initial load of 100000 rows with seven columns, trimed and checked for null and it produce the same values for two different records. 200K rows get 10 rows with 2 each has the same value.
But If I load less than 100000 rows (about 99000 rows) it produce unique values for every single row. My goal here is to run through at least 20 million records.
I understand there is 1 chance in 4 billions rows the CRC32 to produce the same value, but I do not even pass the 100K mark and a I get a dupe value!
![Confused :?](./images/smilies/icon_confused.gif)
Is there a specific way to use this function??
Does anybody knows a free utility out there that will aid in change data capture process??
Any thoughts and ideas is appreciated.
Thanks . . . . Bryceson