 |
 |
Archives of the TeradataForum
Message Posted: Tue, 26 Apr 2005 @ 10:42:35 GMT
| Subj: | | Re: Masking identification numbers |
| |
| From: | | Victor Sokovin |
| | How about calculating a hash value? "select hashrow(creditcardnumber)" or a UDF for V2R5.1, you can download MD5 UDF at
www.teradata.com
. But maybe two different numbers result in the same hash value... | |
I for one would not use the above methods literally. One thing about encryption is that we don't want anybody to easily crack it. The
disadvantage of HASHROW and the MD5 UDF (as currently available from the NCR site) is that they don't have any unknown parameters (secret
encryption keys), so it is only a matter of running them on all possible known parameters until right value is hit. OK, with credit card numbers
it's a long list to run through but with a bit of patience it is still an easy job for a home PC. Imagine the UDF would have a couple of unknown
parameters of length 124 or so which would be creatively incorporated into the algorithm. That would surely make the hackers' task less
attractive.
My suggestion for the OP would be to search for the professional encryption method outside of the standard Teradata tools. Teradata as such has
nothing to do with what you are looking for because NCR do not specialize in this type of products. Once the encryption method has been chosen,
you can implement it on TD.
Regards,
Victor
| |
