Archives of the TeradataForum
Message Posted: Tue, 25 Feb 2003 @ 14:02:00 GMT
| Subj: || || Re: Couting duplicate rows thru multi load |
| From: || || Jose Lora |
You could use Fastload + INMOD (that will add a record id). This will give you the fastest loading time. Once in the database, use your
SQL skills to locate / remove / etc. your duplicated records.
Now, there are some ways to avoid the INMOD, one is using an ETL tool that will allow you to read data from a file (or any other source),
do transformations on the fly (calculate the new record id) and send the data directly to a Teradata Utility (streaming data to Fastload).
The idea of course is avoid intermediate staging files outside Teradata and use parallelism to locate the duplicated rows.