|
|
Archives of the TeradataForum
Message Posted: Thu, 17 Nov 2005 @ 10:32:29 GMT
Subj: | | Re: UPDATE vs DELETE-INSERT |
|
From: | | Agarwal, Rishi |
David Micheletto wrote:
| If table1 is the target of many(?) updates, perhaps an insert select like the following might work better? (check the logic out
carefully) | |
> INSERT newtable
> SELECT a.col1, a.col2,
> case when b.col1 is null then a.col3
> else b.col3 end,
> case when b.col1 is null then a.col4
> else b.col4 end
> from table1 a left join table2 b
> on a.col1 = b.col1
> and a.col2 = b.col2
Assuming Table2 is the actual target table and Table1 has incremental data.
The SQL mentioned above will process the new records and load in into the new table. While applying this newtable to the final table Table2, we
need to delete the existing records. Can we avoid DELETE here? Is this DELETE will be faster than UPDATE on Table2?
| |