Archives of the TeradataForum
Message Posted: Tue, 12 Sep 2000 @ 22:35:14 GMT
| Subj: | | Re: Inserts with Secondary Indexes |
| |
| From: | | Sam Sterling |
Also, it depends on the non-uniqueness of the secondary indices. If you only have a couple of hundred values out of a couple of hundred
million rows, for example, you could have a million rows for each value, and that can take time. Also, if a single value has most of the
occurrences, that causes hot amps and loss of parallelism.
Dropping & recreating can help, but collect statistics and look at the distinct values, then run some queries to determine counts by
value.
Hope this gives you some ideas.
Sam Sterling
Sr. Consultant
www.ssterling.com
Admin Comment: When last tried, the URL in this Post no longer functioned.
| 
