Archives of the TeradataForum
Message Posted: Fri, 11 Apr 2003 @ 18:04:28 GMT
| Subj: | | Re: How to implement the sequence concept in teradata |
| |
| From: | | Anomy Anom |
<-- Anonymously Posted: Friday, April 11, 2003 12:46 -->
| | Teradata doesn't support Sequence concept. I think by using procedure we can implement. Is it right or wrong. If it is wrong how
to implement the sequence concept in teradata. | |
This seems to be the most popular question on this forum. I have the suspicion that this question mostly comes from people with
experience in other databases who are first time users of Teradata. Is that right? For some reason, they have this need to put sequence
numbers on their data. Why is that? Is this for use as an index?
I only have experience with Teradata and we've never had a need for a sequence number. We can usually use the key fields in the primary
index (sometimes plus and minus other fields as needed), and get a good distribution across the amps. Has anyone with a need for a sequence
number as an index tried doing something like this, before bemoaning the lack of a "sequence concept"? Or am I completely off base
here?
Just couldn't keep quiet any longer,
Long time Teradata enthusiast
| 
