SQL Sequence use cases

[David Gibbs <david@xxxxxxxxxxxx>]

Folks:

I'm having a bit of a struggle understanding the SQL Sequence business cases.

I understand the conceptual idea behind SQL Sequences ... but the behavior I'm observing in testing is kind of baffling me.

In interactive SQL (no commitment control), I created a SQL sequence in a schema starting at 1 with a cache value of 10.

I then run the following statement:
select (next value for TESTSEQ1) from sysibm/sysdummy1

I get a result of '1'.

If I run the same SQL statement again, I get the result '2'.

However, if I start interactive SQL in another job, and run the same statement I get a value of '11'.

From what I understand about SQL Sequences, you're supposed to use it to get a sequential number from SQL on an select, insert, or update statement.

What I'm not understanding is why the 'sequential' number isn't so sequential ... and very gappy.

I do know that I could use the no cache option to disable caching ... but I'm just trying to understand the business cases for the sequence concept.

Thanks!

david