Re: SQL Developer - Selects
- From: Thomas Kellerer <YQDHXVLMUBXG@xxxxxxxxxxxxx>
- Date: Wed, 31 Oct 2007 10:38:10 +0100
markus.schreyer@xxxxxxxxx, 31.10.2007 10:20:
When I open a datatable (tab data), SQLD does a coresponding select on
the table. Looking at enterprise manager i see that the select
consists out of ORA_ROWSCN and all the columns from the table. In a
table with million records i get an instant result cause only 55 rows
in my case are fetched. When i take the same select and execute it,
it would take much longer cause i get the whole table. How is it
possible to limit the number of records in the result with such a
select?
SQL Developer uses JDBC. JDBC offers to set the maximum number of rows returned by the query, and I assume that SQL Developer is using that feature - especially as you cannot see any limits when you trace the statement.
I did a quick test with a different (Java/JDBC based) query tool, where I know it uses setMaxRows() and the behavior there is the same as with SQL Developer. A select * from big_table (5 Million rows) executes nearly immediately when max rows is set to 50.
My assumption is that the JDBC driver only sends the defined limit to the server (along with the query of course), and the server is able to take this into account when returning the result. The limit does not happen on the client side as far as I know.
Thomas
.
- References:
- SQL Developer - Selects
- From: markus.schreyer@xxxxxxxxx
- SQL Developer - Selects
- Prev by Date: SQL Developer - Selects
- Next by Date: Re: CBO rows estimate way out in 2 table join - what to do?
- Previous by thread: SQL Developer - Selects
- Index(es):
Relevant Pages
|
