Re: effizient query with jdbc - Mailing list pgsql-performance

From Steve Peterson
Subject Re: effizient query with jdbc
Date
Msg-id 6.2.3.4.0.20051222090735.05452aa8@localhost
Whole thread Raw
In response to Re: effizient query with jdbc  (Dave Cramer <pg@fastcrypt.com>)
List pgsql-performance
Is there a reason you can't rewrite your SELECT like:

SELECT UUID FROM MDM.KEYWORDS_INFO WHERE KEYWORDS_ID IN (a, b, c, d)

Even doing them 100 at a time will make a big difference;  you should
put as many in the list as pgsql supports.  I'm assuming that there's
an index over KEYWORDS_ID.

Retrieving 10000 rows with 10000 statements is generally a Bad Idea.

S

At 08:17 AM 12/22/2005, Dave Cramer wrote:
>The problem is you are getting the entire list back at once.
>
>You may want to try using a cursor.
>
>Dave
>On 15-Dec-05, at 9:44 AM, johannesbuehler@oderbruecke.de wrote:
>
>>Hi,
>>I have a java.util.List of values (10000) which i wanted to use for
>>a query in the where clause of an simple select statement.
>>iterating over the list and and use an prepared Statement is quite
>>slow. Is there a more efficient way to execute such a query.
>>
>>Thanks for any help.
>>Johannes
>>.....
>>List ids = new ArrayList();
>>
>>.... List is filled with 10000 values ...
>>
>>List uuids = new ArrayList();
>>PreparedStatement pstat = db.prepareStatement("SELECT UUID FROM
>>MDM.KEYWORDS_INFO WHERE KEYWORDS_ID = ?");
>>for (Iterator iter = ids.iterator(); iter.hasNext();) {
>>String id = (String) iter.next();
>>pstat.setString(1, id);
>>rs = pstat.executeQuery();
>>if (rs.next()) {
>>uuids.add(rs.getString(1));
>>}
>>rs.close();
>>}
>>...
>>
>>
>>
>>
>>
>>
>>---------------------------(end of
>>broadcast)---------------------------
>>TIP 6: explain analyze is your friend
>
>
>---------------------------(end of broadcast)---------------------------
>TIP 3: Have you checked our extensive FAQ?
>
>               http://www.postgresql.org/docs/faq



pgsql-performance by date:

Previous
From: Dave Cramer
Date:
Subject: Re: effizient query with jdbc
Next
From: Carlos Benkendorf
Date:
Subject: Re: ORDER BY costs