Re: [ITM] Large JDBC result sets with Agent Builder

Julius Loman <[email protected]>
Newsgroups gmane.comp.sysutils.tivoli.general,gmane.comp.sysutils.tivoli.tme10
Message-ID <[email protected]>
On Mon, Dec 10, 2012 at 09:16:49AM -0500, Jay Slay <[email protected]> wrote:
> Hey guys.  Was wondering what sort of performance you have seen with agent
> builder agents pulling more than a few thousand rows back from a query.  It
> appears that over 10k rows is taking one of our agents more than a minute to
> process.   While the agent server itself isn't the beefiest box out there,
> over a minute seems a bit slow to me.  There is only a dozen or so columns
> and the data in each is fairly small.  From the logging, we can see it is not
> the query taking time, but the processing of the result set. 
Internally JDBC processing is done in separate JVM and communicating to the
agent using local TCP connection. Data is transferred in XML and must this XML
must be constructed and and parsed back. This could be the bottleneck you are
facing.

I recall some settings that are limiting the result to 1000 rows - for
performance resons. CDP_JDBC_MAX_ROWS. Did you increase this value?

I think you will to refactor your agent and return less rows anyway. Probably
split the data into more attribute groups and using some filters in the SQL
query.


-- 

[ Julius Loman ][ [email protected] ][ http://lomo.kyberia.net ][ icq:35732873 ]

_______________________________________________
TME10 mailing list
[email protected]
Unsubscribe:[email protected]
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.