MQSeries.net :: View topic - performance report stats for databases updates

MQSeries.net

Tech Exchange

Education

Certifications

Library

Info Center

SupportPacs

FAQÂ Â

Usergroups

RSS Feed - WebSphere MQ Support

RSS Feed - Message Broker Support

MQSeries.net Forum Index » WebSphere Message Broker (ACE) Support » performance report stats for databases updates

performance report stats for databases updates

« View previous topic :: View next topic »

Author

Message

dean

Posted: Wed Jul 15, 2009 12:40 pm Post subject: performance report stats for databases updates

Novice

Joined: 29 May 2009
Posts: 15

The Linux performance report for WMB indicates that 688 messages per second (in the test lab) can be processed using a Database Update (on page 32). But it says that using a stored procedure drops performance to 4 messages per second (on page 34).

It seems that using stored procedures would be more efficient.

Does anyone know why the performance report measured stored procedure utilization so low?

Thanks

rbicheno

Posted: Thu Jul 16, 2009 7:53 am Post subject:

Apprentice

Joined: 07 Jul 2009
Posts: 43

Hi, you need to read the test description abit more carefully

Within the compute node the message headers from the incoming message are copied over
to the outgoing message. Two thousand identical calls are made to an external database
stored procedure. The procedure receives one parameter which is an integer and passes
back zero parameters returning immediately.
This test identifies the cost of calling a Database Stored procedure with one parameter which
is an integer.
The results of running this test are given in the table below. The CPU ms/msg figure has
been adjusted to report a per procedure invocation cost by dividing the CPU cost obtained
from the test results by 2000.

So inreality this means for every message we sent in we made 2000 stored proc calls. This was done to focus costs on the stored proc call itself but i can see why it may not be clear compared to other tests. We can look at changing this test for the next set of reports.

dean

Posted: Thu Jul 16, 2009 8:29 am Post subject:

Novice

Joined: 29 May 2009
Posts: 15

Yes, I didn't understand what was going on in the compute node. It would be nice to compare the 2000 stored procedure calls per message to 2000 database updates per message.

By looking at the minimal processing statitics and doing some rough math, it looks like the stored procedure call is about 8 times faster than the database update.

:thumbup:

thanks for clearing this up for me

dean

Posted: Thu Jul 16, 2009 9:57 am Post subject:

Novice

Joined: 29 May 2009
Posts: 15

I'm still comparing apples and oranges. The stored procedure referenenced doesn't do anything.

I'll just have to experiment...but I expect the stored procedure call to update a database will be faster and cleaner.

Display posts from previous:

Page 1 of 1

MQSeries.net Forum Index » WebSphere Message Broker (ACE) Support » performance report stats for databases updates

Jump to:

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum

Protected by Anti-Spam ACP