High CPU utilization usually means that something is not setup correctly in
your tests. I would echo Chris' sentiment. I'd like to see how you are
initializing the spring/ibatis as well as a breakdown of your testing
metrics.
Brandon
On 11/17/06, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
Please let me know what kind of information you need that will help you I
will be glad to give you the information. I have already provided you our
Sqlmap_config file details. We are doing a simple select statement.
We are basically simulating the load on WAS 6.1 using JMeter and trying to
compare the ibatis to JDBC.What we noticed is that the CPU usage is very
high when running Ibatis. Hence the question.
Thanks.
Regards,
Sanjay
x7273
"Brandon Goodin"
<[EMAIL PROTECTED]
mail.com> To
[email protected]
11/17/2006 11:36 cc
AM
Subject
Re: Fw: ibatis with Spring
Please respond to performance issue.
[EMAIL PROTECTED]
apache.org
The monthly ambiguous performance question. If you are going to make
question/statements on performance, please provide better information
about
how your reached the conclusion that prompted the question. Otherwise your
question is akin to asking how big the universe is.
Thanks :)
Brandon
On 11/17/06, [EMAIL PROTECTED] < [EMAIL PROTECTED]>
wrote:
Any ideas why the CPU usage is more when using Ibatis vs regular JDBC.
Thanks.
Regards,
Sanjay
x7273
----- Forwarded by Sanjay Deshpande/IT/Quixtar on 11/17/2006 09:50 AM
-----
Sanjay
Deshpande/IT/Quix
tar
To
[email protected]
11/10/2006 02:19
cc
PM [EMAIL PROTECTED],
[email protected],
Attila
Halasz/IT/[EMAIL PROTECTED], Nate
Pett/IT/[EMAIL PROTECTED]
Subject
Re: ibatis with Spring
performance
issue.(Document link: Sanjay
Deshpande)
Thanks Larry for your quick Response. Here is our SQLMAP-config file
details.
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE sqlMapConfig PUBLIC "-//ibatis.apache.org//DTD SQL Map Config
2.0
//EN" " http://ibatis.apache.org/dtd/sql-map-config-2.dtd";>
<sqlMapConfig>
<!-- These settings control SqlMapClient configuration details,
primarily to do with transaction
management. They are all optional (more detail later in this
document). -->
<settings cacheModelsEnabled="true" enhancementEnabled="true"
lazyLoadingEnabled="true" maxRequests="512"
maxSessions="256"
maxTransactions="32" useStatementNamespaces="true"
classInfoCacheEnabled="true"
errorTracingEnabled="false"/>
<sqlMap
resource=" SqlMap.xml" />
</sqlMapConfig>
The query in SqlMap.xml is as follows
<resultMap id="detailMap" class="com.mycompany.dto.Detail">
<result property="typeCode" column="typeCode" columnIndex="1" />
<result property="xAmount" column="xTotal" columnIndex="2"
jdbcType=
"DECIMAL" />
<result property="yAmount" column="yTotal" columnIndex="3"
jdbcType=
"DECIMAL"/>
<result property="numberOfCustomers" column="numberOfCustomers"
columnIndex="4" jdbcType="DECIMAL"/>
<result property="numberOfOrders" column="numberOfOrders"
columnIndex
="5" jdbcType="DECIMAL" />
<result property="currency" column="currency" columnIndex="6" />
</resultMap>
<select id="getDetailInformation" resultMap="detailMap" remapResults=
"false"
parameterClass="java.util.Map" >
SELECT JLAWPK as typeCode,SUM(JLAWPL) as xTotal,
SUM(JLAWPM)
as yTotal, SUM(JLAWPN) as numberOfCustomers, SUM(JLAWPO) as
numberOfOrders, JLK2CD as currency
FROM Table1
WHERE JLAEPA = #idNumber:DECIMAL# AND JLAF4H =
#period:DECIMAL# AND JLAWPK = #transactionCode:VARCHAR#
GROUP BY JLAEPA, JLAF4H, JLAWPK, JLK2CD
</select>
The data Base is AS400. We are running this on Websphere 6.1.
When we run the JProfiler for both the JDBC and spring-iBatis
implementation on this test Application , we notice that the CPU
utilisation is very high in case of the spring-ibatis implementation.
Please advise us on how we can configure ibatis to resolve this issue.
If
needed we will provide the detailed Profiler trace.
"Larry Meadors"
<[EMAIL PROTECTED]
org>
To
Sent by: [email protected]
[EMAIL PROTECTED]
cc
il.com
Subject
Re: ibatis with Spring
performance
11/09/2006 04:45 isssue.
PM
Please respond to
[EMAIL PROTECTED] .
apache.org
Heh, well...there are about 4,890,456,354 ways to configure that. :-)
So, I guess if it's within your performance criteria don't worry about
it. If it's not, look at it with a profiler and see what you can do to
make it better. If it is a deficiency in the framework, we'll see what
we can do to improve it.
Larry
On 11/9/06, [EMAIL PROTECTED] <[EMAIL PROTECTED]>
wrote:
>
> We ran performance tests on one of our application with Spring/IBATIS
vs
> regular JDBC. As far as the response time goes it was roughly the
same.
But
> the CPU usage was 4 to 5 times higher in Spring/IBATIS implementation.
Has
> anybody encountered the same issue.? Is that is normal Why?
>
> Thanks.
>
> Regards,
> Sanjay
> x7273
>
>
