SAP Performance Problem with Oracle 10 on Sun SPARC T5240 server

Dear Friends,
We have a performance problem after migrating our SAP ERP 6.0 basis system. We moved to new servers a month ago.
According to SAP EarlyWatchAlert Report, CPU response time is too high, altough CPU utilization is never more than 5%,
The current system is :
Database server : Sun SPARC Enterprise T5240 Server - 2 CPU´s with 6 core, 8 thread, 1.2 Ghz 32 GB RAM
and we use another identical configured server as an application server.
Database is Oracle 10.2.0 and operating system is Solaris 10.
The problem is average CPU response time is 450 ms, and max. CPU load is %5 percent.
Pre-migration configuration with old servers, we had CPU response : 150 ms and max CPU load: 50%.
Old configuration : 2 X HP rp3440, 2 X PA 8800 CPU (2 core, 1.0 GHz).
Have you got any experience with a similar situation, which setting might be wrong for not fully utilizing server CPU´s?
or do you know any similar configuration for benchmark?
Thanks in advance
Uzan

Our organization upgraded an application - The vendor had originally suggested a T2000. When we finally migrated onto it, the performance was worse than the older version of the application. The vendor hadn't yet tested the combination of T2000 and 10.2 with their application when they had made that hardware recommendation, they subsequently revised their recommendations. We ended up with an M4000.
The application performance, IMHO, still stinks. But that's because it's a java based application written for database independence.
The interesting item to note here, is that in the application teams 'selective' testing - everything worked fine. When the testing was extended to include users who didn't follow the 'test plan', they found performance begin to tank. Then with more users, more testing, they found that the T2000 was not going to perform well. The test plan by the apps team was not sufficient to find the issues, and they did not include any load testing!
Edited by: dbtoo on Jun 5, 2009 11:20 AM

Similar Messages

Performance problem with Oracle

We are currently getting a system developed in Unix/Weblogic/Tomcat/Oracle environment. We have developed a screen that contains 5 or 6 different parameters to select from. We could select multiple parameters in each of these selections. The idea behind the subsequent screens is to attach information to already existing data/ possible future data that matches the selection criteria.
Based on these selections, existing data located within the system in a table is searched and those that match are selected. Also new rows are created in the table against combinations that do not currently have a match. Frequently multiple parameters are selected, and 2000 different combinations need to be searched in the table. Of these selections, only about 100 or 200 combinations will be available in existing data. So the system is having to insert 1800 rows. The user meanwhile waits for the system to come up with data based on their selections. The user is not willing to wait more than 30 seconds to get to the next screen. In the above mentioned scenario, the system takes more than an hour to insert the new records and bring the information up. We need suggestions to see if the performance can be improved this drastically. If not what are the alternatives? Thanks

The #1 cause for performance problems with Oracle is not using it correctly.
I find it hard to believe that with the small data volumes mentioned, that you can have perfornance problems.
You need to perform a sanity check. Are you using Oracle correctly? Do you know what bind variables are? Are you using indexes correctly? Are you using PL/SQL correctly? Is the instance setup correctly? What about storage, are you using SAME (RAID10) or something else? Etc.
Facts. Oracle peforms exceptionally well. Oracle exceptionally well.
Simple example from a benchmark I did on this exact same subject. App-tier developers not understanding and not using Oracle correctly. Incorrect usage of Oracle doing a 100,000 SQL statements. 24+ minutes elapsed time. Doing those exact same 100,000 SQL statement correctly (using bind variables) - 8 seconds elapsed time. (benchmark using Oracle 10.1.0.3 on a Sunfire V20z server)
But then you need to use Oracle correctly. Are you familiar with the Oracle Concepts Guide? Have you read the Oracle Application Developer Fundamentals Guide?

Problem with maxprocs 1 on Sun One Web server 6.1

I am using sun one web server on Unix with maxprocs set to 3.
This means that there are 3 jvms running my application under 1 instance. However I have certain components like an application cache that I would like to share between these 3 processes.
Additionally is there a way to configure the server so that requests from 1 session are served by 1 process only.
I am really stuck here. Have searched in all the documentation, but in vain. Would really help if some one of the Sun Developers could answer this.
Thanks
Chetan

Elving,
It is great to hear that there is no predefined limit on number of sessions but if that's the case why the webserver always starts with 1000 allowed by default. Are there any potential performance reasons behind it?
I have your link open in my browser for a few days now as a reference because I am experimenting with different Session Managers. I am trying to find a solution for BheegaChicken to be able to accomodate several thousand simultaneous sessions for his application.
We use V6.0SP2 and SP6 and for BheegaChicken's app specifically V6.1SP2...
I have spent some time in the past looking for the answers how to handle higher then 1000 sessions capacity on one instance and that was even before V6.1 was out. I have found the following reco that talks about optimal performance for sites using threads:
http://docs.sun.com/source/816-5690-10/perf6.htm#17580
*** start qoute
Multi-Process Mode
You can configure the server to handle requests using multiple processes with multiple threads in each process. This flexibility provides optimal performance for sites using threads, and also provides backward compatibility to sites running legacy applications that are not ready to run in a threaded environment. Because applications on Windows NT generally already take advantage of multi-thread considerations, this feature applies to Unix/Linux platforms.
The advantage of multiple processes is that legacy applications that are not thread-aware or thread safe can be run more effectively in iPlanet Web Server. However, because all the Netscape/iPlanet extensions are built to support a single-process threaded environment, they may not run in the multi-process mode, and the Search plug-ins will fail on startup if the server is in multi-process mode.
In the multi-process mode, the server spawns multiple server processes at startup. Each process contains one or more threads (depending on the configuration) which receive incoming requests. Since each process is completely independent, each one has its own copies of global variables, caches, and other resources. Using multiple processes requires more resources from your system. Also, if you try to install an application which requires shared state, it has to synchronize that state across multiple processes. NSAPI provides no helper functions for implementing cross-process synchronization.
If you are not running any NSAPI in your server, you should use the default settings: one process and many threads. If you are running an application which is not scalable in a threaded environment, you should use a few processes and many threads, for example, 4 or 8 processes and 128 or 512 threads per process.
MaxProcs (Unix/Linux)
Use this directive to set your Unix/Linux server in multi-process mode, which may allow for higher scalability on multi-processor machines. If you set the value to less than 1, it will be ignored and the default value of 1 will be used.
Tuning
You can set the value for MaxProcs by:
Editing the MaxProcs parameter in magnus.conf
Setting or changing the MaxProcs value in the Magnus Editor of the Server Manager
Note You will receive duplicate startup messages when running your server in MaxProcs mode.
*** end quote
If it is not relevant anymore please explane.
I am still interested to understand which of the SessionManagers is recommended by Sun to have optimal performance for sites with high number of sessions and using threads. It is not really obvious from http://docs.sun.com/source/817-6251/pwasessn.html. On top of this that's documented just for V6.1. What about V6.0?
Thank you very much for your cooperation!
Sasha aka ttalex

Problem with CMP BEAN ON SUN JAVA APPLICATION SERVER 1.3.1

I am trying to deploy an entity bean with container managed persistence on sun java app. server 1.3.1. When I go for Deploy, the error 'SQL code not generated for following beans' is comming, when verified with verifier tool, the error shown is 'transaction attributes are not specified' , What should be the problem ?

Which appserver are you using exactly? the J2EE 1.3.1 RI, SJAS 7.0, SJSAS 7.1....etc
You need to make sure for that you have specified the correct deployment attributes for your CMP methods. You can pull the ear/jar into deploytool and review it there

Performance problems with Oracle 9.2.0.2 Database with optimizer_mode=RULE

If I make an explain plan for a query of the following form:
SELECT a.*
FROM a, b
WHERE a.pk = b.fk
AND b.pk IN (1, 2);
the explain plan states, that a FULL-TABLE-SCAN on table b is performed and a HASH-JOIN is done to execute the query.
On 8.0.6 Database the Indexes on the tables are used and a NESTED-LOOP join is performed.
Why is the 9.2.0.2 Database using HASH-JOINS when the Database is explicitly set to optimizer_mode=RULE, as I think HASH-JOINS are only possible with CBO?
How can I force the 9.2.0.2 database to perform exactly like the 8.0.6 database with RBO?

some options:
set the hash_join_enabled parameter to false in the init.ora file.
set the compatible parameter to 8.0 in the init.ora file.
documentation:
HASH_JOIN_ENABLED specifies whether the optimizer should consider using a
hash join as a join method. When set to FALSE, hash join is turned off; that is, it is
not available as a join method that the optimizer can consider choosing. When set to
TRUE, the optimizer will compare the cost of a hash join to other types of joins, and
choose it if it gives the best cost.
COMPATIBLE allows you to use a new release, while at the same time guaranteeing
backward compatibility with an earlier release. This is in case it becomes necessary
to revert to the earlier release. This parameter specifies the release with which the
Oracle server must maintain compatibility. Some features of the current release may
be restricted.

Problem with oracle forms,jre & sun thin clients

we have oracle forms using jre 1.5.0 update 11 running on more than 100 thin clients . but the application hangs when navigating through the forms sometime. if all the browser window is closed and started from the begining it again works fine
any support is highly appreciated
thanks
mish

I have a RHEL4(Linux) 64 system I can test and post my notes here as an edit.
For Java based web applications to work, the Firefox 3.6 browser requires you to run Java SE 6 Update 18 or later.
http://java.sun.com/products/archive/j2se/6u22/index.html
jre-6u22-linux-x64.bin
As root
cd /usr
mkdir java
cd java
With a copy of "jre-6u22-linux-x64.bin" in /usr/java/
Change the permission of the file you downloaded to be executable. Type:
chmod a+x jre-6u22-linux-x64.bin
Run the self-extracting binary Type
./jre-6u22-linux-x64.bin
The license agreement is displayed. Review the agreement.
Press the spacebar to display the next page.
At the end, enter yes to proceed with the installation.
Did not display on mine
Double check of mozilla install
find / -name mozilla -print
/usr/share/mozilla
/usr/lib64/mozilla
cd /usr/lib64/mozilla/
mkdir plugins
Exit Firefox browser if it is already running.
Create a symbolic link to the libnpjp2.so file in the browser plugins directory
ln -s /usr/java/jre1.6.0_22/lib/amd64/libnpjp2.so
Restart and test firefox
about:plugins (works)
Best Regards
mseberg
Edited by: mseberg on Aug 11, 2011 11:11 AM
MORE
The application less happy until I changed these lines in my formsweb.cfg
jpi_mimetype=application/x-java-applet;jpi-version=1.6.0_22
jpi_classid=clsid:CAFEEFAC-0016-0000-0022-ABCDEFFEDCBA
Seems it wanted the exact JRE version that was installed.
Edited by: mseberg on Aug 11, 2011 11:18 AM
And finally :
http://www.java.com/en/download/help/linux_install.xml

Performance problems with SAP GUI 7.10 and BEx 3.5 Patch 400?

Hi everybody,
we installed SAP GUI 7.10 and BEx 3.5 Patch 400 and detected hugh performance problems with this version in comparison to the SAP GUI 6.40 and BEx 3.5 or BEx 7.0 Patch 800.
Does anybody detect the same problems?
Best regards,
Ulli

Most important question when you are talking about performance-issues:
which OC are you working on and which excel version?
ciao
Joke

Performance problem with Orcacle / Crystal reports 8.5 / ASP

Hi,
I have a performance problem with reports running crystal reports 8.5 with oracle databases.
Fetching data appears to be slow. Oracle trace reports (tkprof) shows that fetchsize=1.
Fetchsize in Oracle ODBC is 15 (default).
How can I update te fetchsize in crystal reports?
I would appreciate any help.

CR XI R2 is also still available for purchase. This is the last version of CR to use the RDC. Contact sales at 866-681-3435.
Make sure the installed versin is 11.5.12.1838 - if it is not, apply SP6:
https://smpdl.sap-ag.de/~sapidp/012002523100015859952009E/crxir2win_sp6.exe
MSM:
https://smpdl.sap-ag.de/~sapidp/012002523100000634032010E/crxir2sp6_rdc_mm.zip
- Ludek
Edited by: Ludek Uher on Jan 4, 2011 10:05 AM

Performance problem with ojdbc14.jar

Hi,
We are having performance problem with ojdbc14.jar in selecting and updating (batch updates) entries in a table. The queries are taking minutes to execute. The same java code works fine with classes12.zip ans queries taking sub seconds to execute.
We have Oracle 9.2.0.5 Database Server and I have downloaded the ojdbc14.jar from Oracle site for the same. Tried executing the java code from windows 2000, Sun Solaris and Opteron machines and having the same problem.
Does any one know a solution to this problem? I also tried ojdbc14.jar meant for Oracle 10g, that did not help.
Please help.
Thanks
Yuva

My code is doing some thing which might be working well with classes12.zip and which does not work well with ojdbc14.jar? Any general suggestions to make the code better, especially for batch updates.
But for selecting a row from the table, I am using index columns in the where cluase. In the code using PreparedStatement, setting all the reuired fields. Here is the code. We have a huge index with 14 fields!!. All the parameters are for where clause.
if(longCallPStmt == null) {
longCallPStmt = conn.prepareStatement(longCallQuery);
log(Level.FINE, "CdrAggLoader: Loading tcdragg entry for "
+GeneralUtility.formatDate(cdrAgg.time_hour, "MM/dd/yy HH"));
longCallPStmt.clearParameters();
longCallPStmt.setInt(1, cdrAgg.iintrunkgroupid);
longCallPStmt.setInt(2, cdrAgg.iouttrunkgroupid);
longCallPStmt.setInt(3, cdrAgg.iintrunkgroupnumber);
longCallPStmt.setInt(4, cdrAgg.iouttrunkgroupnumber);
longCallPStmt.setInt(5, cdrAgg.istateregionid);
longCallPStmt.setTimestamp(6, cdrAgg.time_hour);
longCallPStmt.setInt(7, cdrAgg.icalltreatmentcode);
longCallPStmt.setInt(8, cdrAgg.icompletioncode);
longCallPStmt.setInt(9, cdrAgg.bcallcompleted);
longCallPStmt.setInt(10, cdrAgg.itodid);
longCallPStmt.setInt(11, cdrAgg.iasktodid);
longCallPStmt.setInt(12, cdrAgg.ibidtodid);
longCallPStmt.setInt(13, cdrAgg.iaskzoneid);
longCallPStmt.setInt(14, cdrAgg.ibidzoneid);
rs = longCallPStmt.executeQuery();
if(rs.next()) {
cdr_agg = new CdrAgg(
rs.getInt(1),
rs.getInt(2),
rs.getInt(3),
rs.getInt(4),
rs.getInt(5),
rs.getTimestamp(6),
rs.getInt(7),
rs.getInt(8),
rs.getInt(9),
rs.getInt(10),
rs.getInt(11),
rs.getInt(12),
rs.getInt(13),
rs.getInt(14),
rs.getInt(15),
rs.getInt(16)
}//if
end_time = System.currentTimeMillis();
log(Level.INFO, "CdrAggLoader: Loaded "+((cdr_agg==null)?0:1) + " "
+ GeneralUtility.formatDate(cdrAgg.time_hour, "MM/dd/yy HH")
+" tcdragg entry in "+(end_time - start_time)+" msecs");
} finally {
GeneralUtility.closeResultSet(rs);
GeneralUtility.closeStatement(pstmt);
Why that code works well for classes12.zip (comes back in around 10 msec) and not for ojdbc14.jar (comes back in 6-7 minutes)?
Please advise.

Performance problems with new Java Tiger style recommendations

Performance problems with jdk 1.5 on Linux plattform
(not tested on Windows, might be the same)
using the new style recommendations.
I need fast Vector loops for high speed mathematical calculations, some
hints about the fastest way to program that loop would be also great!
After refactoring using the new features from java 1.5 (as recommended from
SUN) I lost performance significantly:
using a vector:
public Vector<unit> units;
The new code (recommended from SUN for Java Tiger for redesign):
for (unit u: units) u.accumulate();
runs more than 30% slower than the old code:
for (int i = 0; i < units.size(); i++) units.elementAt(i).accumulate();
I expected the opposite.
Is there any information available that helps?
The following additional information I got from Mr. Shankar Unni:
I got some fairly anomalous results comparing ArrayList and Vector: for the
1.5-style loops, ArrayList was faster then Vector, but for a loop with get()
calls, Vector was faster. Vector was even faster than that using
elementAt(), which was a surprise:
For a million summing iterations over a 100-element array:
vector elementAt loop took 3446 ms.
vector get loop took 3796 ms.
vector iterator loop took 5469 ms.
arraylist get loop took 4136 ms.
arraylist iterator loop took 4668 ms.

If your topic doesn't change, please stay in your original post.

Performance problems with EP 6 and MS IE

Hi everybody,
since a couple of days, we are facing a sever performance problem with our SAP EP 6.0. When I access the system with MS Internet Explorer 6.0, it takes 5-10 minutes after the Login. With Firefox Browser, the performance is ok. Therefore I assume that it must be a problem with the IE settings. Does anybody know a solution?
Best regards,
Michael

There are a few things that this could be. I've seen the setting "Empty Temporary Internet Files folder when browser is closed" cause a lot of performance problems (This is in the advanced settings in your IE).
This will cause your cache to be cleared out each time the browser is closed and cause a lot more data to be downloaded each time you login to the system.
For more analysis I'd recommend putting a tool like HTTPWatch into your IE browser and seeing which requests are using the most time.

Problem with oracle.jbo.domain.Date

Hi there,
I've a problem with oracle.jbo.domain.Date,
I'm doing this code (this part of code is used in my Session Attribute Listener):
* This method is used to add the session id in the database, whenever user login
public void attributeAdded(HttpSessionBindingEvent hsbe){
if(!hsbe.getName().equals("user")){
return;
AmLogin am = (AmLogin)Configuration.createRootApplicationModule(
"com.ahm.pdt001.am.AmLogin",
"AmLoginLocal");
try{
hsbe.getSession().setAttribute("login", new Date(new Timestamp(
System.currentTimeMillis())));
am.createLogin(hsbe.getValue().toString(), hsbe.getSession().
getAttribute("login").toString());
} catch(Exception e){
e.printStackTrace();
System.out.println("Error insert data user: " + hsbe.getValue());
} finally{
Configuration.releaseRootApplicationModule(am, true);
Everything is running well in Jdev 10.1.3 (I'm using ADF Faces technology), and I'm trying to deploy it on OC4J 10.1.2 and it works. But it raised an error when this part of code is runned.
--------------caused and error------------------------
hsbe.getSession().setAttribute("login", new Date(new Timestamp(
System.currentTimeMillis())));
And this is an error:
javax.faces.FacesException: #{pdt001.loginAction}: javax.faces.el.EvaluationException: java.lang.NoSuchMethodError: oracle.jbo.domain.Date.<init>(Ljava/sql/Timestamp;)V at com.sun.faces.application.ActionListenerImpl.processAction(ActionListenerImpl.java:78) at oracle.adf.view.faces.component.UIXCommand.broadcast(UIXCommand.java:211) at javax.faces.component.UIViewRoot.broadcastEvents(UIViewRoot.java:267) at javax.faces.component.UIViewRoot.processApplication(UIViewRoot.java:381) at com.sun.faces.lifecycle.InvokeApplicationPhase.execute(InvokeApplicationPhase.java:75) at com.sun.faces.lifecycle.LifecycleImpl.phase(LifecycleImpl.java:200) at com.sun.faces.lifecycle.LifecycleImpl.execute(LifecycleImpl.java:90) at javax.faces.webapp.FacesServlet.service(FacesServlet.java:197) at com.evermind[Oracle Application Server Containers for J2EE 10g (10.1.2.0.0)].server.http.ResourceFilterChain.doFilter(ResourceFilterChain.java:65) at oracle.adf.model.servlet.ADFBindingFilter.doFilter(ADFBindingFilter.java:332) at com.evermind[Oracle Application Server Containers for J2EE 10g (10.1.2.0.0)].server.http.EvermindFilterChain.doFilter(EvermindFilterChain.java:16) at oracle.adfinternal.view.faces.webapp.AdfFacesFilterImpl._invokeDoFilter(AdfFacesFilterImpl.java:367) at oracle.adfinternal.view.faces.webapp.AdfFacesFilterImpl._doFilterImpl(AdfFacesFilterImpl.java:336) at oracle.adfinternal.view.faces.webapp.AdfFacesFilterImpl.doFilter(AdfFacesFilterImpl.java:196) at oracle.adf.view.faces.webapp.AdfFacesFilter.doFilter(AdfFacesFilter.java:87) at com.evermind[Oracle Application Server Containers for J2EE 10g (10.1.2.0.0)].server.http.EvermindFilterChain.doFilter(EvermindFilterChain.java:20) at com.ahm.filter.AhmFilterSession.doFilter(AhmFilterSession.java:45) at com.evermind[Oracle Application Server Containers for J2EE 10g (10.1.2.0.0)].server.http.ServletRequestDispatcher.invoke(ServletRequestDispatcher.java:645) at com.evermind[Oracle Application Server Containers for J2EE 10g (10.1.2.0.0)].server.http.ServletRequestDispatcher.forwardInternal(ServletRequestDispatcher.java:322) at com.evermind[Oracle Application Server Containers for J2EE 10g (10.1.2.0.0)].server.http.HttpRequestHandler.processRequest(HttpRequestHandler.java:790) at com.evermind[Oracle Application Server Containers for J2EE 10g (10.1.2.0.0)].server.http.HttpRequestHandler.run(HttpRequestHandler.java:270) at com.evermind[Oracle Application Server Containers for J2EE 10g (10.1.2.0.0)].server.http.HttpRequestHandler.run(HttpRequestHandler.java:112) at com.evermind[Oracle Application Server Containers for J2EE 10g (10.1.2.0.0)].util.ReleasableResourcePooledExecutor$MyWorker.run(ReleasableResourcePooledExecutor.java:186) at java.lang.Thread.run(Thread.java:534)Caused by: javax.faces.el.EvaluationException: java.lang.NoSuchMethodError: oracle.jbo.domain.Date.<init>(Ljava/sql/Timestamp;)V at com.sun.faces.el.MethodBindingImpl.invoke(MethodBindingImpl.java:130) at com.sun.faces.application.ActionListenerImpl.processAction(ActionListenerImpl.java:72) ... 23 moreCaused by: java.lang.NoSuchMethodError: oracle.jbo.domain.Date.<init>(Ljava/sql/Timestamp;)V at com.ahm.pdt001.listener.Pdt001AttributeSessionListener.attributeAdded(Pdt001AttributeSessionListener.java:27) at com.evermind[Oracle Application Server Containers for J2EE 10g (10.1.2.0.0)].server.http.EvermindHttpSession.setAttribute(EvermindHttpSession.java:128) at com.ahm.pdt001.bean.Pdt001Bean.loginAction(Pdt001Bean.java:61) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25) at java.lang.reflect.Method.invoke(Method.java:324) at com.sun.faces.el.MethodBindingImpl.invoke(MethodBindingImpl.java:126)
Does anybody have the same problem like this? I've used the same method in my previous project to create oracle.jbo.domain.Date with current time using java.sql.Timestamp, but I don't know why this error happened.
Thx,
Andre

Re: oracle jbo.domain.Date issues in 10.1.3
Sascha

Performance problem between Oracle.DataAccess v1 and v2

Hi, I have serious performance problem with OracleDataReader when I use the GetValues method.
My server is Oracle 9.2.0.7, and i use ODAC v10.2.0.221
I create a dummy table for benchmark :
create table test (a varchar2(50), b number)
begin
for i in 1..62359 loop
insert into test values ('Values ' || i, i);
end loop;
commit;
end;
I use the same code for benchmark Framework v1 and Framework v2.
Code :
try {
OracleConnection c = new OracleConnection("user id=saturne_dbo;password=***;data source=satedfx;");
c.Open();
go(c);
c.Close();
catch (Exception ex) {
MessageBox.Show(ex.Message);
private void go(IDbConnection c) {
IDbCommand cmd = c.CreateCommand();
cmd.CommandText = "select * from test";
cmd.CommandType = CommandType.Text;
DateTime dt = DateTime.Now;
IDataReader reader = cmd.ExecuteReader();
int count = 0;
while (reader.Read()) {
object[] fields = new object[reader.FieldCount];
reader.GetValues(fields);
count++;
reader.Close();
TimeSpan eps = DateTime.Now - dt;
MessageBox.Show("Time " + count + " : " + eps.TotalSeconds);
Result are :
Framework v1 with OracleDataAccess 1.10.2.2.20 "Time 62359 : 0.5"
Framework v2 with OracleDataAccess 2.10.2.2.20 "Time 62359 : 3.57" FACTOR 6 !!!!!
I notice same problem with oleDb provider and Microsoft Oracle Client provider..
It's a serious problem for my production server, the time calculation explode...
Where is the explication ?
Do u know solution ?

Can you please try out following -
1. Create a .NET 1.x DLL with your benchmark code. This will obviously use ODP.NET for .NET 1.x.
2. Call this assembly routine from a .NET 1.x executable and note the results.
3. Now call this assembly routine from a .NET 2.0 executable and note the results.
The idea is to always use "ODP.NET for .NET 1.x" even in .NET 2.0 runtime. This will tell us whether the performance degradation is a runtime issue.

Performance problem with sdn_nn - new 10g install

I am having a performance problem with sdn_nn after migrating to a new machine. The old Oracle version was 9.0.1.4. The new is 10g. The new machine is faster in general. Most (non-spatial) batch processes run in half the time. However, the below statement is radically slower. The below statement ran in 45 minutes before. On the new machine it hasn't finished after 10 hours. I am able to get a 5% sample of the customers to finish in 45 minutes.
Does anyone have any ideas on how to approach this problem? Any chance something isn't installed correctly on the new machine (the nth version of the query finishe, albeit 20 times slower)?
Appreciate any help. Thanks.
- Jack
create table nearest_store
as
select /*+ ordered */
a.customer_id,
b.store_id nearest_store,
round(mdsys.sdo_nn_distance(1),4) distance
from customers a,
stores b
where mdsys.sdo_nn(
b.geometry,
a.geometry,
'sdo_num_res=1, unit=mile',
1
) = 'TRUE'
;

Dan,
customers 110,000 (involved in this query)
stores 28,000
Here is the execution plan on the current machine:
CREATE TABLE STATEMENT cost = 81947
LOAD AS SELECT
PX COORDINATOR
PX SEND QC (RANDOM) :TQ10000
ROW NESTED LOOPS
1 1 PX BLOCK ITERATOR
1 1ROW TABLE ACCESS FULL CUSTOMERS
1 3 PARTITION RANGE ALL
1 3 TABLE ACCESS BY LOCAL INDEX ROWID STORES
DOMAIN INDEX STORES_SIDX
I can't capture the execution plan on the old database. It is gone. I don't remember it being any different from the above (full scan customers, probe stores index once for each row in customers).
I am trying the query without the create table (just doing a count). I'll let you know on that one.
I am at 10.0.1.3.
Here is how I created the index:
create index stores_sidx
on stores(geometry)
indextype is mdsys.spatial_index LOCAL
Note that the stores table is partitioned by range on store type. There are three store types (each in its own partition). The query returns the nearest store of each type (three rows per customer). This is by design (based on the documented behavior of sdo_nn).
In addition to running the query without the CTAS, I am also going try running it on a different machine tonight. I let you know how that turns out.
The reason I ask about the install, is that the Database Quick Installation Guide for Solaris says this:
"If you intend to use Oracle JVM or Oracle interMedia, you must install the Oracle Database 10g Products installation type from the Companion CD. This installation optimizes the performance of those products on your system."
And, the Database Installlation Guide says:
"If you plan to use Oracle JVM or Oracle interMedia, Oracle strongly recommends that you install the natively compiled Java libraries (NCOMPs) used by those products from the Oracle Database 10g Companion CD. These libraries are required to improve the performance of the products on your platform."
Based on that, I am suspicious that maybe I have the product installed on the new machine, but not correctly (forgot to set fast=true).
Let me know if you think of anything else you'd like to see. Thanks.
- Jack

Performance problem with table COSS...

Hi
Anyone encountered performance problem with these table : COSS, COSB, COEP
Best Regards

>
gsana sana wrote:
> Hi Guru's
>
> this is the select Query which is taking much time in Production. so please help me to improve the preformance with BSEG.
>
> this is my select query:
>
> select bukrs
>               belnr
>               gjahr
>               bschl
>               koart
>               umskz
>               shkzg
>               dmbtr
>               ktosl
>               zuonr
>               sgtxt
>               kunnr
>         from bseg
>         into table gt_bseg1
>          for all entries in gt_bkpf
>        where bukrs eq p_bukrs
>          and belnr eq gt_bkpf-belnr
>          and gjahr eq p_gjahr
>          and buzei in gr_buzei
>          and bschl eq '40'
>          and ktosl ne 'BSP'.
>
> UR's
> GSANA
Hi,
This is what I know and please if any expert think its incorrect, please do correct me.
BSEG is a cluster table with BUKRS, BELNR, GJAHR and BUZEI as the key whereas other key will be stored in database as raw data thus SAP apps will need to convert that raw data first if we are using other keys in where condition. Hence, I suggest to use up to buzei in the where condition and filter other condition in internal table level like using Delete statement. Hope its help.
Regards,
Abraham

SAP Performance Problem with Oracle 10 on Sun SPARC T5240 server

Similar Messages

Maybe you are looking for