Performance Tuning: Http Task Time Response Time Threshold

I must analyze the response time of HTTP task on my instance and I found any value from tcode ST03.
For HTTP task with my user I have these values:
N. Step T resp.time A. Resp.Time T. DB Time A. DB Time
3405 3072 902,1 1436 421,7
Now I would like to know if these values are correct or they are very high.
Is there an OSS Note which explain the threshold of all task type??
Thanks in advanced.
Moreno

If its an external ITS,
here is one note:
Note 388198 - ITS performance problems - Monitors and log files
if its an internal ITS,
Note 892442 - Integrated ITS configuration/performance
Regards,
Siddhesh

Similar Messages

Performance tuning for a time-out dump

Hi ,
actually there is a following piece of code(a SELECT query) in my program,due to which TIME OUT dump occurs.plz help me to modify the code to reduce it's execution time.
code is:
DATA: at_seque_extref TYPE zprms_ord_ind OCCURS 0 WITH HEADER LINE.
            SELECT (sv_fields) FROM zprms_ord_ind
              APPENDING CORRESPONDING FIELDS OF TABLE at_seque
                 UP TO lv_max_sequs ROWS
                  WHERE process_type    IN lr_process_type
                    AND   posting_date    IN lr_pos_date
                    AND   sol_date        IN lr_sol_date
                    AND   changed_at      IN lr_chg_date
                  AND   external_ref    NE space
                    AND (ls_where)
                    AND   ib_comp_network IN lr_ibc_netw
                    AND   ib_comp_node    IN lr_ibc_node
                    AND   ib_comp_site    IN lr_ibc_site
                ORDER BY (iv_order_by).
            CLEAR at_seque_extref.
            LOOP AT at_seque.
              IF at_seque-external_ref CP is_bus_trans_search-external_ref .
                MOVE-CORRESPONDING at_seque TO at_seque_extref.
                APPEND at_seque_extref.
              ENDIF.
            ENDLOOP.
            CLEAR at_seque[].
*the table at_seque[] type zprms_ord_ind
Or plz suggest me check points to reduce it's execution time.

Hi Ahmed,
Rarther going to appending go for into corresponding field of table at_seque.
Also check the primary key combination before selecting any data into where condition.
Remove that NE condition from where clause because some time it will take time to extract data. GO for delete statement into at_seque after select statement.
This will definitely work.
<b>Rewards points if useful.</b>
Regards,
Kinjal

Application Performance Tuning

Hi,
Sometimes our application(11.5.10.2) is becoming very slow, I want to know what things I need to check and if there is any Document
Regards

Hi,
We discussed this topic many times in the forum before, please see old threads for details.
Performance Tuning
http://forums.oracle.com/forums/search.jspa?threadID=&q=Performance+Tuning&objID=c3&dateRange=all&userID=&numResults=15&rankBy=10001
Thanks,
Hussein

Performance Tuning in XI

Hi experts,
Can anybody tell me how to do performance tuning in XI? If any one has document plz give the link.
Regards,
venky

Hi,
XI have provided the Performance monitoring options to fine tune the performance in RWB.
Refer
Try performance tuning :
https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uuid/70ada5ef-0201-0010-1f8b-c935e444b0ad
http://help.sap.com/saphelp_nw04/helpdata/en/9e/6921e784677d4591053564a8b95e7d/frameset.htm
https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uuid/489f5844-0c01-0010-79be-acc3b52250fd
XI Performance Testing Tool
https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uuid/70ada5ef-0201-0010-1f8b-c935e444b0ad#search=%22XI%20sizing%20guide%22
/people/sap.user72/blog/2004/11/28/how-robust-is-sap-exchange-infrastructure-xi
XI Performance issues
Thanks
Swarup

Performance tuning of the XI System

Hi Everybody,
We are testing the Interfaces in QA System and I see the messages are getting stuck when we have load.
Is there any performance tuning which can be done to better the performance..
Can somebody suggest me the options?
Helpful answers will be rewarded points.
Thanks & Regards,
Rupash K

Hi,
Check this "SAP Note 857530 "
Integration process performance(in sap XI).
Refer Try performance tuning :
https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uuid/70ada5ef-0201-0010-1f8b-c935e444b0ad
http://help.sap.com/saphelp_nw04/helpdata/en/9e/6921e784677d4591053564a8b95e7d/frameset.htm
Regards
Seshagiri

In sap bi performance tuning option is considered in cube not in ods, why?

details about
indexes,partitions,aggregates,compression,roll up
how these are helping to increase system performance
why ods is not suitable?

Hi,
Generally we do performance tuning on the cubes as most of the reportings are done on the cubes hpowever in case of DSO also we can increase the performance by creating indexes of deactivating SID generation flag if there is no reporting on the DSO.
Following 2 links will show you different aspects of performance tuning :-
http://www.sdn.sap.com/irj/scn/index?rid=/library/uuid/404544e7-83c9-2e10-7b80-a24d5099ce3f
For LID and High Cardinality
http://www.sdn.sap.com/irj/scn/index?rid=/library/uuid/005f3197-d3da-2e10-1a94-a5c62342f2aa
You may create aggregates on the cube if some of IO are being used in the query very frequently to enhance the performce of query execution.
Navesh

ABAP Programs Performance Tuning and Web Services

Hi,
Can anyone give me any good material link or eBook on SAP ABAP programs Performance Tuning. What are the things that needs to be done for performance tuning etc..
Also, any material or simple eBook on web services.
my email is [email protected]
Thanks a ton in advance.
Swetha.

Check this link ABAP Development Performance Tuning
https://www.sdn.sap.com/irj/sdn/wiki?path=/display/abap/performance%2btuning
Check these threads.
How do you take care of performance issues in your ABAP programs?
http://help.sap.com/saphelp_nw04/helpdata/en/9f/db9bd335c111d1829f0000e829fbfe/frameset.htm

Performance tuning SELECTs: Can I force a TRUE db read every time?

Good day everyone!
I've been programming in ABAP for almost 6 years now, and I'd like to begin learning more about performance tuning with respect to database performance.
More specifically, I'm testing some things in a particular SELECT in a report program we have that is timing out in the foreground because of the SELECT. When I first run the program, the SELECT goes against the database, as we all know. Subsequent runs, however, use the buffered data, so the response is a lot quicker and doesn't really reflect that first, initial database read.
Am I correct in assuming that I should be testing my various approaches and collecting performance runtimes against that initial, "true" database read? If that's the case, is there any way I can force the system to actually read the database instead of the buffered data? For those experienced with this kind of performance analysis and tuning, what's the best approach for someone very new to this area such as myself?
Thank you,
Dave

Hi Dave and Rob,
Just my two cents (and yes, I know this is already answered, but..).
I think you might be confusing 2 things: one is SAP buffering, and another one is caching at other levels (database, operating system, etc).
From what I understood Rob was talking mainly about SAP buffering. In that context it is true that if there is a first execution that loads the buffers (for example, some not so small fully buffered tables) then that is an atypical execution, and should be discarded. In real life you will never have execution times like those, except maybe on the very first execution on a monday morning.
Another thing is database caching. If you execute a report twice with exactly the same parameters then you might not be actually making physical reads in the second execution. This second execution will be very fast, but that will not be simulating real life: no user wants a report to be fast the second time you execute it with exactly the same parameters.
To avoid this in Oracle you can empty the so-called SGA, but that is not so useful and it will probably not get you closer to what happens in the real life.
So what to do? In doubt, measure it several times, with different parameters, and probably exclude the extreme values.
Regards,
Rui Dantas

Report running for long time & performance tuning

Hi All,
(1). WebI report is running for long time.so what are the steps i need to check for it ?
(2). Can you tell me about performance tuning in BO ?
please help me.....
Thanks
Kumar

(1). WebI report is running for long time.so what are the steps i need to check for it ?
The first step is to see if the problem lies in the query on the data source or in webi itself. Depending on the data source there are different ways to extract the query and try to run it against the database. Which source does your report uses?
(2). Can you tell me about performance tuning in BO ?
I would recommend to start by reading the administrator's guide. There is a section about how to improve performance.
Regards,
Stratos

Problem in performing multiple Point-In-Time Database Recovery using RMAN

Hello Experts,
I am getting an error while performing database point in time recovery multiple times using RMAN. Details are as follows :-
Environment:
Oracle 11g, ASM,
Database DiskGroups : DG_DATA (Data files), DG_ARCH(Archive logs), DG_REDO(Redo logs Control file).
Snapshot DiskGroups :
Snapshot1 (taken at 9 am): SNAP1_DATA, SNAP1_ARCH, +SNAP1_REDO
Snapshot2 (taken at 10 am): SNAP2_DATA, SNAP2_ARCH, +SNAP2_REDO
Steps performed for point in time recovery:
1. Restore control file from snapshot 2.
     RMAN> RESTORE CONTROLFILE from '+SNAP2_REDO/orcl/CONTROLFILE/Current.256.777398261';
2. For 2nd recovery, reset incarnation of database to snapshot 2 incarnation (Say 2).
3. Catalog data files from snapshot 1.
4. Catalog archive logs from snapshot 2.
5. Perform point in time recovery till given time.
     STARTUP MOUNT;
     RUN {
          SQL "ALTER SESSION SET NLS_DATE_FORMAT = ''dd-mon-yyyy hh24:mi:ss''";
          SET UNTIL TIME "06-mar-2013 09:30:00";
          RESTORE DATABASE;
          RECOVER DATABASE;
          ALTER DATABASE OPEN RESETLOGS;
Results:
Recovery 1: At 10.30 am, I performed first point in time recovery till 9:30 am, it was successful. Database incarnation was raised from *2* to *3*.
Recovery 2: At 11:10 am, I performed another point in time recovery till 9:45 am, while doing it I reset the incarnation of DB to *2*, it failed with following error :-
Starting recover at 28-FEB-13
using channel ORA_DISK_1
starting media recovery
media recovery failed
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of recover command at 03/06/2013 11:10:57
ORA-00283: recovery session canceled due to errors
RMAN-11003: failure during parse/execution of SQL statement: alter database recover if needed
start until time 'MAR 06 2013 09:45:00'
ORA-00283: recovery session canceled due to errors
ORA-00313: open failed for members of log group 1 of thread 1
ORA-00312: online log 1 thread 1: '+DG_REDO/orcl/onlinelog/group_1.257.807150859'
ORA-17503: ksfdopn:2 Failed to open file +DG_REDO/orcl/onlinelog/group_1.257.807150859
ORA-15012: ASM file '+DG_REDO/orcl/onlinelog/group_1.257.807150859' does not exist
Doubts:
1. Why did recovery failed 2nd time, but not 1st time and why is RMAN looking for online redo log group_1.257.807150859 in 2nd recovery ?
3. I tried restoring control file from AutoBackup, in that case both 1st and 2nd recovery succeeded.
However for this to work, I always need to keep the AutoBackup feature enabled.
How reliable is control file AutoBackup ? Is there any alternative to using AutoBackup, can I restore control file from snapshot backup only ?
4. If I restore control file from AutoBackup, then from what point of time/SCN does RMAN restores the control file ?
Please help me out in this issue.
Thanks.

992748 wrote:
Hello experts,
I'm little newbie to RMAN recovery. Please help me in these doubts:
1. If I have datafiles, archive logs & control files backup, but current online REDO logs are lost, then can I perform incomplete database recovery ?yes, if you have backups of everything else
2. Till what maximum time/scn can incomplete database recovery be performed ??Assuming the only thing lost is the redo logs, you can recover to the last scn in the last archivelog.
3. What is role of online REDO logs in incomplete database recovery ? They provide the final redo changes - the ones that have not been written to archivelogs
Are they required for incomplete recovery ?It depends on how much incomplete recovery you need to do.
Think of all of your changes as a constant stream of redo information. As a redolog fills, it is copied to archive, then (eventually) reused. over time, your redo stream is in archivelog_1, continuing into archvivelog_2, then to 3, and eventually, when you get to the last archivelog, into the online redo. A recovery will start with the oldest necessary point in the redo stream and continue forward. Whether or not you need the online redo for a PIT recovery depends on how far forward you need to recover.
But you should take every precaution to prevent loss of online redo logs .. starting with having multiple members in each redo group ... and keeping those multiple members on physically separate disks.

Which approach is having better performance in terms of time

For large no of data from more then two diffrent tables which are having relations ,
In Oracle in following two approaches which is having better performance in terms of time( i.e which is having less time) ?
1. A single compex query
2. Bunch of simple queries

Because their is a relationship between each of the tables in the simple queries then if you adopt this approach you will have to JOIN in some way, probably via a FOR LOOP in PL/SQL.
In my experience, a single complex SQL statement is the best way to go, join in the database and return the set of data required.
SQL rules!

How to display gauge progress refer to HTTP connection time taken

Hi,
I would like to know how can I display the gauge progress depend on the HTTP connection time taken.
I have the source code below but it can't show the gauge progress parallel to the HTTP connection working.
That's kind of those who solve me this problem.
Source code:
import javax.microedition.midlet.*;
import javax.microedition.lcdui.*;
import java.io.*;
import javax.microedition.io.*;
import java.util.*;
public class HttpExample extends MIDlet implements CommandListener
     private Gauge gauge;
     private Command exit, start;
     private Display display;
     private Form form;
     long AUD, CAN;
     public HttpExample()
          display = Display.getDisplay(this);
          exit = new Command("Exit", Command.EXIT, 1);
          start = new Command("Start", Command.EXIT, 1);
          form = new Form("");
          gauge = new Gauge("Updating Currency Rate", false, 100, 0);
          form.append(gauge);
          form.addCommand(exit);
          form.addCommand(start);
          form.setCommandListener(this);
          display.setCurrent(form);
     public void startApp() throws MIDletStateChangeException
          display.setCurrent(form);
     public void pauseApp()
     public void destroyApp(boolean unconditional)
     public void commandAction(Command command, Displayable displayable)
          if (command==exit)
               destroyApp(false);
               notifyDestroyed();
          else if(command==start)
               Updater updater = new Updater(this);
               updater.start();
     class Updater implements Runnable
     private HttpExample MIDlet;
     public Updater(HttpExample MIDlet)
          this.MIDlet=MIDlet;
     public void run()
          try
               transmit();
          catch (Exception error)
               System.err.println(error.toString());
     public void start()
          Thread thread = new Thread(this);
          try
               thread.start();
          catch (Exception error)
     private void transmit() throws IOException
          StreamConnection connection=null;
          InputStream in= null;
          StringBuffer buffer=new StringBuffer();
          try
               connection=(StreamConnection) Connector.open("http://www.geocities.com/cyfan118/index.html");
               in=connection.openInputStream();
               int ch;
               while ((ch=in.read())!=-1)
                    gauge.setValue(gauge.getValue() + 1);
                    buffer.append((char)ch);
               String line=new String(buffer.toString());
               int p=line.indexOf("AUD");
               int p1=line.indexOf("CAN");
               String a=line.substring(p+4,p+7);
               String b=line.substring(p1+4,p1+7);
               AUD=Long.parseLong(a);
               CAN=Long.parseLong(b);
               buffer=new StringBuffer();
          catch (IOException error)
               Alert alert=new Alert("Error","Cannot connect",null,null);
               alert.setTimeout(Alert.FOREVER);
               alert.setType(AlertType.ERROR);
               display.setCurrent(alert);
}

This code looks like it should work to update the progress during the connection but it is not going to start updating the guage until it has made the connection and is downloading data. If you want to show something on the screen during the time the program is doing Connector.open() you are going to need to have the update code in yet another thread or timer.

HTTP Access time

Hi,
I have a HTTP monitor that calculates HTTP access time, ie the time between SYN (send by client) and HTTP 200 OK (recieved by client from server). The value calculated by the HttpMonitor class differs from what's shown in Ethereal. On low responsetimes, it takes ~140ms from creating the new socket() in JAVA til the outgoing packet (SYN) is cacptured in Ethereal. Note that the measuring are started after the new Socket() method. Is it the native connection "socketConnect" thats slow or is there some buffering going on somewhere?
Also, if the recieved datapacket is rather small (say 300byte) , Ethereal and my JAVA class gets the HTTP 200 OK at the same time (differs by some millisec), but if the packet is like 1514byte Ethereal recieves the HTTP 200 OK like 50ms before my java class, even though I change the buffersize to 300byte.
CODE TO SEND SYN:
InetAddress[] addr_list = InetAddress.getAllByName(actual_host);
for (int idx = 0; idx < addr_list.length; idx++) {
try {
if (LocalAddr == null) sock = new Socket(addr_list[idx], actual_port);
else sock = new Socket(addr_list[idx], actual_port, LocalAddr, LocalPort);
break; // success
} catch (SocketException se) {
if (idx == addr_list.length - 1) throw se; // we tried them all
httpSocketsCreatedTime = System.currentTimeMillis();
CODE TO RECIEVE HTTP 200 OK:
//using an instace to BufferedInputStream below we use the method read(...) in order to get the stream
//ie. first we read 7 bytes to determine the reponse code. (ie. HTTP 200 OK). len = 7, off = 0
BufferedInputStream stream = ....;
int rcvd = stream.read(b, off, len);
httpAccessTime = System.currentTimeMillis();
class BufferedInputStream extends FilterInputStream {
private byte[] buffer = new byte[300]; //Normally I have the buffer set to 2000
public int read(byte[] buf, int off, int len) throws IOException {
int got = in.read(buffer, pos, buffer.length - pos); // InputStream in
Basically there is nothing special with the code above...
Thanks in advance
Martin

Hmm, any change if you disable Nagle by socket.setTcpNoDelay(true)?

What happend if we keep http connection time out as 0

what happend if we define http connection time out as 0 or less than 1 ..pls give reply.
i am seriously struck up with some problem
depending on this our enhancement is depended

Are you still facing the same issue ? This majorly relates to network problem or probably some issue on server end , have you tried to contact your host ?
Thanks,
Sanjit

Performance tuning in EP6 SP14

Hi,
We just migrated our development EP5 SP6 portal to NW04 EP6 SP14 and are seeing some performance problems with a limited number of users (about 3).
Please point me to some good clear documentation related to performance tuning. Better yet, please tell me some things you have done in your EP6 portal to improve performance.
Any help is greatly appreciated.
Regards,
Rick

Hi
In general we have experienced that NW04 is faster than earlier versions.
One of the big questions is if the response time is SERVER time (used for processing on the J2EE server), NETWORK time (latency and many roundtrips), CLIENT time (rendering, virus-scanning etc) or a combination.
1) Is the capacity on the server ok? CPU utilization and queue lenght high? Memory swapping?
2) A quick optimazation server-side is logging: Plese verify that that log and trace levels are ERROR /FATAL or NONE on J2EE and also avoid logging on IIS and IIS proxy is used
3) Using a HTTP trace you can see if the NW portal for some reason generetes more roundtrips (more GETS) that the old portal for the same content - what is the network latency? Try to do a "ping <server>" from a client pc and see the time (latency) - if it is below 20 ms the network should do a big difference.
4) On the client try to call the portal with anti-virus de-activated if the delta/difference in response times are HUGE your client could be to old? (don't underestimate the client). Maybee the compression should be set different to avoid compres (server) and uncompres (client) - this is a tradeoff with network latency however.
Also client-side caching (in browser) is important.
These are just QUICK point to consider - tuning is complex
BR
Tom Bo

Performance Tuning: Http Task Time Response Time Threshold

Similar Messages

Maybe you are looking for