Skip to Content
avatar image
Former Member


Hi all,

What is ST01, 02, 03, 04, and 06 how do you monitor?



Add comment
10|10000 characters needed characters exceeded

  • Follow
  • Get RSS Feed

3 Answers

  • Best Answer
    avatar image
    Former Member
    Jul 03, 2007 at 11:10 AM
    Add comment
    10|10000 characters needed characters exceeded

  • avatar image
    Former Member
    Jul 03, 2007 at 12:15 PM


    ST01- System Trace.

    Using this we can do user trace, system trace on a particular date and time.

    Trace On / Trace Off.

    We can trace database. And we can check trace file for further analysis.

    Tune summary, buffer statistics (ST02)

    Menu Path: Tools>Administration>Monitoring>Performance>Setup/Buffers>Buffers.

    In this

    Important statistics:

    High buffer hit ratios (>95%)

    No buffer swaps

    Buffer freespace >>0

    Buffered objects should have status “valid”

    No or little heap memory allocation


    Workload Analysis (ST03)

    Menu Path: Tools>Administration>Monitoring>Performance>Workload>Analysis> Detail…>Workload.

    Note wait times for different types of work processes, variations by instance, high DB request times for particular transactions, substantial fraction of average response time not accounted for by sum of components. Then view by Top Time transactions, Time profile; become familiar with normal distribution of transactions each case for each type.

    Workload analysis is the general and systematic approach to R/3 performance problem resolution. Usage requires an understanding of R/3 architecture, operating system, server and database performance factors.


    Database Performance Overview

    Menu Path: Tools>Administration>Computing Center>Management System>Control>Performace Menu>Database>Activity.

    Important statistics:

    Data Buffer: Quality should be > 95%

    Ratio of Data buffer: Reads/Data buffer: Physical reads should be > 10:1

    Data buffer: Busy wait time should be low

    Data buffer: Busy waits/Data Buffer: read should be < 5%

    Shared Pool: DD-Cache Quality should be > 90%

    Log buffer: Allocation retries/Log buffer: entries should be < 1%

    Calls: Recursive calls/Calls; User calls should be < 10%

    Calls: Rollbacks should be few

    Table scans: Long tables/Calls: User calls should be < 1%

    No disk sorts


    Menu Path: Tools>Administration>Computing Center>Management System>Control>Performance Menu>Operating system>local>Activity

    Indicates overall picture of the hardware and the component subsystems, for example, CPU, memory, and disk information. A high load average and/or paging rate indicate a potential performance bottle neck.

    Important statistics:

    CPU utilization (<80%)

    Load average (<=3)

    Memory utilization and paging out rates (-0 pages/second)

    Disk utilization and access times

    Network utilization (packets/second versus available bandwidth) This assumes that your network is being monitored via R/3.

    I hope this information will help you.

    Note: Points always encourage me to reply !!

    Add comment
    10|10000 characters needed characters exceeded

  • avatar image
    Former Member
    Jul 09, 2007 at 09:26 AM

    Dear All,

    Thanks for Updating the Useful Information . any further material or important points available interview Point of View.

    One more Query asked in Interview.

    i have execute a Background Job it has completed 1500 Ms.. next time i have executed the same program it has taken 2500Ms what could be the Reason.



    Add comment
    10|10000 characters needed characters exceeded

    • Former Member


      Please create new thread for this new question.

      It is the forum rule under one thread , there is only one question.

      Anyway i am giving reply for this.

      1) You can start by analyzing SM50 ont he application server , where the job is running.

      From there, double-click the process running the job to get additional detail regarding the table(s) being accessed or updated.

      And also, you can dig a little deeper,

      1)Debug the process in sm50 ( Program/Mode>Program>Debug)

      2)Trace the b/g job using st05 and evaluate the Select/Update/Insert statements and the calling ABAP statements.

      3)Check the DB performance in ST04-> Detailed Analysis Menu.

      And also,

      you can run STAT/STAD to find out where do you have big response times and you can do ABAP runtime analysis ie SE30.

      Note: Points always encourage me to reply !!