IBM Support

JR44092: PACK FOR SAP R/3, ABAP EXTRACT: IMPROVE EXTRACT PERFORMANCE AND TRANSACTION STABILITY

Subscribe to this APAR

By subscribing, you receive periodic emails alerting you to the status of the APAR, along with a link to the fix after it becomes available. You can track this item individually or track all items by product.

Notify me when this APAR changes.

Notify me when an APAR for this component changes.

 

APAR status

  • Closed as program error.

Error description

  • The ABAP program currently commits work once at the end of the
    extraction and sends the records in packages of 3000; the
    proposed enhancement would send the records in larger packages
    (e.g. 100000) and commit work after every package by
    use of an environment variable which will modify the package
    size being used in a set of jobs.
    Changing the parameter will require restarting the Designer and
    regenerating jobs.
    

Local fix

  • Use the existing ABAP code
    

Problem summary

  • The ABAP program currently commits work once at the end of the
    extraction and sends the records in packages of 3000.
    Enhancement request: Change the ABAP stage ABAP code generator
    towards a configurable package size with intermediate commits.
    This is intended to help improve extract performance and
    transaction stability in certain scenarios.
    

Problem conclusion

  • Added additional functionality to the ABAP code generator on the
     DS Client. The additional functionality is only available for
    ABAP programs designed by using the SQL Builder and with RFC set
    as the data transfer method.
    When this additional behavior is switched on, the client will
    generate the ABAP code as required.
    This new functionality is switched on through the environment
    variable DSR3_COMMIT_PACKAGE_SIZE which can be set to a numeric
    value (the package size).
    
    IMPORTANT NOTE: switching on intermediate commits generates
    increased transfer speed (approximately 65% improvement; for
    example
    job execution time decreased from 52 mins to 34 mins*) but
    requires significantly more memory for the extraction on the SAP
    Server side.
    The requirement of available memory is roughly twice the total
    size of data extracted. For example extracting 2 million rows
    with row size
    1kB would require roughly 2 GB available memory. If the SAP
    Server cannot accommodate the memory requirement, the extract
    will fail.
    There is currently no solution to the excessive memory
    consumption issue.
    The performance and requirement numbers above were obtained on a
    4 node parallel configuration, with the commit package size set
    to 30,000.
    Your numbers will vary based on your actual test conditions.
    The fix is available as a patch on version 6.5.0.2 and 7.0; it
    is included in GA 7.1
    

Temporary fix

Comments

APAR Information

  • APAR number

    JR44092

  • Reported component name

    IS PACK FOR SAP

  • Reported component ID

    5724Q5500

  • Reported release

    650

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt

  • Submitted date

    2012-09-14

  • Closed date

    2014-07-08

  • Last modified date

    2014-07-08

  • APAR is sysrouted FROM one or more of the following:

  • APAR is sysrouted TO one or more of the following:

Fix information

  • Fixed component name

    IS PACK FOR SAP

  • Fixed component ID

    5724Q5500

Applicable component levels

  • R650 PSY

       UP

  • R700 PSY

       UP

[{"Business Unit":{"code":"BU048","label":"IBM Software"},"Product":{"code":"SSCSRJX","label":"DataStage Pack for SAP R\/3"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"6.5.0","Edition":"","Line of Business":{"code":"","label":""}}]

Document Information

Modified date:
08 July 2014