IBM Support

PM39514: Fix async replication where CPU utilization grows as the number of entries increase.

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • A replication query is used to find entries in the cache that
    have changed since the last replication event.  That query was
    O(N) in complexity, it scanned all entries in the partition.
    This results in longer query times as the number of entries
    increases which also causes higher CPU utilization.
    

Local fix

  • N/A
    

Problem summary

  • ****************************************************************
    * USERS AFFECTED:  All users of asynchronous replication.      *
    ****************************************************************
    * PROBLEM DESCRIPTION: When using                              *
    *                      asynchronous replication,               *
    *                      proessor utilization                    *
    *                      grows as the                            *
    *                      number of entries increases.            *
    ****************************************************************
    * RECOMMENDATION:                                              *
    ****************************************************************
    The asynchronous replication path includes both
    intra-domain asynchronous replication and multi-master primary
    to primary
    replication.  When these paths are used, processor  utilization
    grows as the number of entries increases and there are changes
    to be replicated.
    The problem is that the algorithm to calculate the changes to
    send searches for all entries within a window.  The current
    implementation needs to look at every entry to find all of the
    changes within the window.
    

Problem conclusion

  • A new algorithm that searches for entries in constant
    time rather than linear time was implemented to improve
    performance.
    

Temporary fix

Comments

APAR Information

  • APAR number

    PM39514

  • Reported component name

    XD EXTREME SCAL

  • Reported component ID

    5724J3402

  • Reported release

    710

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt

  • Submitted date

    2011-05-18

  • Closed date

    2011-07-27

  • Last modified date

    2011-07-27

  • 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

    XD EXTREME SCAL

  • Fixed component ID

    5724J3402

Applicable component levels

  • R700 PSY

       UP

[{"Business Unit":{"code":"BU053","label":"Cloud & Data Platform"},"Product":{"code":"SSTVLU","label":"WebSphere eXtreme Scale"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"710","Edition":"","Line of Business":{"code":"LOB45","label":"Automation"}}]

Document Information

Modified date:
23 September 2020