Analyzing optimization exceptions for index databases

Analyze the optimization exceptions that were detected for index databases.

Before you begin

Ensure that you can access the Sensor Data Statistics report that contains the latest sensor data. If you cannot access this report, run the FF DB Sensor Printing utility to generate the report.

About this task

If a secondary index database has an overflow data set defined, duplicated index pointer segments are located in the overflow data set. Accessing the index pointer segments in the overflow data set requires extra I/O operations. Consequently, when an excessive number of index pointer segments exist in the overflow data set, the condition might cause performance degradation.

Optimizing secondary index databases typically includes tuning the database definitions to decrease the number of index pointer segments in the overflow data set.

In this task, you will identify the number of index pointer segments in the overflow data set, and determine whether you can reduce these segments.

For HIDAM primary indexes, HALDB partitioned secondary indexes, and secondary index databases without an overflow data set, you do not need to analyze duplicated keys.

Procedure

In this section, instructions to analyze the common optimization exceptions are provided. Follow the instructions for the exceptions that you received.
Exception class name: NUMBER_OF_INDEX_POINTER_SEGMENTS
This exception indicates that the number of index pointer segments (IPS) is high.
This exception does not always indicate a problem. When many index pointer segments exist in the overflow data set, determine whether you need to take an action to resolve this exception. Follow the analysis steps for the EXCESSIVE_INDEX_OVERFLOW exception to determine if the secondary index database needs to be worked on.
Exception class name: EXCESSIVE_INDEX_OVERFLOW
This exception indicates that many index keys are duplicated. This exception is reported for secondary indexes that have an overflow data set.
When index keys are duplicated, only the index pointer segment (IPS) that is generated first is placed in the primary data set. Other IPSs that have the same key are placed in the overflow data set. IPSs that are placed in overflow data sets require extra I/O operations to be read.
Complete these steps to analyze this exception:
  1. Locate the following data elements in the Sensor Data Statistics report:
    • DBX_NUM_IPS
    • DBX_NUM_IPS_OVFL
    • DBX_PCT_IPS_OVFL
    The values for these data elements indicate the numbers of IPSs, IPSs in the overflow data set, and the percentage of IPSs in the overflow data set compared to the total number of IPSs. Analyze these values to determine why you received this exception.
  2. Optional: To obtain further information about the index database or the indexed database, run the HD Pointer Checker utility of IMS HP Pointer Checker, and generate the following reports:
    Scan of Index Database report
    You can obtain various information about index databases such as a summary description of data sets (VSAM KSDS and ESDS) of the index database and the total number of records (of various types) that were involved in the processing of the index database.
    Index Segment Counts Statistics report
    You can obtain information about index counts and indexed databases such as the index database and indexed database names, source and target segment names, number of index pointer and source segments, and the number of suppressed segments.
    This report is contained in the Evaluation of All Pointers to the Same Target report.

What to do next

If you received other exceptions that belong to other exception categories, see the following topics to determine your next step:

If the only exceptions that you received are optimization exceptions, continue with Resolving optimization exceptions for index databases.