15.3.1 Condition analysis by RDAREA (logical analysis)

This utility analyzes the storage condition of all segments and pages for the tables and indexes that are stored in an RDAREA. Note, however, that for tables and indexes that have been partitioned and stored among several RDAREAs, this utility analyzes only those segments stored in the specified RDAREAs.

Organization of this subsection
(1) Purpose
(2) Analysis results
(3) Analyzing the analysis results

(1) Purpose

You can detect a disordered table or index and obtain the capacity status of a single RDAREA.

(2) Analysis results

The following shows the results of condition analysis by RDAREA (logical analysis):

(a) For an RDAREA other than a LOB RDAREA

pddbst VV-RR(Object Option) ** RD Area Logical Analysis **  2008/12/25 15:54:10 [1]
RD Area Name   : user_rdarea_1 [2]
 Server        : bes01 [3]
 Total Segment :        150[4] Segment Size :          5 Pages [6]
 Unused Segment:           9[5] Page Size    :       4096 Bytes [7]
 Original RD Area Name : user_rdarea_1 [21]
 Generation Number :  0[22]  Replica RD Area Count :  0 [23]
 [24]     [25]               [26]               [27]
 History1 Hold Status :      Hold Code :     0  Hold Time :
 History2 Hold Status :      Hold Code :     0  Hold Time :
-------------------------------------------------------------------------------

Table Name : TBL01 [8]
Auth Id    : user1 [9]
Status     :   [20]
Reference Pending Status :   [33]
Check     Pending Status :   [34]
Segment Reuse :          - segments [29]
Segment Reuse Option :          1
 <Base row segment>
 Search Mode : INS [28]
 Branch Row Search Mode :       - [58]
 Reuse Search Failure :          0/         0 [30]
             Used(Full)       Used(      Full)        Sum
             [10]  [11]        [12]       [13]        [14]
 Segment [39]100%(  0%)        107(         0)        107
             [15]  [16]        [17]       [18]        [19]
 Page [40]   100%(  0%)        535(         0)        535
 Collect On Segment :          0 [53]
 Collect Prearranged Page :          0 [31]
 Collect Prearranged Full Page :          0 [59]
 Used Page Ratio       Page(Ratio) [32]
            0% :          0(   0%)
        1- 10% :          1(   1%)
       11- 20% :          0(   0%)
       21- 30% :          0(   0%)
       31- 40% :          0(   0%)
       41- 50% :          0(   0%)
       51- 60% :          0(   0%)
       61- 70% :        118(  22%)
       71- 80% :        416(  78%)
       81- 90% :          0(   0%)
       91-100% :          0(   0%)
 Total                  535
 Del RowID   :                     0 [60]

Table Name : TBL02
Auth Id    : user1
Status     :
Reference Pending Status :
Check     Pending Status :
Segment Reuse :          - segments
Segment Reuse Option :          - [61]
 <Base row segment>
 Search Mode : INS
 Branch Row Search Mode : -
 Reuse Search Failure :          0/         0
         Used(Full)       Used(      Full)        Sum
 Segment 100%(  0%)          1(         0)          1
 Page     60%(  0%)          3(         0)          5
 Collect On Segment :          0
 Collect Prearranged Page :          0
 Collect Prearranged Full Page :          0
 Used Page Ratio       Page(Ratio)
            0% :          2(  40%)
        1- 10% :          1(  20%)
       11- 20% :          1(  20%)
       21- 30% :          0(   0%)
       31- 40% :          1(  20%)
       41- 50% :          0(   0%)
       51- 60% :          0(   0%)
       61- 70% :          0(   0%)
       71- 80% :          0(   0%)
       81- 90% :          0(   0%)
       91-100% :          0(   0%)
 Total                    5
 Del RowID   :                     0
 <BINARY segment>
 Search Mode : INS
 Branch Row Search Mode : -
 Reuse Search Failure :          0/         0
             Used(Full)       Used(      Full)        Sum
             [41]  [42]        [43]       [44]        [45]
 Segment [46]100%( 93%)         14(        13)         14
             [47]  [48]        [49]       [50]        [51]
 Page  [52]  96%( 95%)         67(        66)         70
 Collect On Segment :          0
 Collect Prearranged Page :          0
 Collect Prearranged Full Page :          0
 Used Page Ratio       Page(Ratio)
            0% :          3(   5%)
        1- 10% :          0(   0%)
       11- 20% :          0(   0%)
       21- 30% :          0(   0%)
       31- 40% :          0(   0%)
       41- 50% :          0(   0%)
       51- 60% :          0(   0%)
       61- 70% :          0(   0%)
       71- 80% :          0(   0%)
       81- 90% :          0(   0%)
       91-100% :         67(  96%)
 Total                   70

Index Name : IDX_TBL01_C1
Auth Id    : user1
Status     :
         Used(Full)       Used(      Full)        Sum
 Segment 100%(  0%)         19(         0)         19
 Page    100%(  0%)         95(         0)         95
 Collect On Segment :          0
 Collect Prearranged Page :          0
 Collect Prearranged Full Page :          0
 Collect On Page :          0 [54]
 Used Page Ratio       Page(Ratio) Del RowID [35]
            0% :          0(   0%)         0
        1- 10% :          0(   0%)         0
       11- 20% :          0(   0%)         0
       21- 30% :          1(   1%)         0
       31- 40% :          1(   1%)         0
       41- 50% :          0(   0%)         0
       51- 60% :          0(   0%)         0
       61- 70% :         93(  98%)         0
       71- 80% :          0(   0%)         0
       81- 90% :          0(   0%)         0
       91-100% :          0(   0%)         0
 Total                   95                0 [36]
 Del RowID   :                     0     (  0%) [37]
 RowID Count :                10,000 [38]

Sequence Name : SEQ1 [55]
Auth Id       : user1 [56]
 Current Value :                               10000001​ [57]

Explanation:
When there are multiple RDAREAs to be processed, the information is displayed in the following order:
  • RDAREA names (ascending order)
  • Table names (ascending order)
  • Index names (ascending order)
  • Sequence generator identifiers (ascending order)
  1. Date and time of completion of acquisition of the information required for this condition analysis, in the format YYYY/MM/DD hh:mm:ss:
    YYYY: Year. MM: Month. DD: Date. hh: Hour. mm: Minute. ss: Second.
  2. Name of RDAREA subject to analysis.
  3. Name of server managing the RDAREA.
  4. Total number of segments in the RDAREA (number of used segments + number of unused segments).
  5. Total number of unused segments in the RDAREA.
  6. Segment size (number of pages per segment).
  7. Page size (in bytes).
  8. Name of a table or index contained in the RDAREA.
  9. Authorization identifier of the table or index. For a data dictionary RDAREA, (Data dictionary) is displayed.
  10. Percentage ratio of used segments:
    [Figure]number of used segments/value of 14[Figure][Figure] 100 (%)
  11. Percentage ratio of full segments:
    [Figure]number of full segments/value of 14[Figure][Figure] 100 (%)
  12. Number of used segments. Segments being released by pdreclaim are displayed as used segments.
  13. Number of full segments.
  14. Total number of segments allocated to the table or index (number of used segments + number of unused segments).
  15. Percentage ratio of used pages:
    [Figure]number of used pages/value of 19[Figure][Figure] 100 (%)
  16. Percentage ratio of full pages:
    [Figure]number of full pages/value of 19[Figure][Figure] 100 (%)
  17. Number of used pages.
  18. Number of full pages.
  19. Total number of pages in the segments allocated to the table or index (number of used pages + number of unused pages).
  20. For a table, reload-not-completed data status or being reorganized. If the table is not in either of the following statuses, this field is blank.
    DATA_UNFINISH:
    The table is in reload-not-completed data status, which means that table reorganization was executed, but reload processing has not been completed for some reason such as an error. To release the table from reload-not-completed data status, you must re-execute table reorganization or reload data into the table.
    ON_RORG:
    The table is currently being reorganized.
    For an index, this is the status of an unfinished index. If the index has been completed, this field is blank.
    UNFINISH_0
    See Figure 15-9 Types of unfinished index by table (1/2) and Figure 15-10 Types of unfinished index by table (2/2).
    UNFINISH_1
    See Figure 15-11 Types of unfinished index by RDAREA (1/2) and Figure 15-12 Types of unfinished index by RDAREA (2/2).
    UNFINISH_2
    When a UAP using the plug-in index delayed batch creation function is being executed or after the UAP finished executing, a batch index creation process using the database reorganization utility is not being executed. For details about the plug-in index delayed batch creation function, see the HiRDB Version 9 System Operation Guide.
  21. Name of the original RDAREA.
  22. Generation number of the corresponding RDAREA.
  23. Number of replica RDAREAs.
  24. Shutdown history of the corresponding RDAREA.
    History1: Shutdown information immediately before the shutdown was released (if the RDAREA is currently shut down, this field displays the current shutdown information)
    History2: Shutdown information immediately preceding History1
  25. Shutdown type of the corresponding RDAREA. If there is no history, this field is blank.
    CMD: Command shutdown by HiRDB error detection
    FLT: Error shutdown
  26. Code indicating the cause of the shutdown of the corresponding RDAREA. If there is no history, this field displays 0.
    0: Shutdown release
    10: Input/output error
    20: Page corruption
    40: Open or allocation error
    70: Rollback error
    80: Shutdown due to an error during execution of DROP TABLE or DROP INDEX
    90: No-log shutdown (pdload, pdrorg, and pdrbal commands)
    92: No-log shutdown (UAP)
    93: Shutdown of LOB RDAREA
    96: Invalid time stamp
    97: Invalid object ID
    98: List RDAREA error
    100: Log application-disabled status due to execution of a utility (pdload, pdrorg, or pdrbal) in a mode other than the log acquisition mode
    102: Log application-disabled status due to execution of a UAP in a mode other than the log acquisition mode
    104: Log application-disabled status due to updating of a table for which PARTIAL or NO is specified for the BLOB column recovery restriction
    110: Log application-disabled status due to pdmod
    120: Log application-disabled status due to error shutdown
  27. Shutdown time of the corresponding RDAREA. If there is no history, this field is blank.
  28. Page search mode:
    INS: New page allocate mode
    REU: Free page reuse mode
  29. Number of segments specified for SEGMENT REUSE during execution of CREATE TABLE or ALTER TABLE. If a number of segments for SEGMENT REUSE is not specified, this field displays 0. If NO is specified for SEGMENT REUSE or SEGMENT REUSE is omitted, the field displays -.
  30. Number of times the mode changed from new page allocate to free page reuse and then changed back to new page allocate because there was no reusable free space (n/m).
    n: Idle count
    m: Number of times mode changed from new page allocate to free page reuse
    n and m are reset to 0 at the following times:
    • When the RDAREA is closed
    • When HiRDB is restarted
    • When the PURGE TABLE statement is executed
  31. Number of pages that can be released by pdreclaim (number of used free pages whose usage ratio is 0%).
  32. Number of used pages for each 10% range. The parentheses enclose the percentage ratio to all used pages. Because the fraction part of each ratio is rounded up, the total may exceed 100. Note the following:
    • For a page that stores 201 or more duplicate keys, the percentage ratio of PCTFREE specified during table definition is ignored. Therefore, data is not rearranged according to the PCTFREE value even during reorganization.
    • For an index to which the index key value with no lock is applied, the deletion key value is treated as free space. If all used pages are deletion key values, the usage ratio is 0%.
    • A full page may not always fall in the usage ratio range of 91 to 100%.
    • A free space created by update processing that shortens an existing row is not treated as free space.
    • If the ratio of used pages is less than (100 - free space ratio specified with PCTFREE), the storage efficiency may have decreased. You should consider executing pdrorg. For an area storing branch rows (variable-length character string data with a length of 256 bytes or greater, data for repetition columns, and data for columns of an abstract data type), executing pdrorg has no effect on the storage efficiency.
  33. Check pending status for referential constraint in the table information in the RDAREA. If no referential constraint has been defined for the table, this item is blank.
    P: Check pending status
    Blank: Not check pending status
  34. Check pending status for check constraint in the table information in the RDAREA. If no check constraint has been defined for the table, this item is blank.
    P: Check pending status
    Blank: Not check pending status
  35. Numeric values for the items shown below, when index key values with no locks are used or uncommitted deletion data is locked.
    • Number of rows managed by remaining entries for each indicated range of percentages of used index pages
      (Remaining entries are index entries that still exist in the index due to lock control even after their keys have been deleted as a result of updating or deletion of index key values).
    The utility displays 3 digits, and the maximum length is 5 digits when a decimal point and units character are included. The decimal point and units character are displayed when the number of rows to be displayed requires more than 3 digits. In such a case, rounding occurs at the 4th digit. The table below shows the display format for the number of rows managed by remaining entries:
    Range of numeric valuesDisplay format
    0 to 9990 to 999
    1,000 to 999,4991.00K to 999K
    999,500 to 999,499,9991.00M to 999M
    999,500,000 to 999,499,999,9991.00G to 999G
    999,500,000,000 to 999,499,999,999,9991.00T to 999T
    Units characters:
    K: Kilo, M: Mega, G: Giga, T: Tera
  36. Total number of used pages and total number of rows managed by remaining entries for all percentage ranges.
  37. Number and percentage of managed rows that are in index deleted status (the percentage is of the total number of managed rows).
  38. Total number of rows managed by the index. The utility displays an integer of up to 15 digits in length. If the value consists of more than 15 digits, the utility displays it as an exponent (for example, 1000000000000000 would be displayed as 1E+15). The table below shows the display formats:
    Range of numeric valuesDisplay format
    0 to 9990 to 999
    1000 to 99991,000 to 9,999
    100000000000000​ to 999999999999999​100,000,000,000,000 to 999,999,999,999,999
    1000000000000000​ and greaterz,zzz,zzz,zzz,zzz,zz9E+zz9
    (exponent format with 16 significant digits)
  39. Summary information for segments. This does not include information for binary-only segments.
  40. Summary information for pages.
  41. Percentage of binary-only segments that are used binary-only segments.
    [Figure]Number of used binary-only segments/total number of binary-only segments allocated to the table [Figure] 100 (%)[Figure]
  42. Percentage of binary-only segments that are full binary-only segments.[Figure]Number of full binary-only segments/total number of binary-only segments allocated to the table [Figure] 100 (%)[Figure]
  43. Number of used binary-only segments. Segments being released by pdreclaim are displayed as used binary-only segments.
  44. Number of full binary-only segments.
  45. Total number of binary-only segments allocated to the table.
  46. Summary information for binary-only segments.
  47. Percentage of pages in the binary-only segments that are used pages.
    [Figure]Number of used pages in the binary-only segments/total number of pages in the binary-only segments allocated to the table [Figure] 100(%)[Figure]
  48. Percentage of pages in the binary-only segments that are full pages in used binary-only segments.
    [Figure]Number of full pages in the used binary-only segments/total number of pages in the binary-only segments allocated to the table [Figure] 100(%)[Figure]
  49. Number of used pages in the binary-only segments.
  50. Number of full pages in the binary-only segments.
  51. Total number of pages in the binary-only segments allocated to the table.
  52. Summary information for pages in the binary-only segments.
  53. Number of segments being released if pdreclaim with the -a option specified is running concurrently or terminated with an error. If pdreclaim with the -a option specified is not executed, 0 is displayed.
  54. Number of pages whose release has not been completed because pdreclaim is running or pdreclaim terminated with an error. If pdreclaim terminated with an error, re-execute pdreclaim (0 is displayed for this value). The displayed number of pages is included in the used pages by percentage (0%).
  55. Sequence generator identifier.
  56. Authorization identifier of the sequence generator.
  57. Current value of the sequence generator. If no current value has been set, no value is displayed. If a log output interval has been specified for the sequence generator, the current value might not match the next sequence number to be used for a reason such as forced server termination. In such a case, an asterisk (*) is displayed after the current value. The next time logical analysis is performed, the correct value will be displayed because the current value will be corrected when it is used by a program such as a UAP.
  58. Free area search mode for branch rows:
    INS: New page allocate mode
    REU: Free page reuse mode
    If improve storage efficiency for tables with many branch rows is not specified as the reuse option value in SEGMENT REUSE, "-" is displayed.
  59. Number of full pages that can be released by pdreclaim or pdrorg.
  60. Total number of managed rows in table deleted status.
  61. Reuse option value specified in SEGMENT REUSE in the table definition.
Note 1
Nos. 21 to 23 are displayed when HiRDB Staticizer Option is used. If a generation is specified for analysis and the specified replica RDAREA generation does not exist, the utility issues a warning message, in which case no results are displayed.
Note 2
Nos. 31 to 32 are displayed when the -d option is specified.
Nos. 35 to 38 are displayed when the -d and -z options are both specified.
Note 3
Nos. 39 to 52 are displayed if a column of the BINARY type with a definition length of at least 256 bytes or a column of an abstract data type has been defined.

Figure 15-9 Types of unfinished index by table (1/2)

[Figure]

Figure 15-10 Types of unfinished index by table (2/2)

[Figure]

Figure 15-11 Types of unfinished index by RDAREA (1/2)

[Figure]

Figure 15-12 Types of unfinished index by RDAREA (2/2)

[Figure]
(b) For LOB RDAREA

pddbst VV-RR(Object Option) ** RD Area Logical Analysis **  2003/04/03 12:36:43 [1]
RD Area Name   : user_rdlob_2 [2]
 Server        : bes1 [3]
 Total Segment :      11000 [4] Segment Size :          1 Pages [6]
 Unused Segment:        9178(      9178) [5]
 Page Size    :       8192 Bytes [7]
 Original RD Area Name : user_rdlob_2 [16]
 Generation Number :  0[17]  Replica RD Area Count :  0 [18]
 [19]     [20]               [21]               [22]
 History1 Hold Status :      Hold Code :     0  Hold Time :
 History2 Hold Status :      Hold Code :     0  Hold Time :
--------------------------------------------------------------------------------
Table Name : TBL03 [8]
Auth Id    : user1 [9]
Status     :   [23]
         Used       Used        Sum
 Segment 100% [10]  1822 [11]  1822 [12]
 Segment Over  : N [13] Last Segment :       1822/     11000 [14]
 Lobmap  Over  : N [15]

Explanation
When there are multiple RDAREAs to be processed, the information is displayed in the following order:
  • RDAREA names (ascending order)
  • Table names (ascending order)
  • Index names (ascending order)
  1. Date and time of completion of acquisition of the information required for condition analysis, in the format YYYY/MM/DD hh:mm:ss:
    YYYY: Year. MM: Month. DD: Date. hh: Hour. mm: Minute. ss: Second.
  2. Name of RDAREA subject to analysis.
  3. Name of server managing the RDAREA.
  4. Total number of segments in the RDAREA (number of used segments + number of unused segments).
  5. Total number of unused segments in the RDAREA. The parentheses enclose the total number of unused segments in HiRDB files that are not in the frozen update status in the corresponding RDAREA.
  6. Segment size (number of pages per segment).
  7. Page size (bytes).
  8. Name of a table or index contained in the RDAREA.
  9. Authorization identifier of the table. For a data dictionary LOB RDAREA, (Data dictionary) is displayed.
  10. Percentage ratio of used segments:
    [Figure]value of 11/value of 12[Figure][Figure] 100 (%)
  11. Number of used segments. Segments being released by pdreclaim are displayed as used segments.
  12. Total number of segments allocated to the table (number of used segments).
  13. Whether or not the order is lost in a LOB RDAREA.
    Y: Order is lost.
    N: Order is not lost.
  14. Position information indicating that the segment being used is the last segment; this is indicated in a last position/total number of segments format; always indicates the last segment when {13} is Y.
  15. Whether or not the LOB management entries are all used:
    Y: All are used.
    N: There are unused entries.
  16. Name of the original RDAREA.
  17. Generation number of the corresponding RDAREA.
  18. Number of replica RDAREAs.
  19. Shutdown history of the corresponding RDAREA.
    History1: Shutdown information immediately before the shutdown was released (if the RDAREA is currently shut down, this field displays the current shutdown information)
    History2: Shutdown information immediately preceding History1
  20. Shutdown type of the corresponding RDAREA. If there is no history, this field is blank.
    CMD: Command shutdown by HiRDB error detection
    FLT: Error shutdown
  21. Code indicating the cause of the shutdown of the corresponding RDAREA. For details about the shutdown cause code, see 15.3.1(2)(a) For an RDAREA other than a LOB RDAREA.
  22. Shutdown time of the corresponding RDAREA. If there is no history, this field is blank.
  23. For a table, reload-not-completed data status or being reorganized. If the table is not in either of the following statuses, this field is blank.
    DATA_UNFINISH:
    The table is in reload-not-completed data status, which means that table reorganization was executed, but reload processing has not been completed for some reason such as an error. To release the table from reload-not-completed data status, you must re-execute the table reorganization or reload data into the table.
    ON_RORG:
    The table is currently being reorganized.
    For an index, this is the status of an unfinished index. If the index has been completed, this field is blank.
    UNFINISH_0:
    See Figure 15-9 Types of unfinished index by table (1/2) and Figure 15-10 Types of unfinished index by table (2/2).
    UNFINISH_1:
    See Figure 15-11 Types of unfinished index by RDAREA (1/2) and Figure 15-12 Types of unfinished index by RDAREA (2/2).
    UNFINISH_2:
    Delayed plug-in index updating is underway. This is the status when a UAP using the plug-in index delayed batch creation function is being executed, or after the UAP finished executing, a batch index creation process using the database reorganization utility is not being executed. For details about the plug-in index delayed batch creation function, see the HiRDB Version 9 System Operation Guide.
Note
Nos. 16 to 18 are displayed when HiRDB Staticizer Option is used. If a generation is specified for analysis and the specified replica RDAREA generation does not exist, the utility issues a warning message, in which case no results are displayed.
(c) For DAT format (when the -a option is specified)

The following rules apply in the case of output in DAT format:

#1
When the output format is character string, the output information is enclosed in double quotation marks (""), in which case the maximum length does not include the double quotation marks.
#2
If HiRDB Staticizer Option has not been installed, the null element is output.
#3
What is output as the reuse option value depends on the specification of SEGMENT REUSE in the table definition. The following table shows what is output depending on the value specified for SEGMENT REUSE.
No.Specification of SEGMENT REUSE and specified valueSpecification of SEGMENT REUSE OPTIONOutput
1SpecifiedOther than NOSpecifiedSpecified value
2Omitted0
3NO--Null data
4Omitted----Null data
Legend:
--: Not applicable

#4
What is output as the free page reuse mode depends on the specification of SEGMENT REUSE in the table definition. The following table shows what is output depending on the value specified for SEGMENT REUSE.
No.Specification of SEGMENT REUSE and specified valueSEGMENT REUSE OPTION valueOutput
1SpecifiedOther than NO2,3
  • INS (New page allocate mode)
  • REU (Free page reuse mode)
2Other than 2,3Null data
3NO--Null data
4Omitted----Null data
Legend:
--: Not applicable

(3) Analyzing the analysis results

The results of condition analysis by RDAREA (logical analysis) can be analyzed as shown in the table below.

Table 15-4 Analyzing the analysis results by RDAREA (logical analysis)

Status in the RDAREADescription
Percentage of total unused pages in the table is less than the percentage of free pages per segment that was specified in the CREATE TABLE definition SQL statement.Table data may be in one of the following statuses:
  • There are only a few unused pages because data was added repeatedly.
  • Used free pages are scattered because data addition, deletion, and updating occurred repeatedly.
If necessary, you can correct the table arrangement by expanding the RDAREA that stores the table and then reorganizing the table.#1, #2
The total number of used index storage pages displayed in the analysis results is greater than the number of index storage pages obtained from the number of table storage rows.The index may be in the following status:
  • Because many rows in a specific range were deleted, a large amount of the index was also deleted, resulting in many used free pages.
In this case, you can correct the index arrangement by reorganizing the index. You can adjust the number of used pages to a reasonable value by releasing free index pages. For details about determining the number of index storage pages, see the HiRDB Version 9 Installation and Design Guide.
Number of used pages by percentage rangesThere are many pages whose usage percentage is 0%Used free pages account for at least 30% of the total number of pagesPerformance is affected adversely when there are many used free pages. You should release used free pages.
Another table or index is defined in the same RDAREAUnused pages can be used only by the table or index to which the segments containing those pages have been allocated. This means that if another table or index is defined in the same RDAREA, fewer pages are available to be allocated to the other table or index. You should release used free segments or perform reorganization. The number of unused pages is the difference between the number of pages whose usage is 0% and the number of used free pages.
There are many pages whose usage percentage is lowArea that was once allocated remains as unreleased free area due to data deletion and updating. Such area cannot be reused. You should perform reorganization.
There are many pages whose usage percentage is highThere is only a small amount of free space in the pages that can be allocated during data addition or updating. If either of the following is applicable, you should perform reorganization:
  • You frequently perform update operations that result in longer data lengths.
  • You frequently perform insert operations on a table with a cluster key defined.
Ratio of the number of rows managed by remaining entries to the total number of rows managed by the indexHighThere are many remaining entries. You should release used free pages or perform reorganization to avoid illegal lock-release waits and deadlocks, as well as adverse effects on search performance.
For details about the effects of a large number of remaining entries, see the HiRDB Version 9 UAP Development Guide.
LowIf there are only a few remaining entries but many of the rows managed by the remaining entries are in a search range, search performance might be affected adversely. If the number of pages accessed during searches is large compared to the search hits count, performance is affected adversely and you should release free pages or perform reorganization. You can use the pdbufls command or check the UAP statistical report to determine the number of pages accessed.
Table with clusteringUsed free pagesManyPercentage of free pages per segment specified in CREATE TABLE is too large. If necessary, reorganize the table because used free pages have been created due to data addition, deletion, and updating.#1
FewIf you are adding data, the table should be reorganized.#1
Table without clusteringUsed free pagesManyPercentage of free pages per segment specified in CREATE TABLE is too large. If necessary, reorganize the table because used free pages have been created due to data addition, deletion, and updating.#1
FewIf there are only a few unused segments in the RDAREA, expand the RDAREA as required.#2
End-of-segment is YLOB column search performance may become poor. Reorganize the LOB RDAREA.#1
End-of-Lobmap is YLOB column search performance may become poor. Reorganize the LOB column structure base table.#1
Note
If an RDAREA storing the LOB attribute of a plug-in-provided abstract data type or an RDAREA storing a plug-in index is analyzed, the percentage of used segments in the RDAREA depends on the method used by the plug-in to allocate segments. For example, if a plug-in completes allocation of segments during index definition, most of the segments may become used segments immediately upon definition (before data is stored).
#1: For details about table and index reorganization, see Chapter 8. Database Reorganization Utility (pdrorg). For details about releasing free index pages, see Chapter 11. Free Page Release Utility (pdreclaim).
#2: For details about RDAREA expansion, see Chapter 7. Database Structure Modification Utility (pdmod).