The SYSXPLAIN_SCAN_PROPS table captures information about table/index accesses which occurred during statements that were executed using RUNTIMESTATISTICS with XPLAIN style.
See "Working with RunTimeStatistics" in Tuning Derby for information on how to configure this.
Each row in this table describes a single table/index scan for a particular result set used by a particular statement. Rows in this table are typically joined with rows in SYSXPLAIN_STATEMENTS and SYSXPLAIN_RESULTSETS during analysis:
select st.stmt_text, sp.no_visited_rows
from my_stats.sysxplain_scan_props sp,
my_stats.sysxplain_resultsets rs,
my_stats.sysxplain_statements st
where st.stmt_id = rs.stmt_id and
rs.scan_rs_id = sp.scan_rs_id and
rs.op_identifier = 'TABLESCAN' and
sp.scan_object_name = 'COUNTRIES'
Rows in this table are added automatically when Derby has been configured appropriately. The rows remain in the table until you delete them or drop the table.
The following table shows the contents of the SYSXPLAIN_SCAN_PROPS system table.
| Column Name | Type | Length | Nullable | Contents |
|---|---|---|---|---|
| SCAN_RS_ID | CHAR | 36 | false | A unique identifer for this particular row. Referenced by the foreign key SCAN_RS_ID in SYSXPLAIN_RESULTSETS. |
| SCAN_OBJECT_NAME | VARCHAR | 128 | true | The name of the object being scanned. If this is a scan of a table or index, the table name or index name appears here. If this is a scan of the internal index created for a constraint, the constraint name appears here. For complex join queries, the object being scanned may be an intermediate result, in which case a description such as 'Temporary HashTable' appears. |
| SCAN_OBJECT_TYPE | CHAR | 1 | false | A code indicating the type of object being scanned. Codes include 'T' for Table, 'I' for Index, and 'C' for Constraint. |
| SCAN_TYPE | CHAR | 8 | false | The type of scan being performed. Scan types include 'HEAP', 'BTREE', and 'SORT'. |
| ISOLATION_LEVEL | CHAR | 3 | true | The isolation level being used for this scan. Isolation levels are identified by a code: 'RU' for Read Uncommitted, 'RC' for Read Committed, 'RR' for Repeatable Read, and 'SE' for Serializable. |
| NO_VISITED_PAGES | INTEGER | 10 | true | Number of database pages that this scan touched. For btree scans this number only includes the leaf pages visited. |
| NO_VISITED_ROWS | INTEGER | 10 | true | Number of database rows that were examined by this scan. This number includes all rows, including those rows marked deleted, those rows that don't meet qualification, and those rows which were returned by the scan. |
| NO_QUALIFIED_ROWS | INTEGER | 10 | true | Number of rows that satisfied the qualifiers for this scan. |
| NO_VISITED_DELETED_ROWS | INTEGER | 10 | true | Number of the database rows that were examined by this scan which were found to be rows that were marked deleted. |
| NO_FETCHED_COLUMNS | INTEGER | 10 | true | Number of columns that were fetched from each qualifying row. |
| BITSET_OF_FETCHED_COLUMNS | VARCHAR | 32,672 | true | Description of the columns which were fetched from each qualifying row. |
| BTREE_HEIGHT | INTEGER | 10 | true | For a scan of type BTREE, this column holds the height of the BTREE index. The typical height of a BTREE is 2-4; BTREE heights larger than this should only be seen with very large indexes. A tree with one page has a height of 1. Total number of pages visited in a scan of a BTREE should be (BTREE_HEIGHT - 1 + NO_VISITED_PAGES). For an extremely small BTREE, the btree height may be negative (-1). For other types of scans, this column is NULL. |
| FETCH_SIZE | INTEGER | 10 | true | The number of pages fetched at a time when the scan is retrieving pages from disk. |
| START_POSITION | VARCHAR | 32,672 | true | For index and constraint scans, a textual representation of the operator, if any, which was used to position the beginning of the index/constraint scan. |
| STOP_POSITION | VARCHAR | 32,672 | true | For index and constraint scans, a textual representation of the operator, if any, which was used to position the end of the index/constraint scan. |
| SCAN_QUALIFIERS | VARCHAR | 32,672 | true | If the query specified values which are to be used to limit the rows that are scanned, information about those values is captured in this column. |
| NEXT_QUALIFIERS | VARCHAR | 32,672 | true | If the query specified values which are to be used to limit the rows that are scanned, information about those values is captured in this column. |
| HASH_KEY_COLUMN_NUMBERS | VARCHAR | 32,672 | true | For hash joins, this column contains information about which column is being used to hash the rows that are joined. |
| HASH_TABLE_SIZE | INTEGER | 10 | true | For hash joins, this column contains information about the size of the hash table that will be used to hold the rows being joined. This hash table is an intermediate result, and will be discarded at the end of the query. If the hash table cannot fit in memory, it will automatically spill over to disk. Since the spillover to disk can have significant performance implications, this value can provide a clue that the hash table was unexpectedly too large to fit in memory. |