V$SQL_PLAN_MONITOR

V$SQL_PLAN_MONITOR displays plan level monitoring statistics for each SQL statement found in V$SQL_MONITOR. Each row in V$SQL_PLAN_MONITOR corresponds to an operation of the execution plan being monitored. As with V$SQL_MONITOR, statistics exposed in V$SQL_PLAN_MONITOR are generally updated every second when the statement executes. These statistics are recycled on the same basis as V$SQL_MONITOR.

To eliminate the overhead of SQL plan monitoring, statistics collected for each operation of the plan don't record timing information such as elapsed time, CPU time, or I/O time. Instead, this timing information can be estimated quite accurately by joining V$SQL_PLAN_MONITOR with V$ACTIVE_SESSION_HISTORY on SQL_ID, SQL_EXEC_START, SQL_EXEC_ID, and SQL_PLAN_LINE_ID (simply named PLAN_LINE_ID in V$SQL_PLAN_MONITOR). The result of that join is a sample of the activity performed by each operation in the plan, from which an estimate of CPU time and wait time can be derived. This can be achieved by breaking statement level monitoring time statistics found in V$SQL_MONITOR in proportion to the number of samples found in V$ACTIVE_SESSION_HISTORY for the corresponding activity type.

Column Datatype Description
KEY NUMBER Foreign key to efficiently join V$SQL_PLAN_MONITOR with V$SQL_MONITOR (see V$SQL_MONITOR)
STATUS VARCHAR2(19) SQL execution status:
  • EXECUTING - SQL statement is still executing

  • DONE (ERROR) - Execution terminated with an error

  • DONE (FIRST N ROWS) - Execution terminated by the application before all rows were fetched

  • DONE (ALL ROWS) - Execution terminated and all rows were fetched

  • DONE - Execution terminated (parallel execution)

FIRST_REFRESH_TIME DATE Time when monitoring of the SQL statement started
LAST_REFRESH_TIME DATE Time when statistics were last updated for the SQL statement
FIRST_CHANGE_TIME DATE First time a row was produced by this operation
LAST_CHANGE_TIME DATE Last time a row was produced by this operation
REFRESH_COUNT NUMBER Number of times statistics have been refreshed
SID NUMBER Session identifier executing (or having executed) the SQL statement being monitored
PROCESS_NAME VARCHAR2(5) Process name identifier
SQL_ID VARCHAR2(13) SQL identifier
SQL_EXEC_START DATE Time when the execution started
SQL_EXEC_ID NUMBER Execution identifier
SQL_PLAN_HASH_VALUE NUMBER SQL plan hash value
SQL_CHILD_ADDRESS RAW(4 | 8) Address of the child cursor
PLAN_PARENT_ID NUMBER ID of the next execution step that operates on the output of the current step
PLAN_LINE_ID NUMBER Plan line number for the entry
PLAN_OPERATION VARCHAR2(30) Plan operation name (from V$SQL_PLAN)
PLAN_OPTIONS VARCHAR2(30) Plan option name (from V$SQL_PLAN)
PLAN_OBJECT_OWNER VARCHAR2(30) Name of the user who owns the schema containing the table or index
PLAN_OBJECT_NAME VARCHAR2(30) Name of the table or index
PLAN_OBJECT_TYPE VARCHAR2(20) Type of the object
PLAN_DEPTH NUMBER Depth (or level) of the operation in the tree. It is not necessary to issue a CONNECT BY statement to get the level information, which is generally used to indent the rows from the PLAN_TABLE table. The root operation (statement) is level 0.
PLAN_POSITION NUMBER Order of processing for all operations that have the same PARENT_ID
PLAN_COST NUMBER Cost of the operation as estimated by the optimizer's cost-based approach. For statements that use the rule-based approach, this column is NULL.
PLAN_CARDINALITY NUMBER Estimate, by the cost-based optimizer, of the number of rows produced by the operation
PLAN_BYTES NUMBER Estimate, by the cost-based optimizer, of the number of bytes produced by the operation
PLAN_TIME NUMBER Elapsed time (in seconds) of the operation as estimated by the optimizer's cost-based approach. For statements that use the rule-based approach, this column is NULL.
PLAN_PARTITION_START VARCHAR2(64) Start partition of a range of accessed partitions
PLAN_PARTITION_STOP VARCHAR2(64) Stop partition of a range of accessed partitions
PLAN_CPU_COST NUMBER CPU cost of the operation as estimated by the optimizer's cost-based approach. For statements that use the rule-based approach, this column is NULL.
PLAN_IO_COST NUMBER I/O cost of the operation as estimated by the optimizer's cost-based approach. For statements that use the rule-based approach, this column is NULL.
PLAN_TEMP_SPACE NUMBER Temporary space usage of the operation (sort or hash-join) as estimated by the optimizer's cost-based approach. For statements that use the rule-based approach, this column is NULL.
STARTS NUMBER Number of times this operation was executed. For example, an operation is executed multiple times when it is on the right side of a nested-loop join (once for each row of the left input of that nested-loop join).
OUTPUT_ROWS NUMBER Number of rows produced by this operation since the execution started. This number is cumulated for all executions of this operation. Divide by the value of the STARTS column to compute the average number of rows per execution of the operation. Note that the value in the STARTS column is equal to or higher than the value in the OUTPUT_ROWS column. The value will usually be equal, but depending on internal optimizations a higher value may be seen.
IO_INTERCONNECT_BYTES NUMBER Number of I/O bytes exchanged between Oracle Database and the storage system. Maintained only after Oracle starts to monitor the execution.
PHYSICAL_READ_REQUESTS NUMBER Number of physical read I/O requests issued by the monitored SQL. Maintained only after Oracle starts to monitor the execution.
PHYSICAL_READ_BYTES NUMBER Number of bytes read from disks by the monitored SQL. Maintained only after Oracle starts to monitor the execution.
PHYSICAL_WRITE_REQUESTS NUMBER Number of physical write I/O requests issued by the monitored SQL. Maintained only after Oracle starts to monitor the execution.
PHYSICAL_WRITE_BYTES NUMBER Number of bytes written to disks by the monitored SQL. Maintained only after Oracle starts to monitor the execution.
WORKAREA_MEM NUMBER Amount of memory (in bytes) used by the operation when the query is executing; NULL if the execution is done. This applies only to operations using a work area, such as sort, hash-join, group-by, and so on.
WORKAREA_MAX_MEM NUMBER Maximum value (in bytes) for WORKAREA_MEM; NULL if the operation is not using a work area. When the execution is finished, this value will hold the maximum amount of memory consumed by this operation during the execution of the statement.
WORKAREA_TEMPSEG NUMBER Amount of temporary space (in bytes) used by the operation when the query is executing; NULL if the operation has not spilled to disk or if the execution is finished
WORKAREA_MAX_TEMPSEG NUMBER Maximum value (in bytes) for WORKAREA_TEMPSEG; NULL if this operation never spilled to disk. When the execution is done, this value will hold the maximum amount of temporary space consumed by this operation during the entire execution.