STATEMENT

Obtains information about executed statements (normalized SQL statements) on the current node. You can view all statistics about normalized SQL statements received by the primary database node and other database nodes, whereas you can view only the statistics about normalized SQL statements executed on other database nodes.

The unique_sql_id generated by different savepoint_name values is different. When a large number of savepoint_name is used, the number of unique_sql_id values generated in the system increases rapidly. If the number of unique_sql_id values is greater than the number of instr_unique_sql_count values, the newly generated unique_sql_id information is not counted.

**Table 1** STATEMENT columns
Name	Type	Description
node_name	name	Database process name
node_id	integer	ID of a node
user_name	name	Username
user_id	oid	OID of the user
unique_sql_id	bigint	ID of the normalized SQL statement
query	text	Normalized SQL statement Note: The length is controlled by track_activity_query_size.
n_calls	bigint	Number of calls
min_elapse_time	bigint	Minimum execution time of the SQL statement in the kernel (unit: μs)
max_elapse_time	bigint	Maximum execution time of the SQL statement in the kernel (unit: μs)
total_elapse_time	bigint	Total execution time of the SQL statement in the kernel (unit: μs)
n_returned_rows	bigint	Number of rows in the result set returned by the SELECT statement
n_tuples_fetched	bigint	Number of rows randomly scanned
n_tuples_returned	bigint	Number of rows sequentially scanned
n_tuples_inserted	bigint	Number of rows inserted
n_tuples_updated	bigint	Number of rows updated
n_tuples_deleted	bigint	Number of rows deleted
n_blocks_fetched	bigint	Number of buffer block access times
n_blocks_hit	bigint	Number of buffer block hits
n_soft_parse	bigint	Number of soft parsing times. The value of n_soft_parse plus the value of n_hard_parse may be greater than the value of n_calls because the number of subqueries is not counted in the value of n_calls.
n_hard_parse	bigint	Number of hard parsing times. The value of n_soft_parse plus the value of n_hard_parse may be greater than the value of n_calls because the number of subqueries is not counted in the value of n_calls.
db_time	bigint	Valid DB time, which is accumulated if multiple threads are involved (unit: μs)
cpu_time	bigint	CPU time (unit: μs)
execution_time	bigint	Execution time in the executor (unit: μs)
parse_time	bigint	SQL parsing time (unit: μs)
plan_time	bigint	SQL plan generation time (unit: μs)
rewrite_time	bigint	SQL rewriting time (unit: μs)
pl_execution_time	bigint	Execution time of PL/pgSQL (unit: μs)
pl_compilation_time	bigint	Compilation time of PL/pgSQL (unit: μs)
data_io_time	bigint	I/O time (unit: μs)
net_send_info	text	Network status of messages sent through a physical connection, including the time (unit: μs), number of calls, and throughput (unit: byte). This column is not supported in standalone mode.
net_recv_info	text	Network status of messages received through a physical connection, including the time (unit: μs), number of calls, and throughput (unit: byte). This column is not supported in standalone mode.
net_stream_send_info	text	Network status of messages sent through a logical connection, including the time (unit: μs), number of calls, and throughput (unit: byte). This column is not supported in standalone mode.
net_stream_recv_info	text	Network status of messages received through a logical connection, including the time (unit: μs), number of calls, and throughput (unit: byte). This column is not supported in standalone mode.
last_updated	timestamp with time zone	Last time when the statement was updated
sort_count	bigint	Sorting count
sort_time	bigint	Sorting duration (unit: μs)
sort_mem_used	bigint	Size of work memory used during sorting (unit: KB)
sort_spill_count	bigint	Count of file writing when data is flushed to disks during sorting
sort_spill_size	bigint	File size used when data is flushed to disks during sorting (unit: KB)
hash_count	bigint	Hashing count
hash_time	bigint	Hashing duration (unit: μs)
hash_mem_used	bigint	Size of work memory used during hashing (unit: KB)
hash_spill_count	bigint	Count of file writing when data is flushed to disks during hashing
hash_spill_size	bigint	File size used when data is flushed to disks during hashing (unit: KB)
parent_unique_sql_id	bigint	Unique ID of the parent SQL statement. The value is 0 for a non-stored procedure substatement.

n_calls indicates the actual number of calling times. For the FETCH statement in the stored procedure, the actual number of triggering times of the FETCH statement is the increase times of n_calls of the statement actually executed by the cursor.

Parent topic: Query

Previous topic: Query

Next topic: SUMMARY_STATEMENT