This chapter provides information about the job accounting process and the columns in the Job Accounting table.
If you set the job accounting status system parameter to be active, the notifier component adds a row to the Job Accounting table each time a job completes, either successfully or unsuccessfully.
This information is used by the Tracker utility to provide reports that display a database usage history for queries managed by the DB2 Query Patroller system. For example, this information will be used by Tracker to create reports showing which tables and columns have been accessed most frequently, which tables have returned the most result rows, and which jobs have been completed within a specific time frame.
The following list explains each of the columns in the Job Accounting table.
The ACCOUNT_ID column represents the accounting identifier associated with the user from the User Profile table.
The COMPLETION_STATUS column contains a single letter code indicating how the job completed. Values are D for done (normal completion), C for canceled, or A for aborted.
The DATA_SOURCE column identifies the database instance against which the query was run. Currently, the only supported data source identifier is IWM.
The DTIME_COMPLETED column identifies the date and time that the job was completed.
The DTIME_CREATED column identifies the date and time that the user submitted the job to DB2 Query Patroller.
The DTIME_SCHEDULED column identifies the date and time after which the user scheduled the job to be run.
The DTIME_STARTED column identifies the date and time that the job started running. If the job was canceled or aborted before the job started running, this column contains a NULL value.
The ELAPSED_TIME column identifies the elapsed time in seconds from when the job started running until it completed.
The ELAPSED_TIME_LIMIT column identifies the maximum number of elapsed seconds a job will be permitted to run.
The ERROR_REASON column identifies the reason why the job did not complete successfully.
The ESTIMATED_COST column identifies the estimated cost for the job in accounting units, if one was derived. This column contains the value of the estimated cost for the job multiplied by the cost factor from the Data Source table.
The JOB_ID column contains a number that identifies the job.
The JOB_PRIORITY column contains the priority level number assigned to the job.
The JOB_TEXT column contains either the job's SQL statement or its system command.
The JOB_TYPE column contains either the letter S for SQL jobs or the letter C for jobs containing system commands.
The NODE_ID_EXECUTED column identifies the host name of the node where the job was run.
The NODE_ID_SUBMITTED column identifies the node from which the job was submitted.
If the job was assigned a predecessor job (a job which must have completed before this job could run), the PREDECESSOR_JOB column contains the predecessor's job identifier. If no predecessor was assigned, this column contains a NULL value.
If the user selected an alternative result destination, the RESULT_DESTINATION column contains the identifier of the destination. If the results were to be written to a result table created in the database, this column contains a NULL value.
The RESULT_ROWS column contains the number of database rows that a job selected, updated, deleted, or inserted.
The RESULT_ROWS_LIMIT column identifies the maximum number of rows a SQL SELECT statement is allowed to return.
The RESULT_TABLE_OWNER column contains the database identifier of the owner of the result table, if it was created.
The RESULT_TABLE_NAME column contains the table name of the result table, if it was created.
The SUBMIT_SOURCE column contains the name of the program that submitted the job.
The SUBMITTER_USER_ID column contains the system identifier of the submitter of the job. This identifier could be different from the job's owner if one user submitted the job on behalf of another user.
The SYSTEM_TIME column contains the number of CPU seconds of system time consumed by the job.
The USER_ID column contains the system identifier of the job owner.
The USER_TIME column contains the number of CPU seconds of user time consumed by the job.