Performance: History - Database (DB2 UDB for
UNIX and Windows) 
The system provides day-by-day trend analysis of database activity. You can check the workload of the days and display the workload peak of a single day.
You can access the Performance History screen by calling the DBA Cockpit and choosing Performance ® History ® Database in the navigation frame of the DBA Cockpit.
An overview of all days monitored is displayed:
|
Column |
Description |
|
Partition |
Monitored partition (only displayed if you are using a multi-partition database) |
|
Date |
Day when monitoring was performed |
|
Avg. Phys. Read Time (ms) |
Average physical read time. If you have chosen Total Day, this is the average of all measured average read times. If you have chosen Peak, this is the worst measured read time. |
|
Avg. Phys. Write Time (ms) |
Average physical write time. If you have chosen Total Day, this is the average of all measured average write times. If you have chosen Peak, this is the worst measured write time. |
|
Data Logical Reads |
Number of read accesses to data in the buffer pool |
|
Data Physical Reads |
Number of read accesses to data on disk (I/O). The value includes the number of physical reads that were performed synchronously (by the database manager agents) and asynchronously (by the I/O servers). |
|
Data Physical Writes |
Number of write accesses to data on disk (I/O). The value includes the number of physical writes that were performed synchronously (by the database manager agents) and asynchronously (by the I/O cleaners). |
|
Index Logical Reads |
Number of read accesses to index data in the buffer pool |
|
Index Physical Reads |
Number of read accesses to index data on disk (I/O). Since index data is read only by the database manager agents, this value includes the number of synchronously read index pages. |
|
Index Physical Writes |
Number of write accesses to index data on disk (I/O). The value includes the number of physical writes that were performed synchronously (by the database manager agents) and asynchronously (by the I/O cleaners). |
|
COMMIT Statements |
Total number of COMMIT statements that have been attempted |
|
ROLLBACK Statements |
Total number of ROLLBACK statements that have been attempted. Automatic ROLLBACKs caused by error situations or deadlocks are not included. |
|
Lock Waits |
Total number of times that applications or connections waited for locks |
|
Lock Wait Time (ms) |
Total elapsed time in milliseconds waited for a lock |
|
Deadlocks |
Total number of deadlocks that have occurred |
|
Lock Escalations |
Number of times that locks have been escalated from several row locks to a table lock |
|
Exclusive Lock Escalations |
Number of times that locks have been escalated from several row locks to one exclusive table lock, or the number of times an exclusive lock on a row caused the table lock to become an exclusive lock |
If you choose Total Day in the field Workload in the Performance History - Database group box, the total workload of this day is displayed. The value displayed is calculated using formula maximum value - minimum value. Database restarts are taken into consideration.
If you choose Peak in the field Workload in the Performance History - Database group box, the maximum of all measured values is displayed.
You can display
details for one specific day by double-clicking a field or selecting a row and
choosing 
Details. A detail screen appears with the following
information:
· Snapshot
The measured values of the selected day are displayed.
· Interval
The delta values of the measurements, which are provided under Snapshot, are displayed.
If you select a particular day, snapshots of the database activity are displayed in a two-hour cycle.
If the database is restarted during one day, the interval displayed after the restart does not equal the delta of two measurements because the counter was reset during the restart. In this case the absolute value of the last measurement is displayed.