Reading the contents of the Column Distribution Statistics window

Top area of the window

The top area of the Column Distribution Statistics window shows general information about the column with which the column distribution statistics are associated.

The columns in the window show both those statistics that were gathered at the time of the explain, and those that currently exist in the system catalog tables.

Note:

Distribution statistics do not appear if they are not available or if they are not used by the optimization level chosen.

Rows in the Column Distribution Statistics window

Column distribution information is shown in the bottom area of the window. Each row shows information on one instance of a frequency or quantile value. The left area shows existing explained values; the right area shows current values. (The values with a current count of -1 exist in the configuration settings but are not currently being used.)

The following information is shown:

Type

The type of distribution (frequency or quantile).

Sequence number

How non-uniform distribution statistics are ordered. For frequent values, the sequence number orders the rows for a given column from the most frequent to the least frequent. For quantiles, the sequence number orders rows for a given column by increasing value.

Explained value

The value of the column at the time of the explain.

Explained count

The number of occurrences of the value at the time of the explain.

For quantiles, the value shown is the cumulative count of rows that are less than or equal to the explained value. (If two rows contain the same explained value, the row with the lower sequence number is used to provide a count of the number of rows that are strictly less than the explained value.)

Current value

The current value of the column.

Current count

The number of occurrences of the current value.

For quantiles, the value shown is the cumulative count of rows that are less than or equal to the current value. (If two rows contain the same current value, the row with the lower sequence number is used to provide a count of the number of rows that are strictly less than the current value.)

For more information, see the section on collecting and using distribution statistics in the Administration Guide.

Note:

If no values are shown, it could be for the following reasons: The command runstats WITH DISTRIBUTION was not run. The NUM_FREQVALUES and num-quantiles configuration parameters indicate that distribution statistics should not be collected. The distribution of the data is evenly distributed (for example, each data value in a column is unique). The data type is one for which statistics are not collected (that is, the column is defined as a long field or large object data type).