Help for SAP Data Hub Cockpit

View Profiled Results

Learn more about your data by viewing the minimum and maximum values, unique values, length, and more.

To view the profiled results, select a profiled object, and then click Fact Sheet. When prompted, confirm that it is OK for all users to view the results of this profiling task.

The fact sheet provides metadata about your profiled data. It includes minimum and maximum values, uniqueness, length, and so on. There is a wealth of information and many ways to sort and filter the data. First, we'll describe the information in the graphs.

Graphs

There are two graph features available. One of the graphs is a donut graph that shows the column types available in the data.
  • Character
  • DateTime
  • Numeric
  • Other

Clicking on each type in the graph shows the number of columns that have that type.

The other graph is a line graph that shows up to five recent versions of the fact sheet. This graph is only available after you have profiled the data two or more times. By using the dropdown list, you can show one of these trends.
  • Trend by Row Count: shows any variations in the number of rows
  • Trend by Size: shows any variation in the table or file size
Clicking on the node of each data point shows some metadata about that version of the fact sheet.
Option Description
Version ID Shows the version of the fact sheet. There are up to five versions available depending on how many times the data was profiled.
Profile Date Lists the date and time the data was profiled.
Row Count Shows the number of profiled rows.
Size Shows the file size.
Runtime Shows the processing time to profile the data.
Status

OK means that the data profiling was successful.

ERROR means that there was an issue that occurred during profiling, or the object profiling was canceled by the user. Click the link to view more information about the error.

Table

You can filter the table by clicking in Filter column names and typing a column name.

Column Heading Description
Column Detail Lists the column names from the object.
Type (Length)

Type (precision, scale)

Lists the data type and its length in parentheses. Depending on the profiled object, 'Length' can mean one of several things.
  • The value set for the length of the column
  • The length of the longest value for a character data type
  • The precision and scale for a numeric data type
Minimum - Maximum Includes a range of values, from the smallest value to the largest value, based on the data type.
  • Character data types list the first and last values based on the alphabetical order, for example, aardvark - zebra.
  • Numeric data types list the highest and lowest numeric values, for example, 1-9999.
  • Date data types list the earliest and latest date or time, for example, 01/01/2001 - 08/08/2017.
Average Length The mean length of the values in the specified column. The length is the sum of the lengths of all the values divided by the number of values. Only certain data types have an average length value.
Null / Blank / Zero Lists the percentage of the values occurring as null (empty), blank (contains a blank space) or zero (value is 0).
Distinct Value Lists the amount of values that are different (unique) from other values.
Uniqueness Describes the count and percentage of rows that contain unique data. Uniqueness is classified as one of three types.
  • Sparse: More than 90% of the values of the column are null or blank.
  • Low Cardinality: The number of distinct values in the column is less than 2%.
  • Unique: All of the values in the column are distinct.

Distinct Values Bar Chart

Select one of the columns in the table, and the Distinct Values graph shows the top 10 values that are the most frequently occurring in the data. You can click the value to learn more about the data. Depending on the content type, you may see the value, number of times it appears, and a percentage of the overall values.

Filtering the Profiled Data

You can filter the data in several ways on the fact sheet.
  • Column names: In the filter box above the table, enter a portion or all of a column name. The result shows only the columns that match what you have typed.
  • Column type donut chart: In the graph above the table, click one of the data types, and then click Go. For example, if you chose Numeric, only those columns with the numeric data type are shown in the table.
  • Trend chart (when available): Select a version, and filter the data for that version.
  • Category: Choose one or more categories and define the type of data that you want to see in the table. You can filter on one category, such as Minimum / Maximum, and further refine those results by adding another Minimum / Maximum filter (by clicking the + icon), or by adding a filter to another category such as Uniqueness. All of the set filters are combined together with an AND operator, so the data must meet all of the criteria to be shown in the results table. Click Go.
    Filter Category Description
    Length / Average Length Length is represented in parentheses in the Type column. Average Length is in its own column. Choose Length or Average Length, then choose an operator and enter a number. Only those columns that have a length value are affected by the filter.
    Null / Blank / Zero Three types of data are displayed in the column. Depending on the data type, you can typically filter on two of the types. For example, the numeric data types can show zero and null, while character data types can show null and blank. Choose Null, Blank or Zero, then choose an operator and enter a number. Only those columns that have data in the filter you created are affected by the filter. If a column has 0%, it is still included in the filter.
    Minimum - Maximum There are four types of data available in the column depending on the data type. Use Minimum Length or Maximum Length to filter character type columns. Use Minimum Value or Maximum Value to filter columns of all data types.
    Uniqueness There are three types of data available: Low Cardinality, Sparse Values, and Unique Values. See the descriptions earlier in this topic. Choosing All shows all of the uniqueness data.
    Distinct Value Select the Distinct Value, choose an operator, and enter a value.

To remove filters, click the category, and then click the x next to the filter you want to delete. Click OK. To clear all filters, click the clear filters icon in the upper right corner.

Click Show Filters when you want to create filters. Click Hide Filters to expand the results in the table, so that you can view more rows of data.