Part III: Data Discovery with Informatica Developer > Data Object Profiles > Primary Key Discovery
  

Primary Key Discovery

Primary key discovery generates primary key candidates from the columns you specify.
A primary key is a column or combination of columns that uniquely identify a row in a data source. Primary key discovery identifies the columns and combinations of columns that meet a specific confidence level. You can edit the confidence level, as well as the maximum number of columns to combine for primary key identification.
Primary key discovery can highlight potential data quality issues by identifying the non-unique rows in a primary key candidate. This is especially useful in cases where primary key discovery combines many columns, since non-conforming records are likely to contain duplicate information.

Primary Key Inference Properties

When you create a single data object profile, you can use the Primary Key Profiling view to configure the primary key inference properties.
The following table describes the primary key inference properties in the Primary Key Profiling view:
Property
Description
Override the default inference options
Allows you to configure custom settings for primary key inference.
Max Key Columns
Maximum number of columns that can make up a primary key.
Max Rows
Number of rows to profile.
Conformance Criteria
Minimum percentage or maximum number of rows of key violations that the profile allows when determining primary keys.
Exclude data objects with documented, user defined key
Excludes data objects with documented primary keys or user-defined primary keys.
Excludes data objects with approved key
Excludes data objects with approved primary keys.

Inferred Primary Key Properties

After you run a single data object profile, you can use the Primary Key Profiling view to view the details of the inferred primary keys in the data source.
The following table describes the inferred primary keys properties in the Primary Key Profiling view:
Property
Description
Column
Name of the column in the profile.
% Conforming
Percentage of unique values in the column.
% Duplicates
Percentage of duplicate values for the column.
% Null
Percentage of null values for the column.
Verified
Determines whether the column is a primary key column.
Inference Status
Inference status of the column.
Last Run Time
The date and time that the primary key profile last ran.

Key Violations Properties

After you run a single data object profile, you can use the Primary Key Profiling view to view the details of the primary key violations in the data source.
The following table describes the key violations properties in the Primary Key Profiling view:
Property
Description
Column(s)
Name of the column(s) from which the profile infers a candidate primary key.
Number of Key Violations
Number of key violations in the primary key candidate.