Identifying Candidate Primary Keys
The primary key for a relation variable is an attribute set that uniquely identifies each tuple in that relation variable. This property is true for any alternate key, not just the candidate key that is selected to be the primary key for a relation.
Note: You cannot use XML, BLOB, CLOB, ARRAY, VARRAY, or Geospatial columns to define a key or other database constraint (see Chapter 12: “Designing for Database Integrity”).
Primary keys do not denote either of the following properties.
Tuples within a relation are not ordered in any way.
Keep in mind that normalization is a logical process, not a physical implementation of the database.
Primary keys are not used to access rows on disk, though they are a frequent choice as the primary index for tables (see Chapter 8: “Teradata Database Indexes and Partitioning” and Chapter 9: “Primary Indexes and NoPI Objects”). The primary index does define a storage path and at least one access path for table rows.