Foreign and primary keys question

Short version:

CREATE TABLE Relation
( stu_id INT NOT NULL REFERENCES Student,
  par_id INT NOT NULL REFERENCES Parent,
  PRIMARY KEY (stu_id, par_id)
) ;

Long version:

• Why use short forms for names, like stu_id and par_id? Why not student_id? Saving typing 3-4 characters? How will you differentiate between parent_id and parameter_id? Or school_id and schoolmaster_id?

• The name "Relation" is not very descriptive for a relationship. (Note also that in relational model terminology, "relation" has a meaning very close to "table".) I couldn't come with a good name though, so we could use "Guardian" or "Student_Parent" (this combination is often used in intersection tables)

• The short version above is just an example. While it is working, it uses a lot of shortcuts, like the inline references. It's far better in my opinion, to name all constraints and declare all (primary, unique, foreign key and check) constraints after the column declarations, like in the long version below.

• It's also good to choose some naming conventions anduse them consistently in all the tables, e.g. Tablename_PK for the primary keys, ReferencedTable_referencingTable_FK for the foreign keys, Something_UQ for the unique constraints, etc.

CREATE TABLE Guardian
(                                        -- columns
  student_id INT NOT NULL, 
  parent_id INT NOT NULL,
                                         -- constraints
  CONSTRAINT Guardian_PK                 -- the name of the PK constraint
    PRIMARY KEY (student_id, parent_id),

  CONSTRAINT Student_Guardian_FK         -- the name of the FK constraint
    FOREIGN KEY (student_id) 
    REFERENCES Student (student_id)
      ON UPDATE CASCADE                  -- the actions of the FK
      ON DELETE RESTRICT,
  CONSTRAINT Parent_Guardian_FK          -- second FK
    FOREIGN KEY (parent_id) 
    REFERENCES Parent (parent_id)
      ON UPDATE CASCADE
      ON DELETE RESTRICT
) ;

Taken from the MSDN article on FOREIGN KEY Constraints

A FOREIGN KEY constraint does not have to be linked only to a PRIMARY KEY constraint in another table; it can also be defined to reference the columns of a UNIQUE constraint in another table.

I think it is perfectly fine to define a UNIQUE CLUSTERED INDEX on an IDENTITY column and define the PRIMARY KEY on different columns, often with a NONCLUSTERED INDEX but you don't have to specify an index.