In a collocated join, corresponding records of files being joined exist on the same node.
The values of the partitioning key of the files being joined are partition compatible. No data needs to be moved to another node to perform the join. This method is only valid for queries where all of the fields of the partitioning keys are join fields and the join operator is the = (equals) operator. Also, the nth field (where n=1 to the number of fields in the partitioning key) of the partitioning key of the first file must be joined to the nth field of the partitioning key of the second file, and the data types of the nth fields must be partition compatible. Note that all of the fields of the partitioning key must be involved in the join. Additional join predicates that do not contain fields of the partitioning key do not affect your ability to do a collocated join.
In the following example, because the join predicate involves the partitioning key fields of both files and the fields are partition compatible, a collocated join can be performed. This implies that matching values of DEPTNO and WORKDEPT are located on the same node.
SQL statement:
SELECT DEPTNAME, FIRSTNME, LASTNAME
FROM DEPARTMENT, EMPLOYEE
WHERE DEPTNO=WORKDEPT
OPNQRYF command:
OPNQRYF FILE((DEPARTMENT) (EMPLOYEE))
FORMAT(JOINFMT)
JFLD((DEPTNO WORKDEPT *EQ))
Records returned by this query:
|
SQL: SELECT DEPTNAME, FIRSTNME, LASTNAME
FROM DEPARTMENT, EMPLOYEE
WHERE DEPTNO=WORKDEPT AND MGRNO=EMPNO
OPNQRYF: OPNQRYF FILE((DEPARTMENT) (EMPLOYEE))
FORMAT(JOINFMT)
JFLD((DEPTNO WORKDEPT *EQ) (MGRNO EMPNO *EQ))
Records returned by this query:
|