Indexing for Table Joins - Relational Database Index Design and the Optimizers - page 182

P,C P,C P,C

C1PK C2PK C3PK

C1 C2 C3

1,000 rows 10,000 rows 3,000 rows

P,C F F F

CNO C1PK C2PK C3PK

CUST

1,000,000 rows

Figure 8.26 A four-table join.

8.4. Assume SQL 8.19 is executed a million times in a batch job. Should indexing
be improved? What is the tuning potential? Are there other ways to improve the
performance of this SELECT statement?