Which action would you like to make the query use the best plan for the selectivity?

You are administering a database supporting an OLTP application. The application runs a series
of extremely similar queries the MYSALES table where the value of CUST_ID changes.
Examine Exhibit1 to view the query and its execution plan.

Examine Exhibit 2 to view the structure and indexes for the MYSALES table. The MYSALES table
has 4 million records.

Data in the CUST_ID column is highly skewed.
Examine the parameters set for the instance:

Which action would you like to make the query use the best plan for the selectivity?

A.
Decrease the value of the OPTIMIZER_DYNAMIC_SAMPLING parameter to 0.

B.
Us the /*+ INDEX(CUST_ID_IDX) */ hint in the query.

C.
Drop the existing B* -tree index and re-create it as a bitmapped index on the CUST_ID column.

D.
Collect histogram statistics for the CUST_ID column and use a bind variable instead of literal
values.

Explanation:
Using Histograms
In some cases, the distribution of values within a column of a table will affect the optimizer’s
decision to use an index vs. perform a full-table scan. This scenario occurs when the value with a

where clause has a disproportional amount of values, making a full-table scan cheaper than index
access.
A column histogram should only be created when we have data skew exists or is suspected.