Easy question about indexes

Question

Hello,
I have a table of 100000 rows:
MyTable
MyID number(4,0),
Otherfields ,
Name varchar2(30),
LastName varchar(35),
If I query and just select 60 rows from a total of 100000 rows:
SELECT MyID, OtherFields, Name,LastName
FROM MyTable
WHERE MyID=11
ORDER BY LastName,Name
Actually I need an index for MyID field. But would I need an index for LastName and Name? I think I don't, because I'm not using those fields for filtering. Am I correct?
Thanks!

Answer 1

With this you could see more in detail the issues to consider at the time to index a table.
SQL>
SQL> create table mytable
  2  (
  3  myid number(4,0),
  4  name varchar2(30),
  5  lastname varchar(35));
Table created.
SQL>
SQL> explain plan for
  2  select * from
  3  mytable where
  4  myid=11
  5  order by lastname, name;
Explained.
SQL> select * from table ( dbms_xplan.display );
PLAN_TABLE_OUTPUT
| Id  | Operation            |  Name       | Rows  | Bytes | Cost  |
|   0 | SELECT STATEMENT     |             |       |       |       |
|   1 |  SORT ORDER BY       |             |       |       |       |
|   2 |   TABLE ACCESS FULL  | MYTABLE     |       |       |       |
Note: rule based optimization, 'PLAN_TABLE' is old version
10 rows selected.
SQL> create index i1 on mytable(myid);
Index created.
SQL> explain plan for
  2  select * from
  3  mytable where
  4  myid=11
  5  order by lastname, name;
Explained.
SQL> select * from table ( dbms_xplan.display );
PLAN_TABLE_OUTPUT
| Id  | Operation                    |  Name       | Rows  | Bytes | Cost  |
|   0 | SELECT STATEMENT             |             |       |       |       |
|   1 |  SORT ORDER BY               |             |       |       |       |
|   2 |   TABLE ACCESS BY INDEX ROWID| MYTABLE     |       |       |       |
|   3 |    INDEX RANGE SCAN          | I1          |       |       |       |
Note: rule based optimization, 'PLAN_TABLE' is old version
11 rows selected.
SQL>
SQL>
SQL> create index i2 on  mytable(myid,lastname, name);
Index created.
SQL> explain plan for
  2  select * from
  3  mytable where
  4  myid=11
  5  order by lastname, name;
Explained.
SQL> select * from table ( dbms_xplan.display );
PLAN_TABLE_OUTPUT
| Id  | Operation                    |  Name       | Rows  | Bytes | Cost  |
|   0 | SELECT STATEMENT             |             |       |       |       |
|   1 |  SORT ORDER BY               |             |       |       |       |
|   2 |   TABLE ACCESS BY INDEX ROWID| MYTABLE     |       |       |       |
|   3 |    INDEX RANGE SCAN          | I1          |       |       |       |
Note: rule based optimization, 'PLAN_TABLE' is old version
11 rows selected.
SQL>
SQL> drop index i1;
Index dropped.
SQL>
SQL> explain plan for
  2  select * from
  3  mytable where
  4  myid=11
  5  order by lastname, name;
Explained.
SQL> select * from table ( dbms_xplan.display );
PLAN_TABLE_OUTPUT
| Id  | Operation            |  Name       | Rows  | Bytes | Cost  |
|   0 | SELECT STATEMENT     |             |       |       |       |
|   1 |  SORT ORDER BY       |             |       |       |       |
|   2 |   INDEX RANGE SCAN   | I2          |       |       |       |
Note: rule based optimization, 'PLAN_TABLE' is old version
10 rows selected.
SQL>Creating the index i1 you will have pros and cons regarding to the creation of the index i2 and viceversa. If you want to what are them, you could reply
Joel Pérez
http://www.oracle.com/technology/experts