Understanding simple queries and the cost model

In this section, we will get started with indexes. To understand how things work, some test data is needed. The following code snippet shows how data can be created easily:

test=# DROP TABLE IF EXISTS t_test;
DROP TABLE
test=# CREATE TABLE t_test (id serial, name text); 
CREATE TABLE 
test=# INSERT INTO t_test (name) SELECT 'hans' 
   FROM generate_series(1, 2000000); 
INSERT 0 2000000 
test=# INSERT INTO t_test (name) SELECT 'paul' 
   FROM generate_series(1, 2000000); 
INSERT 0 2000000

In the first line, a simple table is created. Two columns are used; the first is an auto-increment column that just keeps creating numbers, and the second is a column that will be filled with static values.

In all, 4 million rows have been added:

test=# SELECT name, count(*) FROM t_test GROUP BY 1; 
 name | count  
------+--------- 
 hans | 2000000 
 paul | 2000000 
(2 rows) 

These 4 million rows have some nice properties, which we will be using throughout this chapter. IDs are ascending, and there are only two distinct names.

Let's run a simple query:

test=# \timing 
Timing is on. 
test=# SELECT * FROM t_test WHERE id = 432332; 
   id   | name  
--------+------ 
 432332 | hans 
(1 row) 

Time: 176.949 ms 

In this case, the timing command will tell psql to show the runtime of a query.