Re: Query Performance

On 9/17/22 00:28, sivapostgres@yahoo.com wrote:

Hello,

My query is like this   

Select a.field1, a.field2, a.field3

From   (Select a.field1, b.field2, c.field3

        From   table1 a

        Join   table2 b

        on     b.something = a.something

        Join   table3 c

        On     c.something = a.something 

        Where  a.field7 = 'value'

        UNION ALL

        Select a.field4, a.field5, a.field6

        From   table11 a

        Join   table21 b

        On     b.something = a.something

        Where  a.field8 = 'something' ) a

Join   table10 b

On     b.field11 = (Select c.field11

                    From   table10 c

                    Where  c.field10 = a.field1 )                     <- instead of a.field1, if I hardcode value (eg. '100') query runs faster

Join   table21 c

On     c.something = a.something

...

In the above query, If I substitute a value for a.field1, query runs faster.   

Any suggestion/guidance/links to improve the query performance without substituting the value ?

Is there an index on table1.field1?
Have you vacuumed or analyzed table1 lately?
Have you decomposed the query into the smallest possible query that demonstrates the problem?