Question

I have over million records in these tables in both the databases.
I am trying to figure out data in both the tables acros databases.

SELECT COUNT(*) FROM DB1.MYTABLE WHERE SEQ_NO NOT IN(SELECT SEQ_NO FROM DB2.MYTABLE) AND FILENAME NOT LIKE '%{%'  
and PT_TYPE NOT IN(15,24,268,284,285,286,12,17,9,290,214,73) AND STTS=1

The query is taking ages. Is there any way I can make it fast?

Appreciate your help in advance

Answer 1

Do you actually mean different databases? Or do you mean different schemas? You talk about different databases but the syntax appears to be using tables in two different schemas, not two different databases. I don’t see any references to a database link which would be needed if there were two different databases but perhaps DB2.MYTABLE is supposed to be a synonym for MYTABLE@DB2.

It would be helpful if you could post the query plan that is generated. It would also be useful to indicate what indexes exist and how selective each of these predicates is. My guess is that modifying the query to be

SELECT count(*)
  FROM schema1.mytable a
 WHERE NOT EXISTS (
    SELECT 1
      FROM schema2.mytable b
     WHERE a.seq_no = b.seq_no )
   AND a.filename NOT LIKE '%{%'
   AND a.pt_type NOT IN (15,24,268,284,285,286,12,17,9,290,214,73)
   AND a.stts = 1

might be more efficient if most of the rows in SCHEMA1.MYTABLE are eliminated because the SEQ_NO exists in SCHEMA2.MYTABLE.