Costly SQL to run faster
Hi,
Platform info: Datawarehouse Oracle 11.2.0.2 EE on AIX 6.1
Tables are RANGE partitioned by date: 6 tables involved with 3 having 2 millions rows and 3 others with hundreds
Block_size:16K
I have a general question:
I got a case where I had a SQL having a cost of 3000 that never returned because of a missing index (nested loop were involved), during the run I saw that billions of consistent gets occured, then I cancelled the query.
I added the index then the plan changed and the cost became 590,000 and the query returned in 4min...
I know that this is pretty generic info but in case there is a simple explanation
Platform info: Datawarehouse Oracle 11.2.0.2 EE on AIX 6.1
Tables are RANGE partitioned by date: 6 tables involved with 3 having 2 millions rows and 3 others with hundreds
Block_size:16K
I have a general question:
I got a case where I had a SQL having a cost of 3000 that never returned because of a missing index (nested loop were involved), during the run I saw that billions of consistent gets occured, then I cancelled the query.
I added the index then the plan changed and the cost became 590,000 and the query returned in 4min...
I know that this is pretty generic info but in case there is a simple explanation
0