Why is an Oracle bulk operation (insert, update, delete) considered faster than a row by row operation? Unless I tell explicitly which partition to insert,modify,delete, a bulk operation would still need to be done in a row by row fashion. Is this assumtion not right? Because the new rows to be inserted may need to go to different data blocks, so how is a BULK going to work?
Asked
Active
Viewed 95 times
0
-
"Unless I tell explicitly which partition to insert,modify,delete, a bulk operation would still need to be done in a row by row fashion." Why do you think that? – Mat Aug 27 '13 at 19:05
-
Because a bulk update may consist of 1000 rows and they may be scattered all over the place....so how oracle will do the operation in one shot? – Victor Aug 27 '13 at 19:10
-
What do you mean by "bulk DML operations"? Are you talking about set-based operations in SQL? Or PL/SQL operations using collections? – Justin Cave Aug 27 '13 at 19:20
-
3@Victor This might be helpful http://stackoverflow.com/questions/11453215/oracle-bulk-collect-performance – Jacob Aug 27 '13 at 19:26
-
@Justin...using BULK COLLECT – Victor Aug 27 '13 at 19:44