We do bulk processing every night. I noticed that if I run SQL statement,
then flush the shared pool, then run it again it still runs alot faster the
second time. Since the blocks have been flushed from memory this leads me to
believe that its the reuse of the outline leading to the performance
increase.
Now the size of our tables typically increases every night. Now the explain
plan will be the same, however, does the stored outline store more than
that? Such as specific data about which blocks to use? Would this be a good
candidate?
Received on Fri May 02 2003 - 18:29:16 CDT
