Tuning shared_pool_size
We have a production database used by 24 x 7 website. Recently it was upgraded to 11g. The night when the upgrade was finished, the instance crashed because of small size of shared_pool_size. We immediately add more physical RAM to the database system and bumped the shared_pool_size to 3GB now. We still observe the continuouslly dropping of free memory by querying the v$sgastat. It is slowly dropping but it is dropping. We are really concerned about another instance crash. I know the tuning of shared_pool_size based on the nature of our application. Does anyone have any experience on tuning shared_pool_size for