The following are the areas of concern for tuning:
- Memory Usage
- Data Storage
- Data Manipulation
- Physical Storage
- Logic Storage
- Network Traffic
Tuning Memory Usage: In Oracle 10g, we can use the Automatic Workload Repository (AWR) toolset to gather and manage statistical data but in case of Oracle 11g, we can use new initialization parameters such as MEMORY_TARGET to further control the overall memory used by Oracle. It helps us to tune the database automatically when we don’t have time to read the AWR report.
Tuning Data Manipulation: There are two ways to tune Data manipulation in Oracle one way is through Conventional Path mode and second is Direct Path mode.
Tuning Physical Storage: We need to tune the physical storage because it limits the disk in database. This can be done by: Distributing I/O, Striping and Mirroring.
- Clustering means one than one database server configured for the same user connection. When users connect, one of the server’s responds and connects based on availability. The user is completely ignorant of the fact that there are more than one database servers. It demands high cost due to infrastructure needs. Clustering is hence done on server level.
- Mirroring means, one has many configured databases on the same server. Mirrors are basically copies of the original database. Mirroring is hence done on database level.
WATER MARK is a divided segment of used and free blocks. Blocks which are below high WATER MARK i.e. used blocks, have at least once contained some data. This data might have been deleted later. Oracle knows that blocks beyond high WATER MARK don’t have data; it only reads blocks up to the high WATER MARK during a full table scan.