Database Reference
In-Depth Information
SAP Application Dataflow in the Oracle Database
In SAP applications, end-user requests are processed by the Oracle database as follows:
1. The end user logs into the SAP system via the SAP front end and executes a business
transaction.
2. The user request is executed by the SAP dialog work process.
3. The Oracle server detects the incoming call, accepts the call, and sets up a dedicated
server process.
4. There is a one-to-one relationship between SAP work processes and Oracle server
(shadow) processes.
5. The user runs the SQL statement and commits the transaction.
6. The server process looks into the shared pool of the Oracle Database for any reusable
SQL.
7. If it finds a reusable SQL, then it verifies authorization.
8.IfitdoesnotfindreusableSQL,thenanew,sharedSQLareaisallocated andisparsed
and processed.
9. The server process then retrieves data from the SGA or the datafile.
10. The Oracle server process modifies the data blocks in the SGA and, when efficient,
the database writer writes the changes permanently to the data files.
11. The log writer records committed transactions to the online redo log files.
12. A reply is sent back to the user regarding the status of the transaction call (success or
error).
