Business Context
In a Fraud Detection scenario, huge transaction history tables are loaded into memory for real-time model training. The test ensures that invoking endSession immediately following a high-memory operation effectively closes the connection, validating that the command remains responsive even when the session holds significant data.
Data Preparation
Creation of a larger dummy transaction dataset.
Copied!
| 1 | |
| 2 | DATA work.transactions; |
| 3 | DO i=1 to 5000; |
| 4 | trans_id=i; |
| 5 | amount=i*10; |
| 6 | category='RETAIL'; |
| 7 | OUTPUT; |
| 8 | END; |
| 9 | |
| 10 | RUN; |
| 11 |
Étapes de réalisation
1
Load large dataset into global scope.
Copied!
| 1 | PROC CAS; |
| 2 | upload path="work.transactions" casout={name="transactions", replace=true, promote=true}; |
| 3 | TABLE.tableDetails TABLE="transactions"; |
| 4 | RUN; |
2
Execute endSession immediately to verify responsiveness under load.
Copied!
| 1 | PROC CAS; |
| 2 | SESSION.endSession; |
| 3 | RUN; |
Expected Result
The session closes immediately despite the memory footprint of the loaded tables. The action does not hang or timeout.