Examine the partial TKPROF output for an SQL statement:
Which two inferences can definitely be made from this output? (Choose two.)
Array fetch operations were not performed for this query.
No hard parse was performed for this query.
The number of logical I/Os is almost equal to the number of physical I/Os.
Another transaction held a shared lock on the table, thereby causing a significant delay.
An uncommitted transaction made a series of updates to the NAME_ID column just before the execution of this query.
Correct answer: BD
Question 2
Which two actions can reduce library cache latch contention for an OLTP application that repeatedly executes queries containing a mix of literals and bind variables? (Choose two.)
setting the OPEN_CURSORS parameter to hold a sufficient number of concurrently open cursors
coding the application such that a frequently executed statement is parsed only once and executed repeatedly as required
setting the CURSOR_SHARING parameter to EXACT
avoiding the granting of privileges on objects associated with cursors during peak load
enabling Automatic Memory Management and allocating at least 50% of the available memory for SHARED_POOL_SIZE
configuring shared server connections
Correct answer: CE
Explanation:
References:
References:
Question 3
You want to generate statistics for new objects added to an OLTP application without affecting old statistics and the sessions that currently use them.
Which two tasks would you perform to test the new statistics? (Choose two.)
Set the OPTIMIZER_USE_PENDING_STATISTICS initialization parameter to TRUE for the session.
Set the STALE_PERCENT statistics preference to zero and then gather statistics.
Set the PUBLISH statistics preference to FALSE and then gather statistics.
Use the DBMS_STATS.PUBLISH_PENDING_STATS procedure to make pending statistics the current statistics.
Set the NO_INVALIDATE statistics preference to FALSE and gather statistics without affecting old statistics.