054 - Quiz & Trivia

1. You plan to use adaptive thresholds as part of the performance tuning activity. You decide to increase the window size of the default moving window baseline for all metric observations and comparisons in your database. The following error occurs when you try to increase the window size through Enterprise Manager: Failed to commit: ORA-13541: system moving window baseline size (1296000) greater than retention (1036800) ORA-06512: at "SYS.DBMS_WORKLOAD_REPOSITORY", line 601 ORA-06512: at line 2 Which action would allow you to perform the preceding task successfully?

Increasing the flashback retention period

Increasing the retention period for SQL Management Base

Increasing the undo retention period for the database instance

Increasing the Automatic Workload Repository (AWR) retention period

2. Your system has been upgraded from Oracle Database 10g to Oracle Database 11g. You imported SQL Tuning Sets (STS) from the previous version that contains important SQL statements. You are not sure about the elapsed time that the SQL statements will have in the new system due to the version change of the optimizer. Which tool provides a comparative report with respect to the elapsed time in both versions of the optimizer?

SQL Access Advisor

SQL Tuning Advisor (STA)

SQL Performance Analyzer (SPA)

Automatic Workload Repository (AWR) compare period report

3. Which is the correct description of SQL profiling while using SQL Tuning Advisor?

It is a set of recommendations by the optimizer to create new indexes.

It is auxiliary information collected by the optimizer for a SQL statement to eliminate estimation error.

It is a set of recommendations by the optimizer to refresh stale statistics to avoid bad execution plan.

It is a set of recommendations by the optimizer to restructure a SQL statement to avoid bad execution plan

4. You used SQL Tuning Advisor for a long-running SQL statement that suggested a SQL profile which can be used by the query subsequently for a better execution plan. After implementing the profile, you altered the profile to include it in a category as follows: SQL> EXEC DBMS_SQLTUNE.ALTER_SQL_PROFILE (name=> 'SYS_SQLPROF_0146ced728214000', attribute_name => 'CATEGORY', value => 'DEV2'). But the performance has not improved for the user session. The profile is enabled and on investigating the query plan, you find that the query from the user session does not use the SQL profile. View the Exhibit and examine the parameters for the user session. What could be the reason for this? Exhibit:

The CONTROL_MANAGEMENT_PACK_ACCESS parameter is not set to DEV2.

The OPTIMIZER_MODE parameter is set to ALL_ROWS for the user session.

The SQLTUNE_CATEGORY parameter is set to DEFAULT for the user session.

The OPTIMIZER_USE_PLAN_BASELINES parameter is set to TRUE for the user session.

5. View the Exhibit exhibit1 to examine the series of SQL commands and parameter settings. View the Exhibit exhibit2 to examine the plans available in the SQL plan baseline. The first plan (in red) is created when OPTIMIZER_MODE is set to ALL_ROWS and the second plan (in blue) is created when OPTIMIZER_MODE is set to FIRST_ROWS. Which SQL plan baseline would be used if the SQL query in exhibit1 is executed again when the value of OPTIMIZER_MODE is set to FIRST_ROWS? Exhibit 1:

Exhibit 2:

The second plan, because it is a fixed plan

The first plan, because it is an accepted plan

The second plan, because it is the latest generated plan in FIRST_ROW mode

A new plan, because the second plan in FIRST_ROW mode is not an accepted plan

6. You are working on a development database that was upgraded to Oracle Database 11g from Oracle Database 9i. An ADDM finding in this database says that the shared pool is inadequately sized, as shown in the Exhibit. You diagnosed that this is due to different kinds of workloads and this occurs only during peak hours. You tried to resize this by shrinking the database buffer cache but that caused inadequate buffer cache problems. The following are the related parameter settings: SQL> show parameter sga NAME TYPE VALUE ------------------------------ --------------- ----------------- lock_sga boolean FALSE pre_page_sga boolean FALSE sga_max_size big integer 300M sga_target big integer 0 SQL> show parameter target NAME TYPE VALUE ------------------------------ --------------- ----------------- ..... fast_start_mttr_target integer 0 memory_max_target big integer 0 memory_target big integer 0 pga_aggregate_target big integer 100M sga_target big integer 0 You want to balance the memory between the System Global Area (SGA) components within SGA without affecting the size of the Program Global Area (PGA). Which action would solve this problem? Exhibit 1:

Set the SGA_TARGET parameter to 300M

Set the SGA_MAX_SIZE parameter to 400M

Set the MEMORY_TARGET parameter to 100M

Set the MEMORY_MAX_TARGET parameter to 300M

7. You are working on a decision support system (DSS). The index is available on the COUNTRY_ID column of the CUSTOMERS table. View the Exhibit and examine the parameter settings and the query execution plan. Why is the query using a full table scan instead of an index scan? Exhibit:

Because the histogram statistics for the COUNTRY_ID column are not updated

Because the index statistics for the index on the COUNTRY_ID column are not current

Because the DB_FILE_MULTIBLOCK_READ_COUNT initialization parameter is set to a higher value

8. You have the automatic optimizer statistics gathering enabled for a 24x7 database as part of the automatic maintenance task. You observed that a few tables are being updated regularly, but not being queried often. However statistics are being gathered for these tables automatically and regularly, thereby affecting the database performance. What would you do to increase the interval of statistics collection for these tables so that the statistics collection jobs spend less time during the automatic maintenance task?

Set the OPTIMIZER_USE_PENDING parameter statistics to TRUE.

Increase the value of the STALE_PERCENT preference for these tables.

Mark those associated indexes that are less frequently used as invisible.

Increase the value of the ESTIMATE_PERCENT preference for these tables

9. You observe that suboptimal execution plans for the queries are being generated on a table that previously used less resources. You have collected statistics on these tables two days ago. The optimizer statistics retention period is set to 31 days. You are able to find the timestamp information about statistics update from the DBA_TAB_STATS_HISTORY view. Because it is a frequently queried table, you would like the optimizer to generate better plans. Which action would enable you to use the previous set of statistics on the objects that may lead to better execution plans?

Restoring statistics from statistics history up to the desired time

Deleting all AWR snapshots collected after the time of desired statistics collection

Applying the flashback table technique until the time of desired statistics collection

Setting the OPTIMIZER_PENDING_STATISTICS parameter to TRUE to use the previous version of statistics

10. Your company wants to migrate a single instance Oracle 11g Release 1 database to a RAC environment. The database supports middle-tier applications using connection pooling for connecting to database. Being a part of the performance management team, you want to test the performance of the SQL statements and the peak workload of the applications in the new environment before upgrading the production system. Given below are some of the steps in random order to accomplish the task: 1) Capture the real workload on the existing system. 2) Create a SQL Tuning Set (STS) for the statements executed by the application. 3) Transport STS on the test environment and use SQL Tuning Advisor to compare the before and after performance. 4) Replay the workload on the test environment. 5) Analyze the divergence after replay. 6) Use Automatic Workload Repository (AWR) to capture the information for top SQL on the existing system. 7) Transport STS on the test environment and use SQL Performance Analyzer to compare the before and after performance. Identify the correct steps required to perform the test.

2 and 3 only

6 and 7 only

1, 4, and 5 only

2, 3, and 5 only

2, 3, 4, and 5 only

11. You used SQL Tuning Advisor for a long-running SQL statement that suggested a SQL profile which can be used by the query subsequently for better execution plan. View the Exhibit. But you want certain user sessions not to use this SQL profile for their queries. How would you implement this? Exhibit:

Alter the SQL profile to change the category of the SQL profile.

Set the OPTIMIZER_USE_PENDING_STATISTICS to TRUE the desired sessions.

Use database resource manager to prevent the use of the SQL profile by these user sessions.

12. You work as a DBA for a company and you have the responsibility of managing one of its online transaction processing (OLTP) systems. The database encountered performance-related problems and you generated an Automatic Workload Repository (AWR) report to investigate it further. View the Exhibits and examine the AWR report. What could be the problem in this database? Exhibit 1:

Exhibit 2:

Exhibit 3:

Exhibit 4:

Java pool is not configured

The CPU in the system is slow

The shared pool size is inadequate

The database buffer cache is inadequate

The OPEN_CURSORS parameter is set to a small value

13. Examine the initialization parameter values for the instance given below: NAME TYPE VALUE ------------------------------------ ----------- ------------------ optimizer_capture_sql_plan_baselines boolean FALSE optimizer_dynamic_sampling integer 2 optimizer_features_enable string 11.1.0.6 optimizer_index_caching integer 0 optimizer_index_cost_adj integer 100 optimizer_mode string ALL_ROWS db_file_multiblock_read_count integer 64 The index created on the column used in the WHERE clause of the query. You notice that the query is not using the index. Instead of an index scan, a full table scan is used. View the Exhibit and examine the autotrace output for a query. What could be the reason for it? (Choose all that apply.) Exhibit:

The OPTIMIZER_INDEX_COST_ADJ initialization parameter has a low value.

The DB_FILE_MULTIBLOCK_READ_COUNT initialization parameter has a low value.

The statistics for the table and all the indexes associated with the table are not current.

The table has less than DB_FILE_MULTIBLOCK_READ_COUNT blocks under the high-water mark.

Submit

14. The columns CUST_CITY, CUST_STATE_PROVINCE, and COUNTRY_ID are frequently used together in the WHERE clause of the queries. The CUSTOMERS table is a big table with 20 GB of data. You notice that the selectivity for these three columns varies from the selectivity that the optimizer calculates. What would you recommend to influence the selectivity calculated by the optimizer?

Creating function-based indexes by concatenating all the columns

Updating histogram statistics for these columns by using the DBMS_STATS.GATHER_TABLE_STATS procedure

Using the DBMS_STATS.CREATE_EXTENDED_STATS function to create a virtual column and create index on the virtual column

Using the DBMS_STATS.CREATE_EXTENDED_STATS function to create a virtual column and the DBMS_STATS.GATHER_TABLE_STATS procedure to collect statistics on the virtual column

15. The database application developers are planning to make some major schema changes such as creating new indexes and materialized views. They want to check the net impact of these changes on the workload performance. This activity has to be performed in the production database, so they want only the query part of the data manipulation language (DML) statements to be executed so that the side effects to the database or user data can be prevented. What should they use to achieve this?

Database Replay

SQL Tuning Advisor

SQL Access Advisor

SQL Performance Analyzer

16. You have the automatic optimizer statistics gathering enabled for a 24x7 database as part of the automatic maintenance task. Because the automatic optimizer statistics collection runs during maintenance window (in the night), the statistics on tables that are significantly modified throughout the day becomes stale. This leads to poor query performance on these tables and you want to minimize the overhead in statistics collection. Identify a solution to this problem.

Change the STALE_PERCENT preference to zero for the tables.

Set the OPTIMIZER_USE_PENDING parameter statistics to TRUE.

Set the OPTIMIZER_DYNAMIC_SAMPLING parameter to zero during day time.

Gather statistics on the table when the tables have a representative number of rows, and then lock the statistics.

17. The SALES table in your database undergoes frequent changes. While observing the statistics for the table in the middle of the day, you find a change since the last observation. But the information about the statistics collection is not recorded in the DBA_TAB_STATS_HISTORY. Identify the reason for this.

The AWR retention time was set to zero.

The STATISTICS_LEVEL parameter was set to BASIC.

Some of the statistics were collected using the ANALYZE command.

There were user-defined statistics that were set by using the DBMS_STATS.SET_TABLE_STATS procedure.

18. Identify the two direct sources from where SQL plans can be loaded into the SQL plan baselines. (Choose two.)

Cursor cache

Stored outline

SQL Tuning Set

Automatic Workload Repository (AWR) snapshots

Submit

19. Your system is upgraded from Oracle Database 10g to Oracle Database 11g. You imported SQL Tuning Sets (STS) from the previous version. After running the SQL Performance Analyzer for optimizer version change, you observe performance regression for a few SQL statements. What would you suggest to avoid performance regression for these SQL statements?

Include the old plans in the SQL plan baseline.

Use DBMS_SPM.EVOLVE_SQL_PLAN_BASELINE to evolve plans.

Set the OPTIMIZER_USE_PENDING_STATISTICS parameter to TRUE.

Increase the value for the OPTIMIZER_DYNAMIC_SAMPLING parameter.

20. You are a DBA in ABC Corp. You are working on an online transaction processing (OLTP) system. The applications running on the database use connection pooling to connect to the database to perform transactions. The company wants to upgrade the CPU and the memory for the machine on which the production database is running. To test the realistic workload on the new machine with increased CPU and memory, you identified the peak workload time for the database and captured the workload on the production system. You set up the test machine with the upgraded CPU and memory, and performed the following operations for replay: 1) You preprocessed the captured workload. 2) You restored the replay database to match the capture database at the start of the workload capture. 3) You resolved all external references. 4) You set up replay clients. You have the following replay considerations: 1) load testing to be performed 2) all the sessions to connect immediately 3) user waits between issuing calls to be kept to a minimum What replay options would you choose to accomplish the replay?

SYNCHRONIZATION = FALSE, THINK_TIME_SCALE = 0, CONNECT_TIME_SCALE = 0 and THINK_TIME_AUTO_CORRECT = FALSE

SYNCHRONIZATION = TRUE, THINK_TIME_SCALE = 100, CONNECT_TIME_SCALE = 0 and THINK_TIME_AUTO_CORRECT = FALSE

SYNCHRONIZATION = FALSE, THINK_TIME_SCALE = 100, CONNECT_TIME_SCALE = 100 and THINK_TIME_AUTO_CORRECT = TRUE

SYNCHRONIZATION = TRUE, THINK_TIME_SCALE = 100, CONNECT_TIME_SCALE = 100 and THINK_TIME_AUTO_CORRECT = FALSE

21. Examine the query output shown below: SQL> SELECT sum(pins), sum(reloads), sum(reloads)/sum(pins) FROM v$librarycache; SUM(PINS) SUM(RELOADS) SUM(RELOADS)/SUM(PINS)

----------	------------	----------------------
16479	16000	0.9709327

Which statement describes the correct interpretation of the reloads-to-pin ratio and the action to be taken?

This ratio is reasonable. However, decrease the shared pool size for better performance.

This ratio is very high and the shared pool size should be increased immediately to reduce this ratio.

This ratio is reasonable. However, this needs monitoring. Increase the shared pool size only if the ratio crosses 1.

This ratio is reasonable. However, this needs monitoring. Increase the shared pool size only if the ratio falls below 0.1.

22. You work as a DBA for a company and manage an online transaction processing (OLTP) system. You received complaints about the performance degradation of SQL statements executed by the application that uses this database. View the Exhibit and examine the output of the queries executed to investigate the performance issues. Which two actions can improve performance? (Choose two.) Exhibit:

Increasing the size of the shared pool

Setting the CURSOR_SHARING parameter to EXACT

Decreasing the value set for the OPEN_CURSORS parameter

Replacing literals with bind variables in SQL statements with the PLAN_HASH_VALUE 1337874392

Replacing literals with bind variables in SQL statements with the PLAN_HASH_VALUE 1445457117

Submit

23. You work as a DBA for a company and you have the responsibility of managing one of its online transaction processing (OLTP) systems. The database encountered performance-related problems and you generated an Automatic Workload Repository (AWR) report to investigate it further. View the Exhibits and examine the AWR report. Which is the appropriate solution to the problem in this database? Exhibit 1:

Exhibit 2:

Exhibit 3:

Exhibit 4:

Increasing the size of the shared pool

Adding one more CPU to the system

Setting the CURSOR_SHARING parameter to EXACT

Configuring Java pool because it is not configured

24. You work on an online transaction processing (OLTP) database in which the SALES table has 10,000 rows but only four distinct products are sold. View the Exhibit named HIST to check the distribution of values in the table and the histograms on the table. View the Exhibits named QUERY-1 and QUERY-2 that show details in the V$SQL view for the queries executed on the SALES table. You check the plan table and notice that both the queries that executed on the SALES table used index range scan. The second query retrieved most of the rows in the table but used index range scan. Why would the second query use the same plan? Exhibit 1:

Exhibit 2:

Exhibit 3:

Because the plan was bind aware

Because the bind peeking never happened

Because the OPEN_CURSORS parameter is set to a very low value

Because the optimizer did not consider selectivity due to the use of bind variables

25. You are working for a company that uses huge applications to manage its customer details. During your regular performance checks, you executed the following query: SQL> show parameter shared_pool_reserved_size NAME TYPE VALUE ---------------------------------------------------- shared_pool_reserved_size big integer 6920601 SQL> SELECT free_space, used_space, request_misses, request_failures from V$SHARED_POOL_RESERVED; FREE_SPACE USED_SPACE REQUEST_MISSES REQUEST_FAILURES ---------- ---------- -------------- ---------------- 6469776 555528 0 113 You observed that REQUEST_FAILURES is 113 but not increasing, and also that REQUEST_MISSES is not increasing. What would you interpret from this?

The reserved pool is very large. Consider reducing the size of the shared pool.

The reserved pool is large and has free space to satisfy requests. Consider reducing the size of the reserved pool

26. You work for a company as a DBA. The company has an application to manage the details of its business and customer base. However, application users complain that updating the SALES record takes more time than it used to earlier. On investigation, you notice that the application performance degrades when a call is made to the UPDATE_INV procedure. Further, you execute the following query to investigate: SQL> SELECT name, type, sharable_mem, kept 2 FROM v$db_object_cache 3 WHERE sharable_mem > 4000 4 AND EXECUTIONS > 5 5 AND (type='FUNCTION' OR type='PROCEDURE') 6 / NAME TYPE SHARABLE_MEM KEPT ------------------- -------------- --------------- ----------- GETEMKEY FUNCTION 13695 YES UPDATE_INV PROCEDURE 14766 NO SETEMUSERCONTEXT PROCEDURE 13703 YES DECRYPT FUNCTION 17790 YES UPDATE_DEPT PROCEDURE 18765 NO What would you do to improve the performance?

Enable the result cache if not already enabled.

Flush the shared pool to make space for this procedure.

Keep the UPDATE_INV procedure by using the DBMS_SHARED_POOL.KEEP procedure.

Increase the size of the keep buffer pool to accommodate the UPDATE_INV procedure.

Ask the developers to modify and use literals instead of bind variables in the UPDATE_INV procedure.

27. View Exhibit1 and examine the indexes on the CUSTOMERS table. The statistics for the CUSTOMERS table have been updated recently by using the following command: SQL> EXEC DBMS_STATS.GATHER_TABLE_STATS('SH','CUSTOMERS',method_opt=>'FOR ALL INDEXED COLUMNS SIZE AUTO'); View Exhibit2 to examine a query plan. Even though the index is present on the COUNTRY_ID and CUST_GENDER columns, the query uses a full table scan. What could be the reason? Exhibit 1:

Exhibit 2:

Because the histogram statistics for the COUNTRY_ID column are not updated

Because the DB_FILE_MULTIBLOCK_READ_COUNT initialization parameter is set to a high value

28. Examine the following block that executes a SQL Performance Analyzer task: begin dbms_sqlpa.execute_analysis_task(task_name => 'SPA_SKD4', execution_type => 'TEST EXECUTE', execution_name => 'before'); end; Which two statements are true about the execution? (Choose two.)

It generates execution plans and execution statistics.

It executes only the query part of the DML statements.

The test plans are generated and added to plan baseline.

The test plans are executed, but not generated if they are available in the plan baseline.

Submit

29. Identify two correct statements about the Active Session History (ASH) data. (Choose two.)

A part of SGA memory is used to store ASH data as rolling buffer.

The ASH data can be analyzed between any two small time intervals.

All ASH data in memory are flushed to disk by MMON in every 3 seconds.

All ASH data in memory are flushed to disk by MMNL process whenever the buffer is full.

Submit

30. You are working on an online transaction processing (OLTP) system. By day most of the application users perform queries accessing the most recently added or modified rows. The applications have most of the queries based on multiple tables. But at night, some batch processing is also done. Which two actions would you recommend to choose a goal for the optimizer based on the needs of your application? (Choose two.)

Setting the OPTIMIZER_MODE parameter to ALL_ROWS at the instance level

Setting the OPTIMIZER_MODE parameter to FIRST_ROWS_n at the instance level

Asking the developer to add a hint /*ALL_ROWS*/ in the long-running batch processing queries

Asking the developer to add a hint /*FIRST_ROWS_n*/ in the long-running batch processing queries

Submit

31. Which three factors influence the optimizer's behavior while choosing an optimization approach and goal for a SQL statement? (Choose three.)

Parsing of a SQL statement

Operating system (OS) statistics

Object statistics in the data dictionary

The OPTIMIZER_MODE initialization parameter

Optimizer SQL hints for changing the query optimizer goal

Submit

32. A user session executes a query and the plan for the query is shown in the Exhibit as Plan-1. After verifying with the SQL Access advisor, an index is created on the table in the JOB_ID column and the query is executed again. A new plan is generated against the second query shown in the Exhibit as Plan-2. When explaining the plan for the second query, you observe that the optimizer uses the first plan instead of the second. The following parameters are set for the user session: OPTIMIZER_CAPTURE_SQL_PLAN_BASELINES=TRUE OPTIMIZER_USE_SQL_PLAN_BASELINES=TRUE SQLTUNE_CATEGORY=DEFAULT Why does the optimizer not use the second plan? Exhibit:

Because the second plan is not verified

Because the second plan is not part of the SQL plan baseline

Because the SQLTUNE_CATEGORY parameter is set to DEFAULT

33. The OPTIMIZER_CAPTURE_SQL_PLAN_BASELINES parameter is set to TRUE for a 24x7 database instance. The space consumption has significantly increased by the automatically captured plan baselines. You are suggested to run the following block: BEGIN dbms_spm.configure('SPACE_BUDGET_PERCENT', 30); END; After a few weeks of work, you observe warnings on exceeding space usage by SQL Management Base (SMB) in alert log file. Which two actions can be solutions for this? (Choose two.)

Increasing the AWR retention period

Purging the older SQL management objects

Increasing the size of the SYSTEM tablespace

Increasing the size of the SYSAUX tablespace

Submit

34. You are a DBA in ABC Corp. You are working on a DSS system. The applications directly connect to the database to perform transactions. The application users also perform direct loads. The company wants to upgrade from Oracle Database 10g version 10.2.0.1 to Oracle Database 11g. You want to test the performance of the SQL statements in the new environment before upgrading the production database. To accomplish this, you set up the test database with Oracle Database 11g, with the same data and schemas as in production. What would you recommend to accomplish the task?

Gathering all the SQL statements and using SQL Tuning Advisor on the test database for performance analysis

Capturing the workload on the production database, replaying it on the test machine, and comparing the performance

35. Examine the output of the following query: SQL> SELECT c.name,a.addr,a.gets,a.misses,a.sleeps, 2 a.immediate_gets,a.immediate_misses,b.pid 3 FROM v$latch a, v$latchholder b, v$latchname c 4 WHERE a.addr = b.laddr(+) and a.latch# = c.latch# 5 AND c.name LIKE '%latch_name%' ORDER BY a.latch#; LATCH NAME ADDR GETS MISSES SLEEPS IMMEDIATE_GETSIMMEDIATE_MISSES ------------ -------- -------- ------- ------ -------------- ---------------- shared pool 200165448520540 14112 3137 0 0 You calculated the Gets-to-Misses ratio, which is 99834, and this ratio is dropping over a period of time. Which two actions can improve this? (Choose two.)

Reducing hard parses

Increasing the size of the shared pool

Using only dedicated server connections

Setting the CURSOR_SHARING parameter to EXACT

Encouraging the use of more literal SQL statements

Submit

36. You observed that some of the queries are performing poorly on the SALES_RECORDS table in your database. On further investigation, you find that at the end of each day the contents of the SALES_RECORDS table are transferred to the SALES table and deleted from the SALES_RECORDS table. The deleted operations cause the table to be sparsely populated. You decided to use the ALTER TABLE...SHRINK SPACE COMPACT command to shrink the table. Why would you choose this method? (Choose all that apply.)

Because it can be used during peak hours

Because it avoids unnecessary cursor invalidation

Because it adjusts the high-water mark (HWM) immediately

Because you have long-running queries that might span the shrink operation

Because it does not allow any data manipulation language (DML) operations, thereby making the shrink operation faster

Submit

37. You work as a consultant DBA for various clients. A performance issue in one of the online transaction processing (OLTP) systems is reported to you and you received the Automatic Workload Repository (AWR) report generated in the database. The main sections of the AWR report are shown in the Exhibits. View the Exhibit named DBTIME and note % DB Time. View the Exhibit named TIMEMODEL and note what has contributed to % DB Time. View the Exhibit named EFFICIENCY and examine the various percentages shown. Identify the option that has the correct answers for the questions given below: 1) Which factor indicates the main problem? 2) What is the main problem? 3) What solution would you recommend? Exhibit 1 (exhibit):

Exhibit 2:

Exhibit 3:

38. View the Exhibit and examine a portion of the output obtained from the following query: SQL> SELECT * FROM v$sys_time_model; Select three correct interpretations of the time model statistics. (Choose three.) Exhibit:

DB time includes the wait time of all the nonidle and idle user sessions.

SQL execute elapsed time includes the time spent in performing fetches of query results.

DB CPU includes the CPU time spent on database user-level calls and background CPU time.

SQL execute elapsed time includes components of the hard parse elapsed time like bind elapsed time.

DB time includes the connection management call elapsed time excluding the background process time.

Submit

39. A batch workload that historically completed in the maintenance window between 10:00 PM and midnight is currently showing poor performance and completing at 2 AM. To help in the diagnosis of the performance degradation, the senior DBA in your organization asks you to execute the awrddrpt.sql script to produce a Compare Periods report. Which two statements are true about the report produced by this script? (Choose two.)

It is refreshed automatically based on the moving window baseline

It compares details between any two selected time periods of the same duration.

It normalizes the statistics by the amount of time spent on the database for each time period.

It compares details between two consecutive time periods of the same or different durations and is refreshed every 60 minutes.

Submit

40. View Exhibit1 to examine the description of the CUSTOMERS table. You observed that optimizer selectivity is not accurate when the CUST_STATE_PROVINCE and COUNTRY_ID columns are used together in the WHERE clause of a query. View Exhibit2 to examine the query execution plan and the commands executed to gather the statistics. The optimizer predicts that 20 rows will be processed rather than the 3,341 rows, which is the actual number of rows returned from the table. What can you do to make the optimizer detect the actual number of rows? Exhibit 1:

Exhibit: 2

Set the STATISTICS_LEVEL parameter to ALL.

Set the OPTIMIZER_USE_PENDING_STATISTICS parameter to FALSE.

Create extended statistics for the CUST_STATE_PROVINCE and COUNTRY_ID columns.

Increase the STALE_PERCENT value for the CUSTOMERS table by using the DBMS_STATS.SET_TABLE_PREFS procedure.

41. View Exhibit1 to examine the description of the CUSTOMERS table. The CUSTOMERS table has been updated heavily today. In a frequently used SQL statement, you notice that estimated rows and the actual number of rows fetched differ greatly. The COUNTRY_ID column has an index. View Exhibit2 and examine the query execution plan. What would you recommend to improve the optimizer's estimation? Exhibit 1:

Exhibit 2:

Setting the STATISTICS_LEVEL parameter to ALL

Setting the OPTIMIZER_USE_PENDING_STATISTICS parameter to FALSE

Creating extended statistics for the CUST_LAST_NAME, CUST_ID, and CUST_TOTAL columns

Updating the statistics for the CUSTOMERS table by using the DBMS_STATS.GATHER_TABLE_STATS procedure

42. Which three statements are true about performance analysis by SQL Performance Analyzer? (Choose three.)

It detects changes in SQL execution plans.

It shows only the overall impact on workload.

It produces results that can be used to create the SQL plan baseline.

It generates recommendations to run SQL Tuning Advisor to tune regressed SQL statements.

The importance of SQL statements to be analyzed is based on the size of the objects accessed.

Submit

43. You have a range-partitioned table in your database. Each partition in the table contains the sales data for a quarter. The partition related to the current quarter is modified frequently, whereas other partitions undergo fewer data manipulations. The preferences for the table are set to their default values. You collect statistics for the table using the following command at regular intervals: SQL> EXECUTE DBMS_STATS.GATHER_TABLE_STATS('SH','SALES',GRANULARITY=>'AUTO'); You need statistics to be collected more quickly. How do you achieve this?

Set the PUBLISH preference to TRUE for the partition table.

Set the NO_VALIDATE preference to TRUE for the partition table.

Set the INCREMENTAL preference to TRUE for the partition table.

Increase the value of the STALE_PERCENT preference for the partition table.

44. View the Exhibit and examine the result of Automatic SQL Tuning for a period. None of the recommended SQL profiles are implemented. What would you do to allow the Automatic SQL Tuning implement the SQL profiles automatically? Exhibit:

Set the OPTIMIZER_MODE parameter to ALL_ROWS for the database instance.

Set the SQLTUNE_CATEGORY parameter to DEFAULT for the database instance.

Set the OPTIMIZER_CAPTURE_SQL_PLAN_BASELINES parameter to TRUE for the database instance.

Use the DBMS_SQLTUNE.SET_TUNING_TASK_PARAMETERS procedure to set ACCEPT_SQL_PROFILES to TRUE.

45. Identify the type of recommendation that can automatically be implemented by the Automatic Tuning Advisor as part of automatic SQL tuning task?

Recommendation for SQL profiles

Recommendation about effective indexing

Recommendation about stale or no statistics

Recommendation about the structure of SQL statements

46. Examine the output of the query given below: SQL> SELECT mutex_type, location, sum(gets), sum(sleeps) FROM v$mutex_sleep_history GROUP BY mutex_type, location; MUTEX_TYPE LOCATION SUM(GETS) SUM(SLEEPS) -------------- -------------------- ----------------- ---------- ----------- Library Cache kglhdgn1 62 8669586 4538 Library Cache kglget2 2 2016618 24 Cursor Stat kkocsStoreBindAwareStats [KKSSTALOC8] 2975 1 Cursor Pin kkslce [KKSCHLPIN2] 666831 678 Library Cache kgllkdl1 85 3369224 110 Library Cache kglpnal1 90 224199 13 Library Cache kglic1 49 42068 10 Library Cache kglpin1 4 9620087 374 Library Cache kglpndl1 95 2065089 79 9 rows selected. Which statement is true?

Each row in the output represents a SQL statement that had to wait for mutexes.

The Cursor Stat and Cursor Pin SLEEPS indicate that the CURSOR_SHARING parameter is set to EXACT

47. After running SQL Performance Analyzer (SPA), you observe a few regressed SQL statements in the SPA output. Identify the two actions that you would suggest for these regressed SQL statements. (Choose two.)

Running SQL Access Advisor

Adding them to SQL Plan Baseline

Submitting them to SQL Tuning Advisor

Running Automatic Database Diagnostic Monitor (ADDM)

Submit

48. View the Exhibit and examine the plans in the SQL plan baseline. You observe that the last two SELECT statements (outlined in red and blue) are identical, but have two different plans. Also, the second plan (in blue) is not being used by the optimizer. Which two steps do you perform to ensure that the optimizer uses the second plan (in blue) if the plan is more efficient? (Choose two.) Exhibit:

Mark the second plan (in blue) as fixed.

Use SQL Tuning Advisor to tune the plan.

Use SQL Access Advisor to tune the plan.

Use the DBMS_SPM.EVOLVE_SQL_PLAN_BASELINE function to evolve the plan.

Submit

49. A user in a session executed the following SQL statement to set the optimizer mode: ALTER SESSION SET OPTIMIZER_MODE = ALL_ROWS What impact would it have on the goal of the optimizer for that session? (Choose all that apply.)

Statement level OPTIMIZER_MODE hints take precedence over the session-level setting.

The OPTIMIZER_MODE parameter set at instance level takes precedence over the session-level value.

Submit

50. Examine the initialization parameter values for the instance given below: NAME TYPE VALUE ------------------------------------ ----------- ------------------ optimizer_capture_sql_plan_baselines boolean FALSE optimizer_dynamic_sampling integer 2 optimizer_features_enable string 11.1.0.6 optimizer_index_caching integer 0 optimizer_index_cost_adj integer 100 optimizer_mode string ALL_ROWS db_file_multiblock_read_count integer 64 You notice that the one of the queries is using a full table scan (view Exhibit1) instead of index unique scan (view Exhibit2). The index is present on the column that is accessed in the WHERE clause of the query. The cost for a full table scan is more than that for an index unique scan. Why would the optimizer choose full table scan over index unique scan? (Choose all that apply.) Exhibit 1:

Exhibit 2:

The OPTIMIZER_INDEX_COST_ADJ initialization parameter is set to a low value.

The OPTIMIZER_INDEX_COST_ADJ initialization parameter is set to a high value.

The DB_FILE_MULTIBLOCK_READ_COUNT initialization parameter is set to a low value.

The statistics for the table and all the indexes associated with the table are not current.

Submit

51. You receive a recommendation for a SQL statement through the automatic SQL tuning process that suggests implementing a SQL profile. You implement the SQL profile. The tables associated with the SQL statement grow, and indexes are created and dropped on these tables. What would happen to the SQL profile that is created for the SQL statement?

It becomes outdated immediately after the tables change.

It is maintained automatically along with the changes in the tables.

The profile will be used for the changed plans of the SQL statement.

The plan for the SQL statement is frozen until it is regenerated by the SQL Tuning Advisor.

52. You upgraded database from the Oracle Database 10g to Oracle Database 11g. To test the performance SQL on the upgraded database. You want to build the before-change performance data by using SQL Performance Analyzer. Which method would allow the task to execute quickly?

The EXPLAIN PLAN method

The TEST EXECUTE method

The COMPARE PERFORMANCE method

The OPTIMIZER_USE_SQL_PLAN_BASELINE parameter set to TRUE

53. You are a DBA in ABC Corp. You are working on an online transaction processing (OLTP) system. The applications running on the database use connection pooling to connect to the database to perform transactions. The company wants to upgrade the CPU and the memory for the machine on which the database is running. To test the realistic workload on the new machine with increased CPU and memory, you identified the peak workload time for the database and captured the workload on the production system. You set up the test machine with the upgraded CPU and memory and installed the database with the same version as production. Which three actions would you perform to accomplish the realistic replay of the peak workload on the test machine? (Choose three.)

Preprocess the captured workload for the specified target database version.

Set up as many replay clients as there are actual clients to the production system.

Set up all the replay clients on the same machine on which the test database is running.

Resolve all external references such as database links, external tables, and directory objects.

Restore the replay database to match the capture database at the start of the workload capture.

Submit

54. You work as a DBA for a company and as a performance improvement measure, you implemented the result cache in your database. View the Exhibit named SETTING and note the result cache settings. Many users in the company state that performance has improved on the queries they use but some users complain that they have not got any performance benefit on the queries they use. You checked all the queries they use and the following is one of them: SQL> SELECT slnoq.currval as "SLNO", prod_id, pdname, 2 cust_name FROM sales WHERE sl_date < sysdate; View the Exhibit named TEST and examine the testing performed to check this. Why is the result cache not used? (Choose all that apply.) Exhibit 1:

Exhibit 2:

Because the query uses SYSDATE

Because the query uses an alias for a column

Because the query uses the SLNOQ.CURRVAL sequence

Because the CLIENT_RESULT_CACHE_SIZE parameter is set to 0

55. You are working on an online transaction processing (OLTP) system. The CUST table in the SH schema was populated by direct load and after that it has gone through a lot of updates and deletions. The statistics for the CUST and SALES tables were updated recently. View the Exhibit and examine the query plan. The query is performing a lot of I/O for a query that fetches only 168 rows. To investigate further, you queried the ALL_TABLES view to find out PCTUSED, PCTFREE, and the number of rows in the CUST table, as given below: SQL> SELECT table_name ,blocks, pct_used, pct_free, num_rows 2 FROM all_tables 3 WHERE table_name = 'CUST'; TABLE_NAME BLOCKS PCT_USED PCT_FREE NUM_ROWS ---------- ---------- ---------- ---------- ---------- CUST 13768 80 1 11060 What would you recommend to reduce the I/O? Exhibit:

Reorganizing the table to use fewer blocks

Rebuilding the index on the CUST_FIRST_NAME column

Increasing the value for the PCTFREE attribute for the CUST table

Increasing the value for the PGA_AGGREGATE_TARGET initialization parameter

56. You plan to collect statistics for an important table in your database to help improved query performance. You are uncertain about the optimizer behavior after the statistics have been collected because currently there are no statistics for indexes or partitions on the table. You wish to test the performance effect of the newly collected statistics. How would you prevent the optimizer from using the newly collected statistics for other sessions immediately after statistics collection?

Set the PUBLISH preference to FALSE for the table.

Set the NO_VALIDATE preference to TRUE for the table.

Set the OPTIMIZER_DYNAMIC_SAMPLING parameter to 0.

Set the OPTIMIZER_USE_SQL_PLAN_BASELINES parameter to FALSE for the session.

Set the OPTIMIZER_USE_PENDING_STATISTICS parameter to FALSE for the session.

57. You work for a small manufacturing company as a DBA. The company has various applications to manage the details of its business and customer base. The company has only one sales outlet where the operator updates the SALES table in the sales application with the details when a product is sold. The company has many other applications for various other businesses that use the same database. The sales operator, while generating a report, noticed that there are some gaps generated in the serial number in the SLNO column during peak hours of business. On investigating, you find that the SLNO column in the SALES table uses the SLNSEQ sequence to generate the serial numbers when a record is inserted. During peak hours, you noticed the following: When the operator entered the sales details the last time, the sequence number was 1056300, but when you checked the sequence as follows, you found a gap of 100 numbers: SQL> SELECT sequence_name, last_number FROM user_sequences WHERE sequence_name='SLNSEQ'; SEQUENCE_NAME LAST_NUMBER ------------------------------- SLNSEQ 1056400 You observed this behavior many times and only during peak hours. What could be the problem and solution for this?

The reserved pool is not configured. Configure the reserved pool.

The keep buffer pool is not configured. Configure the keep buffer pool.

The sequence was created with a CACHE of 100 numbers. Re-create the sequence with a higher CACHE number.

The sequence is aging out of the shared pool. Keep the sequence in the shared pool using the DBMS_SHARED_POOL.KEEP procedure.

58. You work as a DBA for a company and as a performance improvement measure, you implemented the result cache in your database. Many users in the company say that performance has improved on the queries they use, but some users complain that they have not got any performance benefit on the queries they use. You checked all the queries they use and the following is one of them: SQL> SELECT /*+ RESULT_CACHE */ slnoq.currval as "SLNO", prod_id, pdname, cust_name FROM sales WHERE sl_date < sysdate; View the Exhibit and examine the testing performed to check this. Why is the result cache not used? (Choose all that apply.) Exhibit:

Because the query uses SYSDATE

Because the query uses an alias for a column

Because the query uses the SLNOQ.CURRVAL sequence

Because the table might have an index on the SL_DATE column

59. You are working on an online transaction processing (OLTP) system. You detected free buffer waits events for your database instance running in a machine that has multiple CPUs. You increased the database buffer cache size as the first step. After a few hours of work on the database, further investigation shows that the same event is being recorded. What would be your next step to avoid this event in future?

Decrease the value of the DBWR_IO_SLAVES parameter.

Set the USE_INDIRECT_DATA_BUFFERS parameter to TRUE.

Increase the value of the DB_WRITER_PROCESSES parameter.

Increase the value of the DB_FILE_MULTIBLOCK_READ_COUNT parameter

60. In which case is the database buffer cache NOT used?

When physical I/Os are performed to read from data files

When the database writer process writes to the SYSTEM tablespace

When the server process writes sort data to the temporary tablespace

When the database writer process writes undo data to the undo tablespace

61. You work as a DBA and you have the responsibility of managing a large online transaction processing (OLTP) system. You used the following query during the performance tuning activity: SQL> SELECT (1-((phy.value-phyd.value) / (cur.value + con.value))) * 100 2 "Cache Hit ratio" 3 FROM v$sysstat cur, v$sysstat con, v$sysstat phy, v$sysstat phyd 4 WHERE cur.name = 'db block gets' 5 AND con.name = 'consistent gets' 6 AND phy.name = 'physical reads' 7 AND phyd.name = 'physical reads direct'; Cache Hit Ratio --------------- 98.43 Which understanding of this ratio is correct?

98.43% of times the requests have found free buffers.

98.43% of the total number of requests used the buffer cache.

98.43% of times the requests were satisfied by performing physical I/Os.

98.43% of times the requests have found the required data blocks in the buffer cache.

62. You are working on an online transaction processing (OLTP) system. The middle-tier applications use connection pooling to connect to the database. Presently, you have a single-node database. The company plans to migrate the database to a RAC environment. Before you move to a RAC environment, you want to test the performance of the SQL statements and the peak workload on the new environment. To accomplish the Database Capture for replay, you identified the peak workload period on the existing system and started the Database Capture. Which client requests to the database can be captured as part of the workload capture? (Choose all that apply.)

Flashback query

Distributed transactions

Logging in and logging out of sessions

All DDL statements having bind variables

Direct path load of data from external files

Submit

63. View the Exhibit and examine the output for the WRC command-line utility that you executed to estimate the number of replay clients. Identify three prerequisites for completing the replay. (Choose three.) Exhibit

The replay database should be in the RESTRICTED mode.

You need to start at least 21 replay clients on six CPUs to start the replay.

The replay directory must have all the replay files that have been preprocessed.

The user should be able to use the workload replay package and should have the DBA role.

The replay options for think_time_scale and connect_time_scale must be set to 100 each.

64. You work on an online transaction processing (OLTP) database in which the SALES table has 10,000 rows but only four distinct products are sold. View the Exhibit named HIST to check data distribution in the table and the histograms on the table. View the Exhibit named QUERY-1 that shows details in the V$SQL view for the query executed on the SALES table having product id 1. Further, you query the SALES table thrice more in the following order: - query rows having product ID 4 - query rows having product ID 2 - query rows having product ID 4 View the Exhibit named QUERY-n to see the details about these queries. What do you infer from this? Exhibit 1:

Exhibit 2:

Exhibit 3:

The second plan was created because the first plan was aged out.

The CURSOR_SHARING parameter value was changed to EXACT after the second query was executed.

The third plan was created because the first plan was aged out and the second plan had different selectivity.

65. Which two statements are true about the database buffer cache? (Choose two.)

A buffer can be written to by many processes at the same time.

A buffer containing a block that is identical to the block on disk is called free buffer.

A buffer containing a block that is being accessed by a process is called dirty buffer.

A buffer available to be overwritten with a different database block at any time is called a dirty buffer.