feedburner
Enter your email address:

Delivered by FeedBurner

feedburner count
Custom Search

Part - 17 MCSE Interview

QUESTION 141
You work as the database administrator at Certkiller .com. The Certkiller .com
network consists of a single Windows 2000 domain named Certkiller .com. All servers
on the Certkiller .com network run Windows 2000 Server and all client computers
run Windows 2000 Professional. The Certkiller .com network contains a SQL Server
2000 database server named Certkiller -DB01. Certkiller -DB01 hosts a database
named CK_Sales that stores sales data for the company. The tables in the CK_Sales
database are shown in the following database diagram.



The recent increase in database usage at the company has resulted in several
Certkiller .com users complaining of timeouts when they try to retrieve sales orders
from the CK_Sales database. You need to determine whether partitioning the
Orders table would improve database performance.
What should you do? (Each correct answer presents part of the solution. Choose
TWO.)

A. Run Index Tuning Wizard.
B. Create a performance log file to monitor SQL Server:Buffer Manager:Page
reads/sec.
C. Run SQL Server Profiler to replay the trace file and the log file.
D. Use the SQL Server Profiler Tuning template to create a trace file.
E. Create a performance log file to monitor Logical Disk: Disk Read Bytes/sec

Answer: A, D

Explanation:

You can use the Index Tuning Wizard to determine whether indexing and
partitioning of a table would improve database performance. The Index Tuning
Wizard a file that you can create by running SQL Profiler.

Incorrect Answers:
B: The SQL Server:Buffer Manager:Page reads/sec counter is used to monitor the read
activity on an instance of SQL Server. It does not help you determine whether
partitioning a table will improve database performance.
C: You can use the SQL Profiler to analyze a trace file and a log file to determine which queries are causing excessive resource utilization. However, this information does not help you determine whether partitioning a table will improve database performance.
E: The Logical Disk: Disk Read Bytes/sec counter is used to monitor the read activity on a logical disk. It does not help you determine whether partitioning a table will improve database performance.


QUESTION 142
You work as the database administrator at Certkiller .com. The Certkiller .com
network consists of a single Active Directory domain named Certkiller .com. All
servers on the Certkiller .com network run Windows 2000 Server and all client
computers run Windows 2000 Professional.
The Certkiller .com network contains a SQL Server 2000 database server named
Certkiller -DB01. Certkiller -DB01 hosts a database named CK_Sales that stores
sales data for the company. The tables in the CK_Sales database are shown in the
following database diagram.



Several Certkiller .com users start complaining of slow response time when they run
queries against the CK_Sales database. You run the sys.dm_os_schedulers view on
Certkiller -DB01 and discover that the runnable_tasks_count is consistently at or
above ten. You run System Monitor on Certkiller -DB01 and receive the output as
shown in the exhibit.



You need to improve database performance.
What should you do?

A. Increase the Random Access Memory (RAM).
B. Add an additional processor.
C. Upgrade the disk subsystem.
D. Use the Index Tuning Wizard to suggest new indexes.

Answer: B

Explanation: The Processor:% Processor Time counter in the exhibit indicates that
the processor is often running at over 80%. This indicates that the processor is
causing a bottleneck. Adding an additional processor to the system will improve
overall system performance.

Incorrect Answers:
A: The Memory:Avg. Disk Queue Length counter in the exhibit is low. This indicates
that the RAM is not causing a bottleneck. Therefore, adding additional RAM to the
system will not improve overall system performance.
C: The PhysicalDisk:Pages/sec counter in the exhibit is low. This indicates that the disk subsystem is not causing a bottleneck. Therefore, upgrading the disk subsystem will not improve overall system performance.
D: The Processor:% Processor Time counter in the exhibit indicates that the processor is often running at over 80%. This indicates that the processor is causing a bottleneck. Indexing the tables will not reduce the load on the processor and will not improve overall system performance.


QUESTION 143
DRAG DROP
You work as the database administrator at Certkiller .com. The Certkiller .com
network consists of a single Active Directory domain named Certkiller .com. All
servers on the Certkiller .com network run Windows Server2003. The Certkiller .com
network contains a SQL Server 2005 on a server named Certkiller -DB01.
Certkiller -DB01 has five 120 GB hard disk drives. Two of the hard disks are
configured as a mirrored volume that contains the operating system. The other
three hard disks are configured as a RAID-5 volume. All database files are located
on the RAID-5 volume while the transaction log files are located on the mirrored
volume.

Certkiller -DB01 is running a default SQL Server instance and a named instance
named CK_DB. The default instance hosts an Online Transaction Processing
(OLTP) database named CK_Sales while the named instance hosts historical data
used for reporting.

Certkiller .com users complain that database performance is deteriorating. You
suspect that a disk problem might be causing the bottleneck. You want to determine
which SQL Server instance is using the most disk time.
What should you do? (To answer, select the appropriate performance counters that
you should monitor in the left pane and drag them to the right pane.)



Answers :




Explanation:
The SQL Server:Buffer Manager object is used to monitor resource usage for each SQL
Server instance. You need to monitor the SQL Server:Buffer Manager:Page reads/sec
counter to determine the number of page reads performed by each instance and the SQL
Server:Buffer Manager:Page writes/sec counter to determine the number of page writes
performed by each instance.

Incorrect Answers:

The Logical Disk:Disk Write Bytes/sec and Logical Disk:Disk Read Bytes/sec counters
can be used to monitor the total number of disk reads or writes per second for a volume.

Neither can be used to determine instance is using the most disk time.
The PhysicalDisk counters can be used to monitor a RAID volume but this cannot used
to determine instance is using the most disk time.


QUESTION 144
You work as the database administrator at Certkiller .com. The Certkiller .com
network consists of a single Active Directory domain named Certkiller .com. All
servers on the Certkiller .com network run Windows Server 2003 and all client
computers run Windows XP Professional. The Certkiller .com network contains a
SQL Server 2005 database server named Certkiller -DB01 that hosts a database
named CK_Sales. The CK_Sales database stores sales data for the company.
Certkiller .com users complain that the performance of the CK_Sales database has
deteriorated over the last few weeks. You use System Monitor to monitor the
performance of Certkiller -DB01 and receive the output as shown in the exhibit.




You also notice that the ratio of SQL Recompilations/sec to Batch Requests/sec is
unusually high. You suspect that the number of recompiles is causing the high
processor time value. You need to improve the performance of the CK_Sales
database.
What should you do?

A. Turn off automatic updates of statistics for all tables in the CK_Sales database.
B. Install an additional processor on Certkiller -DB01.
C. Run the Database Engine Tuning Advisor.
D. Use SQL Server Profiler to identify the stored procedures being recompiled.

Answer: D

Explanation: You can use SQL Server Profiler to identify the stored procedures
that are being recompiled. SQL Profiler will indicate which stored procedures are
being recompiled and why each recompilation is occurring.

Incorrect Answers:

A: Turning off automatic updates of statistics for the tables may improve performance
but it does not address the problem caused by the recompiles.
B: The high processor usage could be caused by the high number of recompiles. You
should first reduce the number of recompiles before installing extra hardware on the
server.
C: You can use the Database Engine Tuning Advisor to determine whether indexing and
partitioning of a table would improve database performance but you cannot use it to
identify which stored procedures are being recompiled.


QUESTION 145
You work as the network database administrator at Certkiller .com. The
Certkiller .com network consists of a single Active Directory domain named
Certkiller .com. All servers on the Certkiller .com network run Windows 2000 Server
and all client computers run Windows XP Professional.
Certkiller .com management has recently decided to install SQL Server 2000 on a
newly acquired Windows NT Server 4.0 computer. As a result of the new acquiring
you decide to create a trace in SQL Profiler to capture several events which include
user events, server events and security events. You additionally specified that the
results of the trace be saved in a file and you start the trace.
During the course of the business week the SQL server stops. You are required to
have the SQL Server resume normal operations whilst continuing to audit.
What should you do?

A. SQL Server should be restarted in a single-user mode.
B. The MSSQLServer and the SQLServerAgent services should be restarted.
C. A new disk should be added to the SQL Server computer.
D. The audit logs should be archived and remove them from the disk and restart
MSSQLServer.

Answer: D

Explanation: In the scenario you should remember that the log file used for auditing
would most likely continue to accept new event records as soon as free space
becomes available on the disk drive an then the MSSQLServer service is restarted.

Incorrect Answers:
A: This option could be used for troubleshooting in the scenario but then no other users would be allowed to connect to the instance of SQL Server.
B: You should not consider using this option in the scenario because you are not
required to restart the SQLServerAgent service in the scenario.
C: This option could be used in the scenario but you would still be required to
restart the MSSQLServer service in the scenario.


QUESTION 146
You work as the network database administrator at Certkiller .com. The
Certkiller .com network consists of a single Active Directory domain named
Certkiller .com. All servers on the Certkiller .com network run Windows 2000 Server
and all client computers run Windows XP Professional.
You have received instruction from the management of Certkiller .com and
successfully created an application which network users will use to manipulate data
on two instances of SQL Server which reside on a Windows 2000 Server computer.
Certkiller .com network users use the application to access databases on both
instances of SQL Server. You are required to use SL Profiler to monitor both
instances of SQL Server 2000. You are required to perform this task using the least
amount of administrative effort.
What should you do?

A. One trace should be created for each instance of SQL Server. You should then
configure each trace to monitor both applications and run the traces in separate instances of SQL Profiler.
B. One trace should be created for each instance of SQL Server. You should then
configure each trace to monitor the first applications and run both traces in a single instance of SQL Profiler.
C. One trace should be created to monitor both instances of SQL Server. You should then configure each trace to monitor both applications and run the traces in a single instance of SQL Profiler.
D. One trace should be created for each instance of SQL Server. You should then
configure each trace to monitor both applications and run the traces in a single instance of SQL Profiler.

Answer: D

Explanation: In the scenario you would do well to remember that SQL Profiler is an
administrative tool that can be used to present information about the SQL Server
events.

Incorrect Answers:
A: In the scenario you should not use this option because a single trace can connect
to only one instance of SQL Server but you can run both traces together in the same
instance of SQL Profiler. B, C: In the scenario you should not consider using these options because each instance of SQL Server still requires running a separate trace.


QUESTION 147
You work as the database administrator at Certkiller .com. The Certkiller .com
network consists of a single Active Directory domain named Certkiller .com. All
servers on the Certkiller .com network run Windows 2000 Server and all client
computers run Windows 2000 Professional. The Certkiller .com network contains a
SQL Server 2000 database server named Certkiller -DB01 that hosts a database
named CK_Sales. The CK_Sales database stores sales data for the company.
Certkiller .com users complain that the performance of the CK_Sales database has
deteriorated over the last few weeks. You use System Monitor to monitor the
performance of Certkiller -DB01 and receive the output as shown in the exhibit.




You also notice that the ratio of SQL Recompilations/sec to Batch Requests/sec is
unusually high. You suspect that the number of recompiles is causing the high
processor time value. You need to improve the performance of the CK_Sales
database.
What should you do?

A. Turn off automatic updates of statistics for all tables in the CK_Sales database.
B. Install an additional processor on Certkiller -DB01.
C. Run the Index Tuning Wizard.
D. Use SQL Profiler to identify the stored procedures being recompiled.

Answer: D

Explanation: You can use SQL Profiler to identify the stored procedures that are
being recompiled. SQL Profiler will indicate which stored procedures are being
recompiled and why each recompilation is occurring.

Incorrect Answers:
A: Turning off automatic updates of statistics for the tables may improve performance
but it does not address the problem caused by the recompiles.
B: The high processor usage could be caused by the high number of recompiles. You
should first reduce the number of recompiles before installing extra hardware on the
server.
C: You can use the Index Tuning Wizard to determine whether indexing and partitioning
of a table would improve database performance but you cannot use it to identify which
stored procedures are being recompiled.


QUESTION 148
You work as the network database administrator at Certkiller .com. The
Certkiller .com network consists of a single Active Directory domain named
Certkiller .com. All servers on the Certkiller .com network run Windows 2000 Server
and all client computers run Windows XP Professional. You are responsible for
administering the SQL Server 2000 database of Certkiller .com which is used
extensively to store business information.
Certkiller .com currently has 700 network users who connect to the database to
perform various activities. Certain network employees of Certkiller .com makes use
of an custom application that is used for retrieving data for analysis and performing
INSERT and UPDATE operations. The network users of Certkiller .com ahs recently
started reporting that the custom application responds slowly.
During your maintenance routine you decide to open System Monitor and analyze
the counters below and discover the following values:

1. SQLServer: Access Methods Page Splits/sec has a value of 2
2. SQLServer: BufferManager Buffer cache hit ratio has a value of 95
3. SQLStatistics SQL Re-Compilations/sec has a value of 15
4. SQLServer: Locks Requests/sec has a value of 8,000
You are required to determine that exact cause of the applications poor
performance.
What should you do?

A. The WITH RECOMPILE option should be added to the definition of the stored
procedure that the application uses.
B. You should add additional RAM.
C. On all the tables referenced by the application you should rebuild the clustered
indexes and specify a low value for the fill factor.
D. In Enterprise Manager you should open the Current Activity window to determine
which process are holding locks for resources required by the application.

Answer: D

Explanation: You should remember in the scenario that the SQLServer: Locks
Requests/sec has a value of 8,000 which signifies a high level of activity. In the
scenario the applications performance is slow as it is forced to wait for resources to be released.

Incorrect Answers:

A: This option should not be used in the scenario because there is no indication that the stored procedures should be recompile each time when called. This only degrades the performance further.
B: In the scenario you should not consider adding additional RAM as the
performance object SQLServer: Locks Requests/sec has a value of 8,000 indicating a
high level of activity.
C: This option should not be used in the scenario because there is no indication of excessive page splits so there is no need to rebuild all the clustered indexes.


QUESTION 149
You work as the database administrator at Certkiller .com. The Certkiller .com
network consists of a single Windows 2000 domain named Certkiller .com. All servers
on the Certkiller .com network run Windows 2000 Server and all client computers
run Windows 2000 Professional.
The Certkiller .com network contains a Microsoft SQL Server 2000 database server
which you administer. You discovered recently that the database server is experiencing performance problems. You suspect that the bottleneck is disk access
you run the performance utility and the log indicates the values below:
1. PhysicalDisk: %Disk Time = 94%
2. PhysicalDisk: Avg. Disk Queue Length = 3
You are required to eliminate the lack of physical memory as the bottleneck.
What should you do?

A. Monitor the Memory: Pages/sec counter.
B. Monitor the SQL Server: Buffer Manager: Page reads/sec counter.
C. Monitor the Memory: Cache faults/sec counter.
D. Monitor the SQL Server: Buffer Manager: Total Pages counter.

Answer: A

Explanation: In the scenario you should remember in order to monitor the lack of
physical memory that you should use the Memory: Pages/sew counter as this will
determine how many hard page faults occur.

Incorrect Answers:
B: This option should not be used in the scenario as it will only tell you the amount of the disk access performed by the instance of SQL server.
C: This option is used to report only the number of hard faults and soft faults in a
combination and should not be used in the scenario.
D: This option should not be used in the scenario because it only displays the number of pages in the buffer pool.


QUESTION 150
You work as a data administrator at Certkiller .com. The Certkiller .com network
consists of a single Active Directory domain named Certkiller .com. All servers on
Certkiller .com run Microsoft Windows 2000 Server. The database server named
Certkiller -DB01 is configured with Microsoft SQL Server 2000. Your job
description includes the maintenance of Certkiller -DB01.
The Certkiller -DB01 specifications are as follows:
1. One processor
2. 1 gigabyte (GB) of memory
3. One physical hard drive
You are currently busy establishing the performance threshold for
Certkiller -DB01. To this end you are monitoring Certkiller -DB01 for common
bottleneck conditions. You now need to identify the potential bottleneck conditions.
You thus need to make a choice in the appropriate counters and the appropriate
settings.
What should you do?

A. The PhysicalDisk: %Disk Time counter should be monitored and set the threshold not
to fall below 90 percent.
B. The Processor: %Privileged Time counter should be monitored and set the threshold
not to fall below 20 percent.
C. The PhysicalDisk: Avg. Disk Queue Length counter should be monitored and set the
threshold not to exceed 1.
D. The Processor: %Processor Time counter should be monitored and set the threshold to not exceed 80 percent.

Answer: D

Explanation: In the scenario you should only consider using the option in the
answer because the counter monitors the amount of time that the CPU spends
executing non-idle threads.

Incorrect Answers:

A: In the scenario you should always remember that when looking for bottlenecks that a lower value is better for usage.
B: This option should not be used in the scenario because this should only be used if you identified that the CPU is the bottleneck.
C: You should not make this configuration in the scenario because a bottleneck can be
indicated by a value higher than 1.5 if you have only one hard disk.





0 comments:

Post a Comment

Enter Valid e-mail to get all updates of this sites in mail

Enter your email address:

Delivered by FeedBurner

Cheapest predictive dialer for callcenter power dial

Power Dial predictive dialer for callcenter with less investment and more features based in the heart of Hyderabad AP India.Power Dial has setup more then 200 centers and more then 5000 seats supports is provided.
For best quotes, pricing and other details mail me asap at powerdial.hyd@gmail.com

Vivaan Kumar
http://powerdial.blogspot.com/
Cheapest Predictive Dialer power Dial
cheap pd power dial
predictive dialer
predictive dialer for callcenter with less investment power dial
predictive dialer for callcenter with low price
predictive dialer form india
predictive dialer from hyderabad
power dial predictive dialer