多选题A common approach to solve application performance issues is to add more hardware. Which two  reasons describe why this alone does not always solve performance problems?（）AThere may be a bottleneck in the application that additional hardware would not 

题目
多选题
A common approach to solve application performance issues is to add more hardware. Which two  reasons describe why this alone does not always solve performance problems?（）
A

There may be a bottleneck in the application that additional hardware would not solve.

B

Adding more hardware is costly and time consuming.

C

Adding hardware at the application layer may place additional load on an already overloaded database, network appliance, or storage system.

D

Adding more hardware may introduce new configuration requirements that need to be tested.

相似考题

1.Due to slow response times， a customer purchased additional memory. After the memory was installed，response time was not improved.  They complained to the pSeries technical specialist. Which of the following questions should be asked of this customer？（）A、Who recommended the memory purchase？B、Have more users been added since the memory purchase？C、Why were additional CPUs not ordered at the same time as the memory？D、Did a performance analysis of peak periods show a memory bottleneck？

2. A customer wants to improve the availability of a web application and provide more predictable scalability when scaling out the application.    Which Feature of WebLogic should you recommend to help solve this problem?（）A、Oracle Web GridB、ActiveCacheC、Coherence Grid EditionD、WebLogic Session ReplicationE、Coherence Web Edition

3.Identify three reasons that explain why assigning more than one managed serve achieve better  performance than running one managed server with more resources?（）A、Thread management inside the JVM process might be less effective than context switching between processes on the OS level. It is better to combine both these technologies to achieve full saturation of resources.B、Current CPUs are multicore units. So starting managed servers in the amount of physical cores and pinning them 1:1 to cores will always get the best performance.C、By a combination of prioritization of processes on the OS level and Work Manager in WLS, one  could tune the whole environment more precisely.D、Assigning more than one managed server to hardware will never bring better performance. E、More instances means more connections to databases, so one could serve more concurrent calls on the database level. This is primarily why this approach will give better performance numbers.  F、Depending on the profile of the application, having more JVMs with smaller heaps will lead to more predictable response times than fewer JVMs with a large heap.

4.When your WebLogic Server solution needs to be scaled out with additional capacity and you  Don’t want to add additional hardware, which three techniques should you us?（）A、Assign more than one managed server to a physical hardware that allows better CPU  utilizationB、Assign more than one application to one managed server to better utilize threads within a single JVM processC、Assign the same application to more than one managed server to load balance requests  between serversD、Assign the same heap size to the managed server across the cluster for easier control of  memory footprint  E、Create a virtualized environment with hypervisor for an easier solution

更多“多选题A common approach to solve application performance issues is to add more hardware. Which two  reasons describe why this alone does not always solve performance problems?（）AThere may be a bottleneck in the application that additional hardware would not ”相关问题

  • 第1题：

    While deploying a new application module, the software vendor ships the application software along with appropriate SQL plan baselines for the new SQLs being introduced. Which two statements describe the consequences?（）

    • A、The plan baselines can be evolved over time to produce better performance.
    • B、The newly generated plans are directly placed into the SQL plan baseline without being verified.
    • C、The new SQL statements initially run with the plans that are known to produce good performance under standard test configuration.
    • D、The optimizer does not generate new plans for the SQL statements for which the SQL plan baseline has been imported.

    正确答案:A,C

  • 第2题：

    A customer wants to increase the performance of an existing server. They are running a very network intensive application and feel that increasing the bandwidth to the network will fix their problem.  They are running two 24-port autosensing 10/100 hubs， tied together with a standard Ethernet crossover cable in port 24.  The server is running IPX. After adding a second network card and plugging one card into each hub， they have not seen any performance improvement.  Which two are likely reasons why implementing network adapter teaming has not improved performance？（）

    • A、 Adapter teaming is designed to work with switches.
    • B、 Adapter teaming only functions with gigabit network cards.
    • C、 Adapter teaming only provides fault tolerance， not additional bandwidth.
    • D、 Adapter teaming only works with IPX when using switches and network cards with special features that support teaming.

    正确答案:A,D

  • 第3题：

    Which IBM Resource would best help you determine the proper performance tuning of your System x hardware and the application workload of the software？（）

    • A、xREF Sheets
    • B、IBM Sizing Guides
    • C、Configuration & Options Guide - （COG）
    • D、Standalone Solutions Configuration Tool （SSCT）

    正确答案:B

  • 第4题：

    A common approach to solve application performance issues is to add more hardware. Which two  reasons describe why this alone does not always solve performance problems?（）

    • A、There may be a bottleneck in the application that additional hardware would not solve.
    • B、Adding more hardware is costly and time consuming.
    • C、Adding hardware at the application layer may place additional load on an already overloaded database, network appliance, or storage system.
    • D、Adding more hardware may introduce new configuration requirements that need to be tested.

    正确答案:A,C

  • 第5题：

    Select valid reasons for IT performance issues that cannot be universally solved by adding more hardware：（）

    • A、 adding hardware may exacerbate inefficiency， which can lead to over provisioning and lower utilization
    • B、 performance may degrade due to bottlenecks such as database connections that won’t scale in lock step with other dimensions such as computing capacity or memory
    • C、 both A and B are valid
    • D、 neither A nor B are valid

    正确答案:C

  • 第6题：

    多选题
    A common approach to solve application performance issues is to add more hardware. Which two  reasons describe why this alone does not always solve performance problems?（）
    A

    There may be a bottleneck in the application that additional hardware would not solve.

    B

    Adding more hardware is costly and time consuming.

    C

    Adding hardware at the application layer may place additional load on an already overloaded database, network appliance, or storage system.

    D

    Adding more hardware may introduce new configuration requirements that need to be tested.


    正确答案： D,B
    解析： 暂无解析

  • 第7题：

    多选题
    You are the network administrator for The network consists of a single Active Directory domain named All network servers run Windows Server 2003. Your network includes domain controllers, file and print servers, and application servers. The application servers run a variety of programs, including Microsoft SQL Server 2000 and Microsoft Exchange Server 2003. Your staff are responsible for monitoring current system performance on all servers. You need to enable your staff to use System Monitor to gather performance data for each unique server type. The data will be used for trend analysis and forecasting. Which two actions should you perform? （）(Each correct answer presents part of the solution. Choose two)
    A

    For each server, add the most common performance counters and save them as an HTML file.

    B

    For each server, add the most common performance counters and save them as a counter report file.

    C

    Create trace logs based on the file and schedule and trace logs to gather data.

    D

    Create alerts on the file and schedule the alerts to gather data.

    E

    Create counter logs based on the file and schedule the counter logs to gather data.


    正确答案： D,B
    解析： 暂无解析

  • 第8题：

    多选题
    Identify three reasons that explain why assigning more than one managed serve achieve better  performance than running one managed server with more resources?（）
    A

    Thread management inside the JVM process might be less effective than context switching between processes on the OS level. It is better to combine both these technologies to achieve full saturation of resources.

    B

    Current CPUs are multicore units. So starting managed servers in the amount of physical cores and pinning them 1:1 to cores will always get the best performance.

    C

    By a combination of prioritization of processes on the OS level and Work Manager in WLS, one  could tune the whole environment more precisely.

    D

    Assigning more than one managed server to hardware will never bring better performance.

    E

    More instances means more connections to databases, so one could serve more concurrent  calls on the database level. This is primarily why this approach will give better performance numbers.

    F

    Depending on the profile of the application, having more JVMs with smaller heaps will lead to more predictable response times than fewer JVMs with a large heap.


    正确答案： F,A
    解析： 暂无解析

  • 第9题：

    单选题
    A customer with a single IBM System Storage DS8100 reports severe performance problems for some of their applications when a particular data warehousing application creates a heavy load.  What is the most cost effective way to resolve this issue（）
    A

    add additional disks to the DS8100

    B

    add more cache to the DS8100

    C

    move affected volumes to a different disk enclosure

    D

    implement TPC for Disk to monitor performance


    正确答案： D
    解析： 暂无解析

  • 第10题：

    单选题
    An application server is running slowly. After viewing performance monitoring data it is detected that the disk queue length is high.  Which of the following could BEST resolve the situation？（）
    A

     Add more disk drives.

    B

     Increase the amount of virtual memory.

    C

     Add a faster processor.

    D

     Add more memory to stop paging to disk.


    正确答案： D
    解析： 暂无解析

  • 第11题：

    多选题
    While deploying a new application module, the software vendor ships the application software along with appropriate SQL plan baselines for the new SQLs being introduced. Which two statements describe the consequences?（）
    A

    The plan baselines can be evolved over time to produce better performance.

    B

    The newly generated plans are directly placed into the SQL plan baseline without being verified.

    C

    The new SQL statements initially run with the plans that are known to produce good performance under standard test configuration.

    D

    The optimizer does not generate new plans for the SQL statements for which the SQL plan baseline has been imported.


    正确答案： C,A
    解析： 暂无解析

  • 第12题：

    多选题
    A Windows Azure application connects to a SQL Azure database.  Connections to the database are being dropped due to throttling.   You need to recommend an approach for identifying the root cause of the problem.   Which two actions should you recommend？（）
    A

    Use Dynamic Management Views to analyze performance issues.

    B

    Decrease the connection timeout in the database connection string.

    C

    Make sure the application is closing database connections that are not needed.

    D

    Use SQL Profiler to analyze performance issues by configuring it to connect to SQL Azure.


    正确答案： C,A
    解析： 暂无解析

  • 第13题：

    An application server is running slowly. After viewing performance monitoring data it is detected that the disk queue length is high.  Which of the following could BEST resolve the situation？（）

    • A、 Add more disk drives.
    • B、 Increase the amount of virtual memory.
    • C、 Add a faster processor.
    • D、 Add more memory to stop paging to disk.

    正确答案:D

  • 第14题：

    A customer has an x445 and uses the internal LSI adapter to manage two 146GB drives in the chassis.  The customer is noticing performance issues when deploying a database with significant write activity on the internal drives.  Should they consider using a ServeRAID-6M with 256MB of cache instead， and why or why not？（）

    • A、Yes， the internal LSI SCSI adapter has no write cache.
    • B、Yes， the internal LSI SCSI adapter is unreliable.
    • C、No， the downtime to install the card and reformat the drives would be cost prohibitive.
    • D、No， there will be no performance benefit， because drive performance is the bottleneck.

    正确答案:A

  • 第15题：

    A customer with a single IBM System Storage DS8100 reports severe performance problems for some of their applications when a particular data warehousing application creates a heavy load.  What is the most cost effective way to resolve this issue（）

    • A、add additional disks to the DS8100
    • B、add more cache to the DS8100
    • C、move affected volumes to a different disk enclosure
    • D、implement TPC for Disk to monitor performance

    正确答案:C

  • 第16题：

    Which three JVM options would you add to investigate performance issues in a Hot-Spot JVM?（）

    • A、-xx:+PrintGCDetails
    • B、-xx:+PrintGCTimeStamps  
    • C、-xx:+HandlePromotionFailure
    • D、-verbose:gc
    • E、-server

    正确答案:A,B,E

  • 第17题：

    A Windows Azure application connects to a SQL Azure database.  Connections to the database are being dropped due to throttling.   You need to recommend an approach for identifying the root cause of the problem.   Which two actions should you recommend？（）

    • A、 Use Dynamic Management Views to analyze performance issues.
    • B、 Decrease the connection timeout in the database connection string.
    • C、 Make sure the application is closing database connections that are not needed.
    • D、 Use SQL Profiler to analyze performance issues by configuring it to connect to SQL Azure.

    正确答案:A,C

  • 第18题：

    单选题
    A customer wants to improve the availability of a web application and provide more predictable scalability when scaling out the application.    Which Feature of WebLogic should you recommend to help solve this problem?（）
    A

    Oracle Web Grid

    B

    ActiveCache

    C

    Coherence Grid Edition

    D

    WebLogic Session Replication

    E

    Coherence Web Edition


    正确答案： B
    解析： 暂无解析

  • 第19题：

    单选题
    Select valid reasons for IT performance issues that cannot be universally solved by adding more hardware：（）
    A

     adding hardware may exacerbate inefficiency， which can lead to over provisioning and lower utilization

    B

     performance may degrade due to bottlenecks such as database connections that won’t scale in lock step with other dimensions such as computing capacity or memory

    C

     both A and B are valid

    D

     neither A nor B are valid


    正确答案： C
    解析： 暂无解析

  • 第20题：

    多选题
    When your WebLogic Server solution needs to be scaled out with additional capacity and you  Don’t want to add additional hardware, which three techniques should you us?（）
    A

    Assign more than one managed server to a physical hardware that allows better CPU  utilization

    B

    Assign more than one application to one managed server to better utilize threads within a single JVM process

    C

    Assign the same application to more than one managed server to load balance requests  between servers

    D

    Assign the same heap size to the managed server across the cluster for easier control of  memory footprint

    E

    Create a virtualized environment with hypervisor for an easier solution


    正确答案： D,C
    解析： 暂无解析

  • 第21题：

    多选题
    A customer wants to increase the performance of an existing server. They are running a very network intensive application and feel that increasing the bandwidth to the network will fix their problem.  They are running two 24-port autosensing 10/100 hubs， tied together with a standard Ethernet crossover cable in port 24.  The server is running IPX. After adding a second network card and plugging one card into each hub， they have not seen any performance improvement.  Which two are likely reasons why implementing network adapter teaming has not improved performance？（）
    A

    Adapter teaming is designed to work with switches.

    B

    Adapter teaming only functions with gigabit network cards.

    C

    Adapter teaming only provides fault tolerance， not additional bandwidth.

    D

    Adapter teaming only works with IPX when using switches and network cards with special features that support teaming.


    正确答案： B,D
    解析： 暂无解析

  • 第22题：

    单选题
    A customer has an x445 and uses the internal LSI adapter to manage two 146GB drives in the chassis.  The customer is noticing performance issues when deploying a database with significant write activity on the internal drives.  Should they consider using a ServeRAID-6M with 256MB of cache instead， and why or why not？（）
    A

    Yes， the internal LSI SCSI adapter has no write cache.

    B

    Yes， the internal LSI SCSI adapter is unreliable.

    C

    No， the downtime to install the card and reformat the drives would be cost prohibitive.

    D

    No， there will be no performance benefit， because drive performance is the bottleneck.


    正确答案： A
    解析： 暂无解析

  • 第23题：

    多选题
    Which three JVM options would you add to investigate performance issues in a Hot-Spot JVM?（）
    A

    -xx:+PrintGCDetails

    B

    -xx:+PrintGCTimeStamps

    C

    -xx:+HandlePromotionFailure

    D

    -verbose:gc

    E

    -server


    正确答案： B,D
    解析： 暂无解析

  • 第24题：

    多选题
    You are asked to configure real-time performance monitoring (RPM) between two Junos devices to track traffic in your network. Which two application traffic types can you track?（）
    A

    ICMP

    B

    SNMP

    C

    TCP

    D

    HTTPS


    正确答案： A,D
    解析： 暂无解析

相关内容