| Question 1 | In the Book Seller, Inc. case study, which of the following methods might you use to deploy the storefront application? A. Setup Project B. Web Setup Project C. Merge Module Project D. Cab Project
|
| A1: | The correct answer is B. The standard Setup Project does not automatically deploy IIS settings. A Merge Module Project is used for reusable deployment components. A Cab Project is best suited for ActiveX deployment, not Web applications. |
| Question 2 | Which of the following Book Seller, Inc. elements could you consider placing in merge modules? (Choose all that apply.) |
| A2: | The correct answers are A and E. Business components and data access logic can be built into redistributable components ideal for placement into Merge Module Projects. The user interface should be placed in a Web Setup Project. ActiveX controls are placed in Cab Projects. The inventory management application is not being replaced. |
| Question 3 | When building your deployment project, you should configure it to take what type of action if deployment fails? A. Validate installed components B. Attempt to reinstall C. Enable Admin privileges D. Delete Registry settings E. Not applicable
|
| A3: | The correct answer is E. Windows Installer automatically takes care of uninstalling the partially installed application should it fail, so the other answers do not apply. |
| Question 4 | What are the key benefits Windows Forms No-Touch Deployment offers over traditional Win32 forms deployment? (Choose all that apply.) |
| A4: | The correct answers are A and C. No-Touch Deployment offers offline capabilities, responsiveness, and a rich graphical interface, but these features are a benefit over Web applications, not other Windows Forms applications. |
| Question 5 | If Book Seller, Inc. used licensed components, where would the license be validated? |
| A5: | The correct answer is C. The License Provider is used to build custom license validation. The Common Language Runtime does validate the licensed components in a sense, but is not the best choice because it does not directly validate the license, as License Manager does. The CTS does not pertain to licensing. |
| Question 6 | Which of the following technologies would be best suited for data synchronization between Book Seller, Inc.'s storefront and inventory management applications? |
| A6: | The correct answer is B. XML Web Services could be used, but not without an assumption of a service agent in the inventory management application. BCP is not the best choice for synchronizing multiple tables. A SQL job is applied to the DTS package after it has been created if it is to run at a scheduled time, but a job has no synchronization ability alone. |
| Question 7 | If Book Seller, Inc. wanted to share multiple versions of the same assembly across a single server, which of the following technologies would be the best choice? A. Registry B. GUID C. COM D. Global Assembly Cache
|
| A7: | The correct answer is D. The Registry is not used when storing .NET assembly information. Each .NET assembly has a GUID, but it does not change from version to version. COM is not used to build .NET assemblies. |
| Question 8 | How does the GAC differentiate various versions of one assembly? A. Registry information for the assemblies B. GUID of the assemblies C. Specific references to the assemblies D. Strong name of the assemblies
|
| A8: | The correct answer is C. .NET assembly information is not stored in the Registry. The GUID and the strong name for each assembly do not change per version. |
| Question 9 | Book Seller, Inc. has experienced serious performance issues. Which of the following items could you monitor to capture performance loss information? (Choose all that apply.) |
| A9: | The correct answers are B and D. The SNMP provider, Registry Event Provider, and WDM Provider do not capture performance information. |
| Question 10 | Billington performs the same query about 80% of the time to the Patients table. The Patients table currently has a clustered index on the auto-incrementing primary key. The query returns a list of patients by state. Which of the following indexing choices would be best suited for this query? A. Clustered index on the primary key B. Non-clustered index on the primary key C. Clustered index on the State column D. Non-clustered index on the State column
|
| A10: | The correct answer is D. A non-clustered index on the State column is the best choice because the Patients table can contain only one clustered index; therefore, a non-clustered index is your only option. A clustered index on the primary key is not the best choice for this question, but could slightly increase performance. A non-clustered index on the primary key would not help in this situation. A clustered index is best suited for non-repetitive data, and the State column should contain mostly repetitive data. |
| Question 11 | What does the Index Tuning Wizard generate from the supplied trace information? A. Index recommendations B. Non-clustered indexes C. Clustered indexes D. All indexes
|
| A11: | The correct answer is A. The Index Tuning Wizard makes recommendations only; it doesn't actually build any indexes. |
| Question 12 | Given Billington's diverse geographic locations, which of the following forms of partitioning could you use to divide and share data? |
| A12: | The correct answer is B. Horizontal, vertical, and disk partitioning do not pertain to geographic locations. |
| Question 13 | If Billington wanted to increase the speed of queries to the Patients table, which of the following forms of partitioning could be used if the queries typically exclude the Comments and Description columns? A. Vertically partition the Comments and Description columns. B. Horizontally partition the Comments and Description columns. C. Use federated databases. D. Use disk partitioning.
|
| A13: | The correct answer is A. Horizontal partitioning relates to rows, not columns. Federated databases and disk partitioning could be used after vertical partitioning was applied to increase performance, but used before vertical partitioning, they actually degrade the speed of queries to the Patients table. |
| Question 14 | If Billington infrequently accessed a few tables, how could you increase performance to other tables in the database? A. Store the tables in delimited text files. B. Partition the infrequently used tables to another database instance. C. Denormalize the infrequently used tables. D. Partition the infrequently used tables to another disk.
|
| A14: | The correct answer is D. Partitioning the infrequently used tables to another physical disk decreases the size of the database, thus slightly increasing the performance of queries placed to other tables. Storing the tables in text files creates too much obscurity and does not represent the best choice. You could partition the tables into another database if disk space was utilized, but this option was not stated. Denormalizing infrequently accessed tables has no impact on other tables in the database. |
| Question 15 | Billington's Patients table has to be joined with six other tables to return the requested patient list, and queries are taking longer than expected. Which of the following actions could you take to increase the performance of queries placed to the Patients table? (Choose all that apply.) A. Use non-clustered indexes on foreign keys. B. Use clustered indexes on foreign keys. C. Use vertical partitioning for the foreign key relationships. D. Denormalize foreign key relationships. E. Use a non-clustered index on the primary key of each table that the foreign keys of the patients table relate to.
|
| A15: | The correct answers are A and D. There can be only one clustered index per database. Vertically partitioning the foreign key relationships would do nothing to increase performance. Using a non-clustered index as the primary key for the reference tables would not help this situation in general, but using a clustered index on the reference table's primary key could increase performance. |
| Question 16 | If Billington's network could transfer continuously for 16.5 hours and the bandwidth of the transfer was 13KB per second, which of the following would be the transfer capacity? A. 214.5KB B. 12,870KB C. 21,450KB D. 12,480KB
|
| A16: | The correct answer is B. Answers A, C, and D are incorrect because you multiply 16.5 hours times 60 seconds and then multiply again by 13 to get the total transfer capacity (12,870) in kilobytes. |
