2. 计算机
  3. You are the administrator of a SQL Server 2000 computer. Your company purchased an accounting application from a vendor. The app

You are the administrator of a SQL Server 2000 computer. Your company purchased an accounting application from a vendor. The app

admin2009-05-19  61
问题 You are the administrator of a SQL Server 2000 computer. Your company purchased an accounting application from a vendor. The application stores its data in a database named Accounting on the server. The tables in this database contain columns that function as primary keys, but PRIMARY KEY and FOREIGN KEY constraints are not used.

You need to replicate data from this database to another SQL Server computer. This server will use the replicated data to generate reports. Most reports will run each month, but the accounting department needs to have the ability to run reports at any time. Reports should be accurate through the last full working day.

You cannot make any changes to the database, but you need to implement replication. Which two actions should you take? (Each correct answer presents part of the solution. Choose two)
选项 A、Implement merge replication.
B、Implement snapshot replication.
C、Implement transactional replication.
D、Schedule replication to run continuously.
E、Schedule replication to run during off-peak hours.
答案B,E
解析 Explanation:
B: Snapshot replication is the best solution since we will not be making changes to the Accounting database but need to replicate data from the database to another SQL Server computer and as the accounting department needs to have the ability to run reports that only need to be accurate through the last full working day.

E: Only one snap-shot replication would be necessary per day, and to make as little impact on the daily use of the server this replication should be scheduled during off-peak hours.

Note: SQL Server 2000 provides three types of replication that can be used in distributed applications. These are snapshot replication, transactional replication and merge replication. Each type provides different capabilities.

Snapshot replication is the process of copying and distributing data and database objects exactly as they were at the time the snapshot was taken. Snapshot replication does not require continuous monitoring of changes because changes made to published data are not propagated to the Subscriber incrementally. Subscribers are updated with a complete data set and not individual transactions. As snapshot replication replicates an entire data set at one time, it may take longer to propagate data modifications to Subscribers. This type of replication in helpful in situations when data is mostly static and does not change often; it is acceptable to have copies of data that are out of date for a period of time; replicating small volumes of data; and when sites are often disconnected and high latency (the amount of time between when data is updated at one site and when it is updated at another) is acceptable.

In transactional replication, an initial snapshot of data is propagated to Subscribers, and then when data modifications are made at the Publisher, the individual transactions are captured and propagated to Subscribers. SQL Server 2000 monitors INSERT, UPDATE, and DELETE statements, and changes to stored procedure executions and indexed views. SQL Server 2000 stores the transactions affecting replicated objects and then it propagates those changes to Subscribers continuously or at scheduled intervals. Transaction boundaries are preserved. Transactional replication is used when data modifications must be propagated to Subscribers, often as they occur; transactions must be atomic; subscribers are mostly connected to the Publisher and when the application will not tolerate high latency for Subscribers receiving changes.

Merge replication allows various sites to work autonomously and merge data modifications made at multiple sites into a single, uniform result at a later time. The initial snapshot is applied to Subscribers and then SQL Server 2000 tracks changes to published data at the Publisher and at the Subscribers. The data is synchronized between servers either at a scheduled time or on demand. Updates are made independently at more than one server, so the same data may have been updated by the Publisher or by more than one Subscriber. Therefore, conflicts can occur when data modifications are merged. This type of replication is helpful when multiple Subscribers need to update data at various times and propagate those changes to the Publisher and to other Subscribers; Subscribers need to receive data, make changes offline, and synchronize changes later with the Publisher and other Subscribers; the application latency requirement is either high or low, and when site autonomy is critical.

Incorrect Answers:
A: No data at the second server will not be changed so there is no need to use merge replication.

C: We don’t need the accuracy of transactional replication. In this scenario high latency is permissible as the reports must be accurate only through the last full working day.

Note: Transactional replication is used when data modifications must be propagated to Subscribers, often as they occur; transactions must be atomic; subscribers are mostly connected to the Publisher and when the application will not tolerate high latency for Subscribers receiving changes.

D: As the reports must be accurate through the last full working day, it not necessary to run replication continuously. Running replication continuously will place locks on the destination database, which would hamper the performance of queries and actions run against the destination database.
转载请注明原文地址:https://www.kaotiyun.com/show/ZuhZ777K
本试题收录于： 微软70228题库微软认证分类
0

微软70228

微软认证

相关试题推荐
随机试题
最新回复(0)