The Real MTCS SQL Server 2008 Exam 70/432 Prep Kit- P84 ppt

397 Exam objectives review: ˛ Summary of Exam Objectives ˛ Exam Objectives Fast Track ˛ Exam Objectives Frequently Asked Questions ˛ Self Test ˛ Self Test Quick Answer Key Managing Replication Chapter 9 MCTS SQL Server 2008 Exam 432 Exam objectives in this chapter:  Replication Scenarios  Replication Types  Replication Configuration  Conflict Resolution  DDL Replication  Security Considerations  Replication Performance  Monitoring 398 Chapter9•ManagingReplication Introduction Replication technology has been with SQL Server for several versions now. With each new version of SQL Server Replication you will see improvements; the same applies to SQL Server 2008 as well. Replication is copying and distributing data between databases and then synchronizing them to maintain consistency and integrity. Head of the Class… Why Replication and Which Replication Type Because there are several methods for disaster recovery in SQL Server, most of the time, users are confused about which method to use when. Most of the users are more aware of replication than the other types (i.e., clustering, mirroring and log shipping) available to them, so they will decide to use replication without knowing its features and limitations. However, replication has several configurations to do compared with other disaster recovery methods. More importantly, there are three main types of replication, and users may not know which type of replication will be better suited for their problem. Replication is one of four methods available for high availability of your SQL Server: Clustering  Mirroring  Log shipping  Replication  Replication has the advantage over the other technologies because it has the ability to filter rows or columns in a given table. Unlike mirroring and log shipping, replication has the ability to provide readability in the primary and secondary databases. ManagingReplication•Chapter9 399 You can implement replication for the following scenarios:  Improving Performance By implementing replication, data is distributed among several database servers. You can point your applications to these servers to distribute load. For example, you can use read-only databases for reporting and analysis and therefore, the main database will handle only the transactions.  Improving Availability In case of disaster, you need to have an additional database so that it takes the load. Replication can be used for such scenarios. However, unlike in database mirroring, in replication you need to find a mechanism to change the connection of application from the failed database node to the active database node.  Connecting with Mobile Devices In case you have users who are using mobile devices, you can use replication. For example, mobile users can connect to database and synchronize data, then discount from the database. After doing his normal activities (i.e., quotations, sales), the user can connect to the replicated database and synchronize it again.  Transferring Data between Database Servers In today’s business environment, you need to transfer data between servers. For example, the Financial Division needs sales information of all the retail locations, the HR division needs employees’ attendance of all the factories around the country, and so on.  Using Replication as a Staging Database Server in a Data Warehouse In a data warehouse application, most of the time you are not allowed to modify schema in the source system. In this type of situation, you can use replication to load source data into a staging server and apply any schema changes you need to the staging database. Te s T Da y Tip Replication is not only a disaster recovery (DR) method—you can use replication as load balancing as well. Therefore when selecting replication as a technology, you should consider the other options available with replication. Replication is analogous to the publishing industry. A publisher can collect a set of articles and create any type of publication, and has the sole rights to decide on 400 Chapter9•ManagingReplication the content. There is no rule to prevent a publisher from creating any publication he or she wants. After the publication is done, publisher will transfer the publications to the distributor, and it is distributor’s duty to ensure that the relevant publication has passed to the subscriber depending on the subscriber’s subscription. Replication Scenarios In replication, there are few concepts and scenarios we need to know before jumping into configuring replication in SQL Server 2008. Article Article is a database object that you want to replicate; it can be a Table, View, or Store procedure, or other objects. In some cases of replication, Article can be subset of a database object. For example, you can replicate a selected number of columns instead of all columns. Moreover, you can replicate a selected number of rows by adding a where clause. Publication Publication is a collection of articles. Article is a database object that you want to replicate. Database object can be a Table, View, Store procedure or other objects. One article can be in one or more publications. Publication can be treated as a logical grouping for the articles. For example, if you want to replicate only the employee master data, you can include employee table article, address article, and any other relevant views and stored procedures relevant can be added to the publication. Subscription Subscription is a request for a copy from the publication. Subscription defines where to receive the publication, what the frequency for the receiving is, and whether data is pulled from the subscriber end or data is pushed from publisher end. Publisher Publisher is a database instance that holds publications for other database instances to receive. There can be several different types of publication in one publisher. ManagingReplication•Chapter9 401 Distributor Each publisher has a distributor to store replication status data, and metadata about the publication. Sometimes the distributor acts as a data queue for data moving from the publisher to the subscribers and vice versa. There are two types of distributors, local and remote. Local distributor is where the publisher and distributor are in one database server instance. Subscriber Subscriber is a database instance that receives replicated data from the publisher. In some replication types the subscriber also passes data to the publisher. Replication Types Unlike other disaster recovery methods, replication itself has several configuration types. You can select the correct replication type to suite your requirement. Snapshot As the name specifies, Snapshot Replication distributes data as it appears in the publisher at a specific moment of time. Whenever the Subscriber called for subscrip- tion, it will send the snapshot of the current Publisher and send it to the Subscribers. Because of the fact that every time Snapshot Replication sends the entire snapshot, there are limitations for using Snapshot Replication. Following are the scenarios where you can use Snapshot Replication: Replicating a small volume of data. If you want to replicate less data  volume (like system data, which is less than 100 rows of data), you can use snapshot replication. Infrequently changing data. Sometimes there are data that rarely change.  For example, some organizations won’t change their price list frequently, maybe just once a year. For those cases, you can configure snapshot replication. When it is acceptable to have out-of-data copy at the subscriber end.  Every time you don’t need subscriber data you are allowed to have out-of-data at the subscriber end. Replicating table that does not have a primary keys. Snapshot replication  does not require primary article tables. Though it is very rare that you will not see a primary key in your tables, if there is no primary key, you can implement Snapshot Replication. . technology has been with SQL Server for several versions now. With each new version of SQL Server Replication you will see improvements; the same applies to SQL Server 2008 as well. Replication. and then synchronizing them to maintain consistency and integrity. Head of the Class… Why Replication and Which Replication Type Because there are several methods for disaster recovery in SQL Server, . high availability of your SQL Server: Clustering  Mirroring  Log shipping  Replication  Replication has the advantage over the other technologies because it has the ability to filter rows