uCertify Articles » 70-443 http://www.ucertify.com/article Just another WordPress weblog Wed, 04 Nov 2009 00:00:00 +0000 http://wordpress.org/?v=2.8.4 en hourly 1 What is the SystemProcessor Queue Length counter? http://www.ucertify.com/article/what-is-the-system-processor-queue-length-counter-2.html http://www.ucertify.com/article/what-is-the-system-processor-queue-length-counter-2.html#comments Wed, 30 Sep 2009 00:00:00 +0000 uCertify 3601467

The SystemProcessor Queue Length counter displays the number of threads ready in the processor queue but not currently able to use the processor. The Processor: %Processor Time counter should be measured with a counter. If its value is two, it indicates a CPU bottleneck. This results in slow performance of the server. To eliminate the performance problem, several steps can be used such as reducing the number of processors, using = multiple processors, updating the processors, or installing new and updated CPUs.

Share/Bookmark]]> http://www.ucertify.com/article/what-is-the-system-processor-queue-length-counter-2.html/feed 0 What are the types of capacity planning? http://www.ucertify.com/article/what-are-the-types-of-capacity-planning-3.html http://www.ucertify.com/article/what-are-the-types-of-capacity-planning-3.html#comments Wed, 30 Sep 2009 00:00:00 +0000 uCertify 3601480

The following are the types of capacity planning:

  • Preconfiguration capacity planning: This capacity planning involves the planning and implementation of hardware resources that will be used to accomplish a given task.

  • Postconfiguration capacity planning: This capacity planning involves monitoring performance of the current hardware resources. It is also used to analyze trends and plan the upgrade or replacement of hardware as per requirements.
Share/Bookmark]]> http://www.ucertify.com/article/what-are-the-types-of-capacity-planning-3.html/feed 0 Replication http://www.ucertify.com/article/replication_3601460-2.html http://www.ucertify.com/article/replication_3601460-2.html#comments Tue, 29 Sep 2009 00:00:00 +0000 uCertify 3601460

Replication is a type of configuration used to copy data and database objects from one database instance to another. A database administrator has to configure a server for replication, and monitor it in order to detect and resolve issues regarding replication.

Requirements for replication: The following roles are required for replication configuration:

  • Distributor: The server that is configured to the Distributor role distributes data to Subscribers. It creates tables and files to store the data for replication. The computer to be used as a Distributor should have a large amount of storage space.

  • Publisher: The server that is configured to the Publisher role stores data for replication. It detects data changes and maintains information about all publications.

  • Subscriber: The server that is configured to the Subscriber role receives data published by either Publisher or Distributor. The Subscriber configuration depends on the type of replication chosen to replicate data. Each replication type should have a different configuration for a Subscriber.

Replication objects: After the configuration of Publisher, Distributor, and Subscriber servers, various objects are needed to replicate the data. These objects are as follows:

  • Article: Article is an object that includes various database objects such as tables, views, stored procedures, etc. Each of the database objects can be selected as an article. All the articles are combined to create a publication.

  • Publication: Publication is a combination of articles. The data to be received by the Subscriber is sent as a publication. An article must be placed in a publication in order to send the data to the Subscriber.

  • Subscription: Subscription receives the publication sent by the Publisher or Distributor. Subscription is of the following two types:
    • Push subscription: A push subscription is created at the Publisher. Data is pushed to the Subscribers by the Merge or the Distribution agent.

    • Pull subscription: A pull subscription is created at the Subscriber. Data is pulled by the Merge or the Distribution agent from the Publisher.

Replication Models: A DBA has to decide the type of replication model before replicating the data. He has to decide the configuration for the Publisher, Distributor, and Subscriber databases. The following replication models can be applied before data replication:

  • Peer-to-Peer: It is a type of replication model in which a single database is hosted by multiple servers. The purpose of this model is to maintain a database and manage database failures. If a server containing the database is unavailable, the applications can access the data through other servers hosting the database.

  • Central Publisher: The Central Publisher is a replication model in which the Publisher and the Distributor roles are maintained on the same SQL Server 2005/2008 computer. The Subscriber role is configured on other servers. The maintenance of the server is easy, but it adds additional load on the server.

  • Central Publisher with Remote Distributor: The Central Publisher with Remote Distributor is a replication model in which the Publisher and the Distributor roles are maintained on separate SQL Server 2005/2008 computers. The Subscriber role is configured on other servers. The workload of the Publisher server is distributed, but an additional Distributor server is to be maintained.

  • Central Subscriber: Central Subscriber is a replication model in which the Subscriber role is configured on a single server. It is used to collect data from several Publisher servers. It can also be configured to send data to other servers.

  • Publishing Subscriber: Publishing Subscriber is a replication model in which data is obtained from a Publisher and transferred to other Subscribers. This replication model can be used with other replication models for the replication of data.

Types of replication: There are different types of replication and a DBA has to choose which type of replication he needs to configure. Some applications need only a snapshot of the database, whereas others need incremental data changes. The different types of replication are as follows:

  • Transactional Replication: Transactional replication starts with a snapshot of the Publication database objects and data. The modifications made in the data and in the schema are delivered to the Subscriber database, simultaneously. Transactional replication is used in a server-to-server environment in the following conditions:

    • When a Publisher or Subscriber is a non-SQL Server database, such as Oracle.

    • When incremental changes are to be made at the Subscriber.

    • When an application requires low latency.

    • When an application requires access to data changes.

    • When the Publisher has a high volume of insert, update, or delete operations.
  • Merge Replication: Merge replication starts with a snapshot of the Publication database. Triggers track the modifications in the data and the schema at the Publisher or the Subscriber server. Merge replication is performed in a server-to-client environment. It should be performed in the following conditions:
    • When same data is to be updated by multiple Subscribers.

    • When data changes are to be propagated to the Publisher and other Subscribers.

    • When offline data changes are to be synchronized on the Publisher and other Subscribers.

    • When conflicts are to be detected and resolved.

    • When each Subscriber needs different partitions of data.

    Merge replication does not take place if a Publisher or Subscriber is a non-SQL Server database.

  • Snapshot Replication: Snapshot replication distributes data at a specific moment. It does not monitor the modifications in data or schema. The entire snapshot is sent to Subscribers when synchronization occurs. The snapshot replication can be performed in the following conditions:
    • When data changes infrequently.

    • When small volumes of data are to be replicated.

    • When large volumes of changes occur over a short period of time.

    It does not take place when a Publisher or a Subscriber is a non-SQL Server database.

Replication Agents: Replication agents are used during the replication of data. They store data published by a Publisher. They also initialize and monitor the replication. Replication agents are as follows:

  • Queue Reader Agent: The Queue Reader agent runs with a transactional replication. It runs at the Distributor database and transfers the changes from the Subscribers to the Publisher database. Only one instance of this agent exists to serve all the Publishers.

  • Snapshot Agent: The Snapshot agent is used with all types of replication. It is used to prepare snapshot files that contain the schema and the data of published tables and database objects in the publication database. It also records synchronization jobs in the distribution database. It runs at the Distributor database instance.

  • Log Reader Agent: The Log Reader agent is used with a transactional replication. It is used to monitor the transaction log of each database configured for transactional replication. It copies the transactions that are marked for replication to the Distribution database.

  • Distribution Agent: The Distribution agent is used with transactional and snapshot replications. It is used in a transactional replication to move transaction files and data to the Subscribers database. It runs at the Subscribers database for pull subscriptions or at the Distributors database for push subscriptions.

  • Merge Agent: The Merge agent is used with the merge replication. It applies the snapshot to the Subscriber database instance, and the changes in the Subscriber are reflected in the Publisher database. The Merge agent uploads data changes from Subscribers to Publishers, and downloads data from Publishers to Subscribers.
Share/Bookmark]]> http://www.ucertify.com/article/replication_3601460-2.html/feed 0 What is Database Engine Tuning Advisor? http://www.ucertify.com/article/what-is-database-engine-tuning-advisor.html http://www.ucertify.com/article/what-is-database-engine-tuning-advisor.html#comments Thu, 03 Sep 2009 00:00:00 +0000 uCertify 7101039

Database Engine Tuning Advisor is a tool that tunes a database for better query performance by analyzing a workload that is a set of T-SQL statements. It suggests clustered indexes, nonclustered indexes, indexed views, and partitioning. Database Engine Tuning Advisor in SQL Server 2005 replaces the Index Tuning Wizard of SQL Server 2000. This tool provides two user interfaces: a graphical user interface (GUI) and the dta command prompt utility.

Share/Bookmark]]> http://www.ucertify.com/article/what-is-database-engine-tuning-advisor.html/feed 0 What are the disadvantages of database server consolidation? http://www.ucertify.com/article/what-are-the-disadvantages-of-database-server-consolidation.html http://www.ucertify.com/article/what-are-the-disadvantages-of-database-server-consolidation.html#comments Fri, 21 Aug 2009 00:00:00 +0000 uCertify 3601486

The disadvantages of database server consolidation are as follows:

  • It requires high-availability options and failover clustering.

  • It results in the performance deterioration of the server.

  • It requires a disk with high storage and high input-output capacity.

  • It requires more memory.

  • It increases CPU usage.

  • It increases network traffic.
Share/Bookmark]]> http://www.ucertify.com/article/what-are-the-disadvantages-of-database-server-consolidation.html/feed 0 What are SQL Server Securables? http://www.ucertify.com/article/what-are-sql-server-securables.html http://www.ucertify.com/article/what-are-sql-server-securables.html#comments Fri, 21 Aug 2009 00:00:00 +0000 uCertify 3601489

SQL Server Securables contains three scopes, which are used to assign permissions to users. The securables are nested and each securable contains various other securables. The securable scopes are as follows:

  • Server: It includes server roles, logins, etc.

  • Database: It includes database users, application roles, database roles, etc.

  • Schema: It includes various database objects such as tables, views, procedures, etc.

The securables are used to assign permissions to the users based on scope and the tasks assigned. The issues related to the connectivity to databases, accessing database objects, etc., can be resolved by granting or denying the permissions to the users.

Share/Bookmark]]> http://www.ucertify.com/article/what-are-sql-server-securables.html/feed 0 What is the IS_SRVROLEMEMBER function? http://www.ucertify.com/article/what-is-the-is_srvrolemember-function.html http://www.ucertify.com/article/what-is-the-is_srvrolemember-function.html#comments Fri, 21 Aug 2009 00:00:00 +0000 uCertify 3601490

The IS_SRVROLEMEMBER function provides information about whether or not a login is a member of a specific role. A database administrator can use this information to assign permissions to users. The syntax of using the function is as follows:

IS_SRVROLEMEMBER (’role’, ‘login’)

The name of a server role is to be provided in the role argument. The server roles are sysadmin, dbcreator, bulkadmin, diskadmin, processadmoin, serveradmin, setupadmin, securityadmin.

The login argument is optional. It is used to provide the SQL Server 2005 login that is to be checked for its membership. If this argument is not provided, the membership of current user is checked.

Share/Bookmark]]> http://www.ucertify.com/article/what-is-the-is_srvrolemember-function.html/feed 0 Joins: Its Types and Uses http://www.ucertify.com/article/joins-its-types-and-uses.html http://www.ucertify.com/article/joins-its-types-and-uses.html#comments Fri, 21 Aug 2009 00:00:00 +0000 uCertify 3601509

The JOIN clause is used with the SELECT statement to retrieve data from more than one table. This clause matches data from multiple tables. This results in the creation of a table, which is a combination of columns from multiple tables.

Uses of Joins: Join is a feature of SQL Server. It joins the tables by means of the combination of a primary key and a foreign key. The result of the query generates a temporary table containing the columns from two or more tables. An example of using the JOIN clause is as follows:

Suppose there is a database named Manufacture, which contains the tables Car and Dealer. The Car table has the following columns:

  • CarID

  • CarName

  • CarModel

  • CarPrice

The Dealer table has the following columns:

  • DealerID

  • DealerName

  • Address

  • Phone

  • Mobile

  • Commission
  • CarID

The CarID column in the Car table is defined as the primary key, and the CarID column in the Dealer table is defined as the foreign key. If a user wants to retrieve the car name, car price, car model, dealer name, and commission, the following query should be executed:

SELECT CarName, CarModel, CarPrice, DealerName, Commission FROM Car JOIN Dealer WHERE Car.CarID=Dealer.CarID

Aliases: If two tables contain the same columns, it results in a conflict. To avoid this conflict, a user can use either the table names or aliases. The aliases improve the readability of a SQL statement. Aliases can be defined by using letters, numbers, or a combination of these. The following example will demonstrate the use of aliases:

SELECT c.CarName, c.CarModel, c.CarPrice, d.DealerName, d.Commission FROM Car c JOIN Dealer d WHERE Car.CarID=Dealer.CarID

In the above example, c is the alias for the Car table, and d is the alias for the Dealer table.

Types of Joins: The join types used in SQL Server are as follows:

Inner Joins: In this type of join, data in various columns are compared by using the comparison operators. The inner joins are also known as equi-joins. It returns only those rows that have an equal value. The following example will demonstrate the use of the inner join:

SELECT c.CarName, c.CarModel, c.CarPrice, d.DealerName, d.Commission FROM Car c INNER JOIN Dealer d ON c.CarID=d.CarID

Outer Joins: Outer joins return rows from one of the tables specified in the query. There are three types of outer joins as follows:

LEFT OUTER JOIN: When this join is used in a query, all the rows from the left table will be retrieved, and only the matching rows from the right table will be retrieved. If a row from the left table does not match with any of the rows of the right table, NULL is inserted in that row of the right table. The following example will demonstrate the use of the LEFT OUTER JOIN:

SELECT c.CarName, c.CarModel, c.CarPrice, d.DealerName, d.Commission FROM Car c LEFT OUTER JOIN Dealer d ON c.CarID=d.CarID

In the above example, all rows from the Car table will be retrieved, and only the matching rows from the Dealer table will be retrieved.

RIGHT OUTER JOIN: When this join is used in a query, all the rows from the right table will be retrieved, and only the matching rows from the left table will be retrieved. If a row from the right table does not match with any of the rows of the left table, NULL is inserted in that row of the left table. The following example will demonstrate the use of the RIGHT OUTER JOIN:

SELECT c.CarName, c.CarModel, c.CarPrice, d.DealerName, d.Commission FROM Car c RIGHT OUTER JOIN Dealer d ON c.CarID=d.CarID

In the above example, all rows from the Dealer table will be retrieved, and only the matching rows from the Car table will be retrieved.

FULL OUTER JOIN: When this join is used in a query, all the rows from all the tables specified in the query will be retrieved. The following example will demonstrate the use of the FULL OUTER JOIN:

SELECT c.CarName, c.CarModel, c.CarPrice, d.DealerName, d.Commission FROM Car c FULL OUTER JOIN Dealer d ON c.CarID=d.CarID

In the above example, all rows from the Dealer and Car tables will be retrieved.

Cross Joins: When this join is used in a query, the result is the multiplication of the number of rows from the left table with the number of rows from the right table. The following example will demonstrate this:

SELECT c.CarName, c.CarModel, c.CarPrice, d.DealerName, d.Commission FROM Car c CROSS JOIN Dealer d ON c.CarID=d.CarID

If the Car table contains ten rows and the Dealer table contains twelve rows, the result will be 120 rows.

Self Join: This join is used to join a table with itself. In this case, a user will have to use two unique aliases for the table. The following example will demonstrate this:

SELECT c1.CarName, c1.CarModel, c1.CarPrice FROM Car c1 INNER JOIN Car c2 ON c1.CarID=c2.CarID

Share/Bookmark]]> http://www.ucertify.com/article/joins-its-types-and-uses.html/feed 0 How to create a partition for an existing table and index? http://www.ucertify.com/article/how-to-create-a-partition-for-an-existing-table-and-index.html http://www.ucertify.com/article/how-to-create-a-partition-for-an-existing-table-and-index.html#comments Fri, 21 Aug 2009 00:00:00 +0000 uCertify 3601539

Take the following steps to create a partition for an existing table and index:

  1. Create a partition function: A partition function can be created by using the CREATE PARTITION FUNCTION STATEMENT. An example is given below:

    CREATE PARTITION FUNCTION myfunc
    RANGE LEFT FOR VALUES(100, 200, 300)

  2. Create a partition scheme: A partition scheme can be created by using the CREATE PARTITION SCHEME statement. An example is given below:

    CREATE PARTITION SCHEME myps AS myfunc TO (mv1), (mv2), (mv3)

  3. Drop the clustered index: The clustered index can be dropped by using the DROP INDEX statement. The statement is as follows:

    DROP INDEX index_name ON table_name.column_name

  4. Recreate the clustered index on the partition scheme: The index can be created on a partitioned scheme as follows:

    CREATE CLUSTERED INDEX ON employee.empid ON myps(empid)

The table is automatically partitioned according to the partition scheme of the clustered index.

Share/Bookmark]]> http://www.ucertify.com/article/how-to-create-a-partition-for-an-existing-table-and-index.html/feed 0 What are the types of capacity planning? http://www.ucertify.com/article/what-are-the-types-of-capacity-planning.html http://www.ucertify.com/article/what-are-the-types-of-capacity-planning.html#comments Thu, 20 Aug 2009 00:00:00 +0000 uCertify 3601480

The following are the types of capacity planning:

  • Preconfiguration capacity planning: This capacity planning involves the planning and implementation of hardware resources that will be used to accomplish a given task.

  • Postconfiguration capacity planning: This capacity planning involves monitoring performance of the current hardware resources. It is also used to analyze trends and plan the upgrade or replacement of hardware as per requirements.
Share/Bookmark]]> http://www.ucertify.com/article/what-are-the-types-of-capacity-planning.html/feed 0