Skip to main content

How to Optimize Indexes in SQL Server

SQL Server automatically maintains indexes whenever insert, update, or delete operations are made to the underlying data. These modifications cause the information within the index to become scattered in the database. Fragmentation exists when indexes within the index to become scattered in the database.

Fragmentation exists when indexes have pages where the logical ordering does not match the physical ordering within the data file. Heavily fragmented indexes can degrade the query performance and cause your application to respond slowly.

Fragmentation normally occurs when a large number of insert and update operations are performed on the table. Fragmented data can cause the SQL Server to perform unnecessary data reads. This affects the performance of the query. Therefore, index defragmentation is necessary to speed up the query performance.

The first step in deciding which defragmentation method to use is to determine the degree of index fragmentation. You can detect index fragmentation by using the sys.dm_db_index_physical_stats system function.

The following statement displays a list of all the indexes on the HumanResources.Employee table with their fragmentation level:

SELCET a.index_id AS IndexID, name AS IndexName,
Avg_fragmentation_in_percent AS Fragmentation
FROM sys.dm_db_index_physical_stats (DB_ID (N’AdventureWorks’),
OBJECT_ID (‘HumanREsources.Employee’), NULL, NULL, NULL ) AS a
JOIN sys.indexes AS b ON  a.object_id = b.object_id AND a.index_id = b.index_id ORDER BY Fragmentation desc

In the preceding output given by this query, you can notice that the AK_Employee_LoginID index shows a high level of fragmentation.

After the degree of fragmentation is known, the fragmentation needs to be corrected. The following table lists the actions to be taken at different fragmentation levels to defragment the index.

Fragmentation Level                         Action to be Taken
>5% and <=30%  ALTER INDEX REORGANIZE
>30%  ALTER INDEX REBUILD WITH (ONLINE – ON)

Method of Degragmentation

You can execute the following command to defragment the AK_Employee_LoginID index:

ALTER INDEX AK_Employee_LoginID ON HumanResources.Employee REBUILD
After executing the preceding command, the degree of fragmentation is reduced.

Comments

Popular posts from this blog

difference between structure and union in C Language

In c language article we will see the difference between union and structure. Both are the user define datatype in c language. See the table which is mentioned below: ASP.NET Video Tutorial Series Structure Union1.The keywordstruct is used to define a structure 1. The keyword union is used to define a union. 2. When a variable is associated with a structure, the compiler allocates the memory for each member. The size of structure is greater than or equal to the sum ofsizes of its members. The smaller members may end with unused slack bytes. 2. When a variable is associated with a union, thecompiler allocates thememory by considering the size of the largest memory. So, size of union is equal to the size of largest member. 3. Each member within a structure is assigned unique storage area of location. 3. Memory allocated is shared by individual members of union. 4. The address of each member will be in ascending order This indicates that memory for each member will start at different offset v…

Difference between Linear search and Binary Search in c language

SQL Video Channel : Download all SQL Video



Binary Search Linear Search Works only on sorted items. such as  1,2,3,4,5,6  etc
Works on sorted as well as unsorted items. 12,4,5,3,2,1 etc Very efficient if the items are sorted Very efficient if the items are less and present in the beginning of the list. such as Suppose your list items are : 12,3,4,5,1 and you want to search 12 number then you get beginning in the list. Works well with arrays and not on linked lists. Works with arrays and linked lists.
Number of comparisons are less More number of comparisons are required if the items are present in the later part of the array or its elements are more.

Memory representation of Linked List Data Structures in C Language

Memory representation of Linked List

             In memory the linked list is stored in scattered cells (locations).The memory for each node is allocated dynamically means as and when required. So the Linked List can increase as per the user wish and the size is not fixed, it can vary.

               Suppose first node of linked list is allocated with an address 1008. Its graphical representation looks like the figure shown below:


      Suppose next node is allocated at an address 506, so the list becomes,



  Suppose next node is allocated with an address with an address 10,s the list become,


The other way to represent the linked list is as shown below:




 In the above representation the data stored in the linked list is “INDIA”, the information part of each node contains one character. The external pointer root points to first node’s address 1005. The link part of the node containing information I contains 1007, the address of next node. The last node …