Full-text indexing is a feature of SQL Server 2008
that allows you to carry out sophisticated searches of text-based data
called a full-text search.
A full-text search is different from a normal search of data through a
normal index because it enables you to use linguistic-based searches.
For example, you could search a text-based column for inflectional
forms of the word run, which would return results including running and ran. You could also search for similar words using the thesaurus feature, so searching for bicycle might return results including bike, pushbike, tandem and tricycle.
the exam you need to focus on configuring and managing full-text
indexes to support full-text searches rather than how to implement the
Configuring Full-Text Indexing
databases in SQL Server 2008 are enabled for full-text indexing by
default, so the first step you need to make is to create a full-text catalog,
which is a logical object for grouping together full-text indexes.
Microsoft recommends that full-text indexes with similar update
activity patterns are grouped together in a full-text catalog, so that
population schedules can be applied at the catalog level to reduce
resource usage during population.
Full-text catalogs can be created from the right-click menu in SQL Server Management Studio at <instance> | Databases | <database> | Storage | Full Text Catalogs or by executing the following T-SQL:
CREATE FULLTEXT CATALOG <name>
you can create a full-text index for a table, you’ll need to make sure
it has an existing unique, single column, nonnullable index. The
full-text index will base its index keys on this.
you’ve fulfilled all the requirements you can create a full-text index
using the CREATE FULLTEXT INDEX t-sql command, or use Management Studio
by navigating to the table and selecting Full-Text Indexes from the right-click menu.
Managing Full-Text Indexes
Once you’ve created a full-text index, the process of filling it is referred to as populating
the index. This is done initially when you create it, and by default
the index will stay up to date as the underlying data changes. There
are scenarios, however, where this default behavior is undesirable. As
the population process is resource intensive, if you have frequent
updates to your underlying text data, it might be prohibitive for you
to keep the index automatically updated.
In this scenario, you can modify the default behavior of change tracking, which is to automatically track changes and populate the index. You can configure it to manual, which specifies that changes will be tracked but not propagated until you run or schedule the ALTER FULLTEXT INDEX ON <tablename> START UPDATE POPULATION t-sql command or set it to off, in which case changes will not be tracked or propagated until you run a FULL or INCREMENTALtimestamp data type to be present on the underlying table.
population. An incremental population will update the index with
changed rows since the last population, but it requires a column with
the timestamp data type to be present on the underlying table.