Share via

  • Article

Full-Text Engine

The SQL Server Full-Text Engine is a full-text indexing and search engine. In SQL Server 2008, the Full-Text Engine has been fully integrated into the Database Engine. The Full-Text Engine now resides in the SQL Server process (sqlservr.exe), rather than residing in a separate process (msftesql.exe). Integrating the Full-Text Engine into the Database Engine has improved full-text manageability, optimization of mixed query, and overall performance. For information about full-text search components that reside in the SQL Server process, see Full-Text Search Architecture.

SQL Server 2008 provides side-by-side installs of the Full-Text Engine. For each instance of SQL Server, there is a dedicated instance of the Full-Text Engine, including dedicated components such as word breakers and filters, resources such as memory, and configuration such as service-level settings at the instance level.

Function of the Full-Text Engine

The Full-Text Engine is responsible for reading from and writing to full-text indexes, which are now stored in SQL Server. The Full-Text Engine supports the following:

  • Indexing
    The Full-Text Engine performs a variety of indexing tasks, among them gathering data from the base table and sending the data to the filter daemon host for processing. The filter daemon host filters and tokenizes data provided by the full-text gatherer. After the data has been filtered and the words have been broken, the SQL Server process receives the results and indexes them as an inverted index structure that is stored in database as an internal table.  

  • Querying
    The Full-Text Engine processes full-text search queries and determines which entries (rows or documents) in the base table meet the full-text selection criteria. For each entry that meets the selection criteria, it returns the identity of each row along with an optional ranking value. This information is used to construct the query-result set.

    The Full-Text Engine supports both simple and sophisticated word or phrase searches. For more information, see Supported Forms of Query Terms