Action disabled: source

Integration: Relational Databases

Querying Relational Databases

Query Optimization

Omnidex allows the relational database tables and the Omnidex indexes to be queried using standard SQL statements using an ODBC or JDBC interface. The query optimizer within Omnidex will evaluate the SQL statement and choose the best approach for processing. Queries will only access the indexes or the database tables if needed.

Some queries will be processed entirely within the Omnidex indexes with no need to access the underlying relational database, as shown below.

Some queries may be processed entirely in the relational database with no need to access the Omnidex indexes, as shown below.

Some queries will be processed using a combination of Omnidex indexes and the underlying relational database, as shown below.

Omnidex SQL

Queries can be submitted using the Omnidex SQL language. Omnidex SQL consists of ANSI SQL, supplemented with Omnidex extensions and functions. Omnidex also supports a limited number of proprietary functions found in the popular relational databases. Access to these relational database extensions can be controlled using the SET SQL_SYNTAX statement. Applications using a particular SQL dialect from a relational database will need to have the syntax adjusted to Omnidex SQL.

Omnidex will attempt to optimize all queries; however, like the relational databases, queries can sometimes be tuned by restructuring or reorganizing the query. Omnidex does favor certain approaches to queries. Additionally, many poor-performing relational queries have been customized by talented and hard-working database administrators to achieve the best performance for the underlying relational database. This may not lead to the best performance with Omnidex. Consult the documentation on optimizing queries to understand more about how queries can be optimally structured for Omnidex.

Additional Resources

See also:

 
Back to top
integration/rdbms/queries.txt ยท Last modified: 2016/06/28 22:38 (external edit)