This is an old revision of the document!


DRAFT

Omnidex Indexing

Overview

Omnidex adds performance and flexibility to database applications. These benefits are achieved by a combination of the indexes and the Omnidex SQL Engine. The indexes provide the basic speed and flexibility, while the SQL Engine analyzes statements and determines the best way to use the indexes. Omnidex is different than most SQL engines in that many indexes are used to fulfill a query. It is common for all criteria, table joins, aggregations and ordering to be satisfied in a coordinated effort between many indexes. Complex queries may even be satisfied without ever accessing the underlying data.

Three basic types of indexes allow SQL statements to be optimized: Omnidex indexes, QuickText indexes and FullText indexes. They also provide features such as textual searches, fuzzy searches and geographical searches. Omnidex indexes provide the bulk of Omnidex functionality, optimizing criteria, table joins, aggregations and ordering. QuickText indexes provide basic textual searches on columns like name and address fields, product descriptions and general text. FullText indexes provide sophisticated text indexing designed for large blocks of textual data.

ActiveCounts is an Omnidex indexing approach that allows applications to show criteria values and their counts at each step of a search. Users can see their choices and know how the counts would change before making their selection. This feature is especially valuable for businesses that allow online searching of their database through web applications.

PowerSearch is an Omnidex indexing approach that allows applications to activate a fuzzy search. The user can enter their best guess at the criteria, and then PowerSearch will use misspellings, phonetic searches, synonyms lists and word forms to locate the data. This feature is especially valuable for applications that search for specific data based on a user's search terms.

Omnidex also has advanced options that optimize complex or unusual situations. One example is Index Groups, in which multiple columns are indexed as though they were one column. Another example is geographic indexes which index latitude and longitude coordinates in preparation for geographic radius searches.

Indexes are installed by simply designating indexes on the columns involved in a query. There are times when indexing options are valuable, and there are also times when multi-column indexes are created. Once the indexes are designated, all indexes can be built concurrently, greatly reducing the time required to build the indexes.

To learn more about Omnidex indexing, read the following sections:




The next section on Omnidex Indexing is Omnidex Indexes.

 
Back to top
admin/indexing/home.1260637055.txt.gz · Last modified: 2016/06/28 22:38 (external edit)