This is an old revision of the document!
There are several different types of Omnidex indexes. Each is designed for a different scenario, and all can be used in the same search. For each column that is indexed, choose the index type that best meets the needs of the queries.
Omnidex indexes provide the bulk of Omnidex functionality. These indexes are placed on columns that are involved in Omnidex criteria, table joins, aggregations and ordering. Omnidex indexes can also be used to retrieve data without having to access the database.
Typically, administrators place Omnidex indexes on all columns that are involved in the SQL statement. The Omnidex SQL Engine will use each index in different ways depending on how the corresponding column is used in the SQL statement. In a given query, some indexes will satisfy criteria or support table joins, while other indexes will fulfill aggregations or ordering. In a subsequent query, those same indexes may get used for different purpose.
An Omnidex index can also be created with multiple columns. These indexes are valuable for optimizing SQL statements with multiple columns in the GROUP BY clause, ORDER BY clauses or the aggregation functions (COUNT, SUM, MIN, MAX and AVERAGE).
QuickText indexes are specialty indexes designed for small blocks of text, such as names, addresses, descriptions and general text. QuickText indexes parse the contents of the column so that each word is indexed separately. Queries can then mention just a few of the words in that column rather than having to mention the whole field. For example, a field containing the name “John Q. Doe” can be located by just asking for “John” or “John Doe”. QuickText indexes provide basic text searches at the same speed as Omnidex indexes.
FullText indexes are specialty indexes designed for large blocks of text, such as abstracts, articles and text documents. FullText indexes parse the contents of the column so that each word is indexed separately. FullText indexes also track the position of each word in the field to aid in providing relevancy scores. Queries can also use special syntax to require that one word be a certain distance from other word. FullText indexes necessarily have more overhead than QuickText indexes.
Omnidex also supports Custom Indexes which allow more granular control over the options used on an index. Custom indexes can also be used to index a portion of a column or a concatenation of multiple columns.
See also: