Omnidex Indexing

Overview → Indexes → Strategies → Advanced → Installation → Maintenance

The basic strategy for creating Omnidex indexes is to index all of the columns that are used in criteria, table joins, aggregations and ordering. This is a different approach than is taken with relational databases. Database administrators are usually trained to identify selected columns that are most frequently use and create indexes only on those tables. Their expectation is that the relational database will choose an index to access a table and will process the rest of the statement by directly evaluating the data.

Omnidex approaches indexing differently. Omnidex indexes all of the columns and then coordinates searches across all of the indexes to fulfill the different aspects of a query. Some indexes will be use to satisfy table joins or criteria. Other indexes will be used to fulfill aggregations or ordering. Most database administrators are surprised to learn that complex SQL statements that join many tables and contain intricate criteria can often be fulfilled without ever accessing the underlying data. This approach allows Omnidex to satisfy queries very quickly. It also reduces the load on the servers since accessing the data is a common cause of performance problems.

1. Indexes for Table Joins
2. Indexes for Criteria
3. Indexes for Aggregations
4. Index for Ordering
5. Other Uses for Indexes

The next section on Omnidex Indexing is Basic Indexing Strategies.