Indexes are used to improve data retrieval performance. Traditional indexes use a B-tree index strategy to point to the data records. That strategy is valuable only if many unique data values are used to filter down to a very small set of records, as with columns of order numbers or customer names, as you would encounter in a transaction processing system.
Sybase IQ indexes actually represent and store the data so that the data can be used for processing queries. This strategy is designed for the data warehousing environment, in which queries typically examine enormous numbers of records, often with relatively few unique values, and in which aggregate results are commonly required.
