Traditionally, high performance database servers have relied heavily on the presence of a knowledgeable, dedicated, database administrator. This person spent a great deal of time adjusting data storage and performance controls of all kinds to achieve good database performance. These controls often required continuing adjustment as the data in the database changed.

SQL Anywhere learns and adjusts as the database grows and changes. Each query betters its knowledge of the data distribution in the database. SQL Anywhere automatically stores and uses this information to optimize future queries.

Every query both contributes to this internal knowledge and benefits from it. Every user can benefit from knowledge that SQL Anywhere has gained through executing another user's query.

Statistics-gathering mechanisms are thus an integral part of the database server, and require no external mechanism. Should you find an occasion where it would help, you can provide the database server with index hints. These hints ensure that certain indexes are used during optimization, thereby overriding the decisions made by the optimizer based on selectivity estimations. If you encode these into a trigger or procedure, you then assume responsibility for updating the hints whenever appropriate. See Updating column statistics, and Working with indexes.