15.00 - Hash Index Storage - Teradata Database

Teradata Database Design

Teradata Database
User Guide

Hash Index Storage

For the most part, hash index storage is identical to base table storage except that hash index rows are, in most cases, row compressed automatically. Hash index rows are hashed on their primary index (which is always defined as nonunique). Hash index tables can be indexed explicitly, and their indexes are stored just like nonunique primary indexes for any other base table. Unlike join indexes, hash index definitions do not permit you to specify secondary indexes.

The major difference in storage between hash indexes and base tables is the manner in which the repeated field values of a hash index are stored. Repeated field value storage is too complex to describe in this summary. See Chapter 11: “Join and Hash Indexes” under “Join Index Storage” on page 598 for details about how repeated field values are stored. A hash index can be stored either in hash-order or in value-order, depending on its intended use.

Related Topic

For additional usage information and design tips about hash indexes, see Chapter 11: “Join and Hash Indexes”.