15.00 - Built-In Support for Fully-Normalized Databases - Teradata Database

Teradata Database Design

Product
Teradata Database
Release Number
15.00
Content Type
User Guide
Publication ID
B035-1094-015K
Language
English (United States)

Built‑In Support for Fully-Normalized Databases

Teradata Database parallel technology is optimized to perform relational database management tasks in a normalized physical environment that other commercially available relational DBMSs cannot match operating with a denormalized physical schema. Among the special performance advantages built into Teradata Database are the following:

  • Star join and other join optimizations (see SQL Request and Transaction Processing)
  • Full-table scan optimization (see SQL Request and Transaction Processing)
  • Specially designed index types to optimize parallel query operations (see Chapter 8: “Teradata Database Indexes and Partitioning,” Chapter 9: “Primary Indexes and NoPI Objects,” Chapter 10: “Secondary Indexes,” and Chapter 11: “Join and Hash Indexes”)
  • Row and column partitioning
  • A full complement of SQL aggregate and ordered analytical functions (see SQL Functions, Operators, Expressions, and Predicates)
  • A sophisticated parallel‑aware SQL query optimizer (see SQL Request and Transaction Processing)
  • Teradata was founded at the time relational technology had initially emerged as the dominant force in database management, and the company has, from the outset, focused solely on delivering an enterprise‑wide decision support system with high performance, direct-access, application‑neutral, and extensible facilities that present a single version of the truth.