MATRIX_SPEC Element | Teradata Vantage - Matrix Specifications - Teradata Vantage

Teradata® VantageCloud Lake
Deployment
VantageCloud
Edition
Lake
Product
Teradata Vantage
Published
January 2023
ft:locale
en-US
ft:lastEdition
2024-12-11
dita:mapPath
phg1621910019905.ditamap
dita:ditavalPath
pny1626732985837.ditaval
dita:id
phg1621910019905
A matrix is a two-dimensional array that has rows and columns. A matrix is identified by its MATRIX_ID fields, and is indexed by its ROW_AXIS and COLUMN_AXIS fields. A matrix can be a one of the following types:
  • Row-major matrix: Each row is processed as a wavelet that is grouped by its MATRIX_ID and ROW_AXIS, and ordered by its COLUMN_AXIS.
  • Column-major matrix: Each column is processes as a wavelet that is grouped by its MATRIX_ID and COLUMN_AXIS, and ordered by its ROW_AXIS.

A MATRIX_SPEC is used to reference any Teradata accessible table or a view, a derived-table, or a particular layer residing within an ART. When used to reference a layer of an ART, the LAYER parameter must be included, except the case in which the layer is ARTPRIMARY, in that case the LAYER parameter can be omitted.

The identifier and PAYLOAD fields can be renamed using AS clauses. However, this may not affect the output field names for the PAYLOAD fields because the PAYLOAD fields are input fields. The output payload field name is only affected if the function uses the input payload field name as the output field name or as part of the output field name.

The identifier columns cannot reference a complex data type or LOB column. Dot notation is allowed.