15.00 - Generic NUSI Read Process Flow - Teradata Database

Teradata Database Design

Teradata Database
User Guide

Generic NUSI Read Process Flow

The following flowchart illustrates the logic of a NUSI read. The graphic also illustrates a somewhat more concrete NUSI read example where the rowID value B2,1 read (where B2 is the row hash value and 1 is the uniqueness value for the row and assuming that the table being accessed is an nonpartitioned base table) is followed in turn through the master index, the cylinder index, and the data block to access the base table row set matching that value.

The following process describes the process flow for locating a row using a NUSI.

1 Read the first rowID from the subtable row.

In the graphic example, the value read is B2, 1.

2 Scan the Master Index for the required Cylinder Index.

3 Scan the Cylinder Index for the required data block.

4 Scan the rows in the data block until the row having the requested rowID is located.

5 Return the retrieved row set to the requestor.