Archives of the TeradataForum

Message Posted: Thu, 10 Nov 2005 @ 21:06:10 GMT

	<Prev	Next>		<<First	<Prev	Next>	Last>>

This question could be considered ambiguous as it could cover a great deal of ground. That said, I will take a stab at it ....Statistics aid (any) DBMS engine in making informed decisions about how best to develop and execute the query plan. In the absence of up-to-date statistics, the DBMS engine has to make assumptions, which can sometimes be poor.

The fact that in Teradata, data is distributed across amps makes for the ability of Teradata to efficiently partition up certain operations and perform them in parallel. In short this means there is the potential that Teradata might arrive at the answer set faster than running a similar transaction in a DBMS environment without its underlying parallel architecture.

Cornelius Meade III
EDS - Consulting Services/BI/CRM

	<Prev	Next>		<<First	<Prev	Next>	Last>>