Bulk Load Sessions - Teradata Unity

Teradata Unity User Guide

Product
Teradata Unity
Release Number
16.00
Published
March 2017
Language
English (United States)
Last Update
2018-03-29
dita:mapPath
ipb1476817227329.ditamap
dita:ditavalPath
ft:empty
dita:id
B035-2520
lifecycle
previous
Product Category
Analytical Ecosystem

A Bulk Load job includes control sessions and data sessions.

Processing Bulk Load jobs in Unity differs from standard SQL processing. For example, the Sequencer does not process Bulk Load data as for other SQL clients. The Sequencer and Dispatcher processes are only involved with Bulk Load when processing the control session. The Endpoint processes are the main components of the Bulk Load job and open the Load data sessions to the database.

Bulk load data is temporarily stored on Endpoint servers, under /data/processid . This data automatically deletes after the load data is completed on all systems. The load job fails if there is insufficient space available in this directory.

If you deactivate and reactivate all systems in the cluster, or individual tables that have stored Bulk Load data, delete the associated contents of the /data/processid directory.
Process Bulk Load Job Role
Sequencer Manages load operations by processing SQL executed by control sessions, coordinating session distribution, monitoring the load progress on each system, and initiating load recovery as needed. Does not assist in processing data for loading.
Dispatcher Negotiates the initial logon and number of sessions to open the job on each system.
Endpoint Manages receiving data, storing data, and loading data to systems.