Bulk Load - Teradata Unity

Teradata Unity User Guide

Product
Teradata Unity
Release Number
16.10
Published
January 2018
Language
English (United States)
Last Update
2018-03-29
dita:mapPath
anz1496162519675.ditamap
dita:ditavalPath
ft:empty
dita:id
B035-2520
lifecycle
previous
Product Category
Analytical Ecosystem

When a bulk load client submits data for loading, Unity controls the data load operation on each target Teradata system. Bulk load operations are run independently across multiple Teradata systems, and automatic recovery and restarts occur if data loading is interrupted. Systems that fail to load data are automatically reloaded when the system is restored. For consistency, Unity compares the results of the data load operation across Teradata systems.

Unity supports different sized Teradata systems that load at different rates and minimizes impact on the bulk load client when a system is offline. Unity manages an internal disk array to accommodate the disk space requirements for landing data before loading the data to Teradata systems. Data is landed to disk for any of the following reasons:
  • Data cannot be loaded to one or more of the target systems due to target, error, or restart tables in the interrupted state or an unavailable target system.
  • A data checkpoint did not occur on all systems, and there is insufficient memory to store the data in memory.
  • A data checkpoint is successful on one system, but not on the others.

Unity deletes landed data stored on the disk array after the data successfully loads to the Teradata systems, or the begin loading record of the bulk load job is overwritten in the recovery log. Management of the disk space by Unity reduces the total amount of disk space required for the load files.