Limit Load Slots | Performance | Teradata Data Mover - About Limiting the Load Slots Used on the Target System - Teradata Data Mover

Teradata® Data Mover User Guide

Product
Teradata Data Mover
Release Number
16.20
Published
November 2021
Language
English (United States)
Last Update
2021-11-04
dita:mapPath
wph1512683331242.ditamap
dita:ditavalPath
4101_UG_dm_1620.ditaval
dita:id
B035-4101
lifecycle
previous
Product Category
Analytical Ecosystem
In the target system load slot limit feature, Data Mover limits the number of job tasks running at the same time based on their use of the Teradata PT API LOAD and UPDATE operators. These API operators, which load data to the target system, restrict the number of active jobs running at a specific time against a specific Teradata Database.
Some limits external to Data Mover can override the Data Mover sessions:
  • TASM - If Teradata Active System Management (TASM) session rules are enabled on a particular source or target system, the TASM rules may override the Data Mover sessions.
  • DBSControl - If you limit the MaxLoadTask and MaxLoadAWT fields of the DBSControl utility, this could override the Data Mover sessions.
  • The load slot restriction affects Aster to Teradata jobs. Aster to Teradata jobs use the SQL-MR function load_to_teradata to move data, which internally uses the Teradata PT API Load operator. Aster to Teradata jobs share the load slots with the Teradata to Teradata jobs.
  • The load slot restriction affects jobs that are moving tables from Hadoop to Teradata when the transfer method is internal_fastload.

You can easily exceed operator limits for a specific target Teradata Database because each Data Mover job can use several instances of these operators, and multiple Data Mover jobs can run at the same time. Data Mover jobs can subsequently fail or hang as attempts to create new instances of the operators are rejected or put on hold by the Teradata Database.

To avoid such failures, Data Mover limits the number of concurrent job tasks, based on the operator being used by the tasks. The limit is enforced independently for each target Teradata Database system. For example, if a job has 10 tasks that use the restricted operators, and the target system load slot limit is 5, Data Mover allows up to 5 of the 10 tasks from that job to run concurrently against the target Teradata Database system. If 2 jobs that each has 10 tasks using the restricted operators are run against the same target Teradata Database system at the same time, Data Mover allows only 5 of the combined 20 tasks from both jobs to run concurrently. If 2 jobs are run against different target Teradata Database systems, and the target system load slot limit is 5 for both systems, Data Mover allows 5 tasks from each job to run at the same time. Data Mover jobs and tasks that use non-restricted operators are not affected by this feature.

The target load slot feature does not affect the number of Data Mover jobs that can run at a specific time or determine which utility Data Mover chooses when creating new jobs.

The target.system.load.slots configuration property, which can be modified through the list_configuration and save_configuration commands, controls the target system load slot limit. Use the target.system.load.slots property to set the following:
  • A default load slot limit for all target Teradata Database systems
  • Different limits for specific target Teradata Database systems
For the target system load slot limit to work correctly, all Data Mover jobs run against a specific target Teradata Database system must use the same name or IP address for the same target system. Otherwise, Data Mover assumes that the jobs are using different target systems and limits their tasks separately, rather than collectively.