The TeradataAsterR package contains a set of "runner" functions that allow users to run open-source or custom R functions directly on the data inside the Aster Database. These runner functions are analogous to the apply() functions in open-source R, and can be used to execute any R functions defined in the current session over a virtual data frame. No SQL commands are required when using the runner functions.
These runner functions automatically handle all data distribution and parallel processing, and also take care of many other tasks that users must pay attention to when using the Aster Stream functionality, such as installing and uninstalling scripts and managing input and output.
This table briefly describes the runner functions.
Function | Description |
---|---|
ta.eval() | Calculates an R function without any input from an Aster data frame. |
aa.apply() | Applies an R function on a given Aster data frame.
ta.apply() is deprecated.
|
ta.aggregateApply() | Applies a given R function on tables of an Aster data frame. |
ta.colApply() | Applies a given R function on columns of an Aster data frame. |
ta.rowApply() | Applies a given R function on rows of an Aster data frame. |
aa.tapply() | Applies an R function on a partitioned Aster data frame.
ta.tapply() is deprecated.
|
ta.source() | Runs scripts and functions written in other languages, including R, Python, Ruby, Perl, and C#. The script must be installed on the cluster before execution. If the script uses functions from the Aster R package, then the TeradataAsterR package must be installed on the Aster cluster. Refer to Appendix C Installing TeradataAsterR Package on the Aster Cluster for information on how to install the TeradataAsterR package on the Aster cluster. |
ta.by() | Partition-oriented shortcut for the ta.tapply() function. |
ta.partitionApply() | Partition-oriented shortcut for the ta.tapply() function. |