Data acquisition and transformation

Applies to: TBM Studio 12.0 and later

About the table transform pipeline

To modify data in a table, you can add or modify steps in the transform pipeline. In the pipeline, you can:
  • See all of transform steps in the order they are executed.
  • Add, edit, and delete transform steps to modify the output.
  • Drag to reorder individual steps.
  • Browse videos in the Apptio community (link requires TBM Connect credentials).
Note: Editable tables are not table transforms and are not covered in this article. For more information, see Editable tables: Accommodating user input.

Table transform pipeline steps

When creating a new table, the following pipeline steps will automatically populate regardless of the source of the data:
  • Source: Determines the type of data on which to base a table.
  • Table: Previews the data table created by the transform.

Depending on the Source option selected, other pipeline steps will automatically be added:

File Upload source options:
  • Upload: Allows the user to select the file to upload.
  • Import: Displays the settings used to define how the system should import a file.
Existing Table source option:
  • Existing Table: Defines the table transform on which the table should be based.
You can add the following pipeline steps to any table:
  • Append: Adds the rows from one table to the rows in another table. See Append data.
  • Assign Rows: Automatically added when using Map Columns to map to master data sets with machine learning capabilities and can be re-added to valid tables through the Add Step process. Uses machine learning and user created rules to map to ATUM concepts. See Assign cost pools with machine learning.
  • Date Partition Use dates from the file, either in rows or columns, to automatically distribute the data across time periods. See Partition data by date.
  • Filter: Remove specific rows based on values in one or more columns.
  • Flatten Hierarchy: Provides the greatest flexibility when using slicers. This pipeline step adds one column for each level in the hierarchy to a table. You can then create slicers for each of the columns. See Flatten a data hierarchy.
  • Formulas: Add new formula columns based on data, change column types (for example, change a number to a label), or override existing values. See Add and edit columns and Formulas and functions.
  • Group: Group the table by the values in one or more text columns.
  • Hide and Rename: Hide or rename columns in the data.
  • Join: Combine data from two or more tables into the same rows based on common values in one or more columns. See Join data.
  • Map Columns: Conforms the table's output to match the ATUM schema and adds the rows to the master data set. See Map Columns.
  • Model: Allows the table to be used as an object in a model where users can define drivers and allocate values. See About models.
  • Pivot: Converts rows to columns. See Pivot data.
  • Remove Duplicates: Filters out rows in a table that have the same value in a column or set of columns.
  • Row-Level Security: Allows users to filter tables and reports based on the current user. See Apply Row-Level Security.
  • Unpivot: Convert columns to rows.