Dataset refers to a group of tables whose inter-relationship is defined by specifying a join condition between the various tables. It is a basic building block to create a query and execute on a data warehouse for a large number of functions and to generate reports.
Dataset function within the Infrastructure system facilitates you to create Datasets and specify rules that fine-tune the information for querying, reporting, and analysis. Datasets enhances query time by pre-defining the names of tables required for an operation (such as aggregation), and also provides the ability to optimize the execution of multiple queries on the same table set. For more information, refer to Scenario to Understand the Dataset Functionality.
The roles mapped to Datasets are Dataset Access, Dataset Advanced, Dataset Authorize, Dataset Phantom, Dataset Read Only, and Dataset Write.
The Datasets Summary window
displays the list of pre-defined Datasets with their Code, Short Description,
and Long Description. You can add, view, edit, copy, and delete the required
Dataset. You can also make use of Search and Pagination options to search
for a specific dataset based on the Code, Short Description, and Authorization
status. By clicking the Column header names, you can sort the column names
in ascending or descending order. Click if you want
to retain your user preferences so that when you login next time, the
column names will be sorted in the same way. To reset the user preferences
click
.
You can create Dataset by defining the Dataset Details, Entities, and Dataset Definition. You need to have DATADD function role mapped to create Datasets.
To create Dataset in the Datasets screen:
From
the Dataset Summary window,
click Add
from the Datasets toolbar.
The Dataset Details window
is displayed.
Enter the details in the Dataset Details section as tabulated.
Field |
Description |
Fields marked in red asterisk (*) are mandatory. |
|
Code |
Enter a distinct code to identify the Dataset. Ensure that the code is alphanumeric with a maximum of 8 characters in length and there are no special characters except underscore "_". Note the following:
|
Short Description |
Enter a Short Description based on the defined code. Ensure that the description is of a maximum of 8 characters in length and does not contain any special characters except underscore "_". |
Long Description |
Enter the Long Description if you are creating subject-oriented Dataset to help users for whom the Dataset is being created or other details about the type/subject. Ensure that the description is of a maximum of 100 characters in length. |
From the Entities pane, you can perform the following:
To select all entities, click button.
To remove an entity, select the entity from the Selected
Values grid and click button.
To remove all entities from the Selected Values grid, click
button.
Additionally, you can rearrange the entities by moving an
entity up (), down (
), to the top (
),
or to the bottom (
).
Specify the required table-join condition in the Dataset Definition grid as tabulated.
Field |
Description |
ANSI Join |
The ANSI Join condition defines which set of data have been joined along with the type of join condition. It also describes the exact operations to be performed while joining the Datasets. In ANSI join, the join logic is clearly separated from the filtering criteria. |
Join/Filter Condition |
The Join/Filter Condition facilitates the objective of creating Datasets. Datasets with linked tables using the join conditions help in reducing the query time. There are two ways of defining the join condition:
|
Date Filter |
The Date Filter condition enables you to cascade the cubes that are using the Dataset with the defined Date Filter. |
Order By |
The Order By condition enables you to sort the dimension data in order. The order of the Dimension nodes will be maintained only for Business Intelligence enabled hierarchies. The Order By condition is specific to the Essbase database. |
Enter the required expression or click button to define an expression using the Expression window. For more information, see Expression Builder .
Click Preview. The Data of Preview Dataset Query window is displayed. This window displays an error message if the Query execution fails. Up to 400 records of data is displayed in the Summary Grid.
Click Show Query to view the query.
Enter the values for $MISDATE and $RUNSK parameters.
Click Save and save the Dataset definition details.
You can view individual Dataset details at any given point. You need to have DATVIW function role mapped to view the Datasets.
To view the existing Dataset definition details in the Datasets screen:
From the Dataset Summary window, select the checkbox adjacent to the required Dataset code.
Click View from the Datasets
toolbar.
The View Datasets screen displays the details of the selected Dataset definition. The User Info grid at the bottom of the screen displays the metadata information about the Dataset definition created along with the option to add comments.
You can update the existing Dataset definition details except for the Code and Short Description. You need to have DATMOD function role mapped to modify the Datasets.
To updated the required Dataset details in the Datasets screen:
From the Dataset Summary window, select the checkbox adjacent to the required Dataset code.
Click Edit
from the Datasets
toolbar. The Edit Datasets
screen is displayed.
Update the required details. For more information, see Create Dataset.
Click Save and update the changes.
You can copy the existing Dataset details to quickly create a new Dataset. You can later modify the Dataset Code or Short Description, add/remove tables, and also define the join conditions. You need to have DATADD function role mapped to copy the Dataset definitions. To copy an existing Dataset definition in the Datasets screen:
From the Dataset Summary window, select the checkbox adjacent to the required Dataset code.
Click Copy
from the Datasets toolbar.
The Dataset definition details are copied and a confirmation message is displayed.
You can remove the Dataset definitions which are created by you and which are no longer required in the system by deleting from the Datasets screen. You need to have DATDEL function role mapped to delete a Dataset. Delete function permanently removes the Dataset details from the database. Ensure that you have verified the details as indicated below:
A Dataset definition marked for deletion is not accessible for other users.
Every delete action has to be Authorized/Rejected by the authorizer.
On Authorization, the Dataset details are removed.
On Rejection, the Dataset details are reverted back to authorized state.
You cannot update Dataset details before authorizing/rejecting the deletion.
An un-authorized Dataset definition can be deleted.
To delete an existing Dataset:
From the Dataset Summary window, select the checkbox adjacent to the required Dataset code.
Click Delete
from the Datasets toolbar.
A confirmation dialog is displayed.
Click OK. The Dataset details are marked for delete authorization.