Creating a dataset

You can define and manage domain-specific dataset by creating tables and defining relationships between them.

About this task

In the Dataset page, define custom domain-specific dataset by creating tables and establishing relationships between them. The Dataset page displays the following information about the created datasets.

Table 1. Dataset details
Fields	Description
Domain Name	Name of the domain object.
Table List	Associated dimension and fact tables
Created	Timestamp when the domain was create
Last Updated	Timestamp of the most recent update.
Actions	Edit Domain, Add New Table and Domain Metadata: These options are available for custom domain objects.

You can create new dataset and define relationship between data by using the following procedure.

Procedure

In the main menu, click Dataset > New Dataset to open the domain ownership form.

In the Domain ownership tab, enter the required details to create a new domain as follows.


Fields	Description
Domain name	Enter a unique identifier for the domain. For example, MY_CUSTOM_DOMAIN.
Owner email	Email address of the domain owner or responsible user.
Owner name	Full name of the domain owner for reference.

Click Submit.
On the Dataset page, domain names display in uppercase for standardization, though they are case-insensitive.

In the Table and domain metadata tab, define the database tables and domain metadata as follows.

Table 2. Database table fields
Fields	Description
Table name	Enter a unique name for the table. For example, `CUSTOM_DIM`, `CUSTOM_FACT`.
Type	Select one of the following types. Dimension: Stores descriptive data about domain entities. Fact: Stores relationship data between dimensions and includes a classification column.
Domain	Updates automatically as the Domain name from the Domain ownership tab.
Time driven	Select Yes if the table tracks time-based information. For example, `snapshot interval`.
Language	Select the language for metadata labels. This is used for localization of the user interface elements.
Table description	Enter a brief explanation about the purpose of the database table.

Table 3. Database table column-level fields
Fields	Description
Column name	A unique name for the column. For example, `app_id`,`scan_timestamp`.
Data type	Select one of the supported data types from the list.
Business name	A business-friendly label for the column. For example, `Application ID`.
Description	Description of the purpose of the column.
Unique	Select if the column values should be unique across rows.
Nullable	Sect if the column must allow NULL values.
Primary key	Select if the column must be a primary key.
Domain key	Use this field if the column should be marked as the domain key value for the current domain.
Display in UI	Select if you want the column to display in dashboards and filters.

Note: The system adds a classification column for the Fact tables automatically.

Click Submit.

In the Data integrity tab, create a relationship between the domain and the table as follows.

Once the dimension and fact tables are defined, you can establish relationships between them.

Table 4. Metadata details
Fields	Description
Domain	Select the relevant domain for which the relationship applies.
Source table	The table that contains the foreign key (typically a fact table).
Source column	The foreign key column in the source table
Target table	The primary key column in the dimension table.
Relationship type	Select a relationship type from the list.
Constraint name	Provide a unique and descriptive name for the relationship constraint.

The relationships that are defined enforce referential integrity and support accurate joins for analytics and reporting.

Click Append column + to add new columns.
Click Save integrity rule.

Results

A dataset is created with table and domain metadata details and relationships are defined between the database tables.

What to do next

Verify the created dataset to help ensure the custom domain metadata is correctly defined. For more information, see Verifying custom domain metadata and integrity.