In this Article
Overview
Offline Assets allow you to catalog and manage data assets that are not connected to a live data source. These assets are created and maintained through manual imports instead of automated scans.
Offline Assets follow the same logical hierarchy as online assets and support metadata management, relationships, classifications, and lineage. However, because there is no live connection, features such as automated scans, profiling, and Data Quality metrics are not available.
This capability is commonly used to document legacy systems, external platforms, third-party datasets, mainframes, flat files, or any environment where direct connectivity is not possible or not permitted.
Offline Data Source Types
Offline assets are created using Offline Data Source Types. A source type defines how assets are structured, what hierarchy is used, and how imported assets are organized within the catalog.
From Asset Management, you can create a new offline source type by providing a name, description, data hierarchy, and an optional icon. The icon helps visually differentiate the source type in the user interface.
Once created, the offline source type appears alongside other source types and can be used to create offline data sources. Source types can be enabled or disabled. When a source type is disabled, it cannot be selected for creating new sources, but existing sources remain unaffected.
Supported Data Hierarchies
Offline data source types support multiple hierarchies to accommodate different data models.
The supported hierarchies include:
Flat Table: Database Table -> Database Column
Schema Based: Database Schema -> Database Table -> Database Column
Database without Schema: Database -> Database Table -> Database Column
Full Database: Database -> Database Schema -> Database Table -> Database Column
The selected hierarchy determines how assets are imported and displayed in the catalog. This selection is fixed at the time of source type creation and cannot be changed later. Choosing the correct hierarchy upfront is important to ensure consistency and accurate asset organization.
Creating an Offline Data Source
After creating an offline source type, an offline data source can be created from the Data Sources section.
Offline data sources are clearly marked as offline. Automated scanning, scheduling, and scan-related configurations are disabled for these sources. All assets under an offline source must be created or updated through imports.
Once the source is created, the source overview page is displayed. Tabs related to scanning, scan history, and logs are not available, while metadata-related tabs remain accessible.
Importing Offline Assets
Assets for offline sources are imported using the standard import workflow available in DvSum.
Supported asset types include databases, schemas, tables, and columns. Imports can be performed incrementally, allowing assets to be added or updated over time as information becomes available.
During import, you can choose to:
Add new records only
Update existing records only
Add and update records in a single import
This flexibility allows offline catalogs to evolve without requiring a complete reimport each time.
Identification Fields and Matching
Identification fields are used to uniquely identify assets during the import process. These fields are based on the selected hierarchy and may include source name, database name, schema name, table name, and column name.
When updating existing assets, records can be matched using asset ID, full name, or identification fields. Using identification fields is recommended, as it ensures consistent matching even when naming conventions or delimiters vary.
Proper use of identification fields helps prevent duplicate asset creation and ensures accurate updates.
Import Options and Behavior
Several import options control how metadata is handled during the import process.
These options include skipping empty values, overwriting or appending tags, overwriting or appending relationships, and automatically creating parent assets if they do not already exist.
More information about Data import here.
Automatically creating parent assets is particularly useful when importing tables or columns that reference databases or schemas that are not yet present in the catalog.
After the import completes, a summary email is sent with details about the import results. This includes the total number of records processed, the number of records added or updated, any skipped or invalid records, and a downloadable summary file for review.
Viewing Offline Assets
Offline assets appear in the Asset Dictionary and asset detail pages alongside online assets. From a user perspective, offline assets can be browsed, searched, and filtered in the same way.
Supported capabilities include viewing and editing metadata, managing relationships, applying tags and classifications, and adding manual lineage.
The following capabilities are not available for offline assets:
Automated scans
Data profiling
Sample data grids
Data Quality metrics
This distinction ensures clarity between manually maintained assets and those backed by live connections.
Managing Offline Sources and Assets
Offline source types and offline data sources can be edited, enabled, disabled, or deleted as needed.
Deleting an offline data source removes all assets created under that source. Deleting an offline source type removes all associated sources and their assets. In both cases, a confirmation prompt is displayed to prevent accidental deletion.
These controls allow administrators to manage offline catalogs throughout their lifecycle.
Additional Notes
Classification rules and access controls apply to offline assets in the same way as online assets. Manual lineage can be added through asset editing to document upstream or downstream relationships.
FAQ's
What is the recommended limit for asset import?
The recommended limit is up to 10,000 entries per import, which is applicable to both single-asset and multi-asset uploads. Exceeding this limit may lead to processing issues.
How long does the import process take?
The import processing time may vary depending on the file structure, number of fields, and total record count.For files within the recommended limit and standard structure, processing typically completes and a notification email is generated within 15 minutes.
What should be done if no email notification is received?
If no email is received after 15 minutes, the request may have timed out due to high record volume or file complexity. In such cases, reduce the number of records per file and re-import in smaller batches to ensure successful processing and email notification.
0 Comments