Frequently asked questions (FAQ) about Microsoft Purview data governance solutions

Many organizations lack a holistic understanding of their data. It's challenging to understand what data exists, where data is located, and how to find and access relevant data. Data lacks context such as lineage, classification, and comprehensive metadata, making it difficult for business users to search for the right data and use that data appropriately. As a result, only a small fraction of collected data is used to inform business decisions. Finally, identifying data security issues and protecting sensitive data is inconsistent. It requires ongoing time and effort, especially while maintaining data agility.

Microsoft Purview data governance solutions help customers gain deep knowledge of all their data while maintaining control over its use. With our tools, organizations discover and curate data. They gain insights into their data estate and centrally govern access to data.

This FAQ answers common questions that customers and field teams often ask about Microsoft Purview data governance solutions: Microsoft Purview Data Map and Microsoft Purview Data Catalog.

If you have an issue with Microsoft Purview Data Map or Data Catalog, reach out to support.

If you have feedback about the documentation, reach out using the feedback button at the bottom of every documentation page.

There are many sources available, and we're adding more all the time. For a full list of sources and all their available capabilities, see our supported data sources article.

Currently, the Microsoft Purview Data Map supports scanning GZIP files that contain a single CSV. For more information, see supported file types for scanning.

It depends. The amount of load on a database will be dependent on the number of assets being scanned, and complexity of the table schema (like number of columns, etc.). The Microsoft Purview Data Map only samples a subset of the data to determine classification. For sampling related details, see the supported data sources documentation. It's recommended to schedule scans outside of normal business hours. For scanning best practices, see the scanning best practices documentation.

No. Number of rows scanned is determined by the data source being scanned. For more information, see our sampling documentation.

Our lineage offering updates often. For a full list of systems, see our lineage user guide.

Microsoft Purview (formerly Azure Purview) originally began as ADC Gen 2 but has since broadened in scope. It now embraces the advanced catalog capabilities of ADC Gen 2 combined with the data classification, labeling, and compliance policy enforcement capabilities of Azure Information Protection. Microsoft Purview now houses compliance and data governance tools to manage your full data estate. For more information, see our overview documentation.

Microsoft Purview is the focus of all product innovation in the catalog solution space for Microsoft. ADC Gen 1 is being retired in 2024.

Use the Microsoft Purview APIs to extract from ADC Gen 1 and ingest into Microsoft Purview. For the glossary, we support bulk tools based on CSV.

Data encryption is done at the data source level. Microsoft Purview stores only the metadata. It doesn't preview data.

A glossary uses a naming convention followed by non-technical/business users of the data, also known as data consumers. These types of people are business analysts or data scientists who use Microsoft Purview to search for certain types of data, based on business usage. For instance, supply chain analysts might need to search for the terms SKU types and shipment details. They search the glossary for these terms to find relevant data. Classification is a tag applied to a data asset at the table, column, or file level that identifies what data exists in the asset. Classification can be applied automatically or manually, based on the type of data found. Typically, you use classification tags to identify whether an asset contains sensitive data, and what type of sensitive data that might be.

There's a Azure-managed scanning infrastructure. For most Azure/AWS resources that we support, you don't need to deploy a scanning infrastructure.

Yes, PowerShell commands, a .NET script, a Python script, an ARM template, and a Bicep template are all available.

Microsoft Purview is compatible with Atlas API. If you're migrating from Atlas, it's recommended to scan your data sources first using Microsoft Purview. Once the assets are available in your account, you can use similar Atlas APIs to integrate such as updating assets or adding custom lineage. Microsoft Purview modifies the Search API to use Azure Search so you should be able to use Advance Search.

No, currently in order to scan another tenant's data source, you need to create a separate Microsoft Purview account in that tenant.

Yes, Microsoft Purview supports column level lineage.

Yes, Microsoft Purview supports Soft Delete for Azure subscription status management perspective. Microsoft Purview can read subscription states (disabled/warned etc.) and put the account in soft-delete state until the account is restored/deleted. All the data plane API calls will be blocked when the account is in soft delete state and only GET/DELETE control plane API calls will be allowed. You can find additional information in Azure subscription states page Azure Subscription Status

No. For more information about the SHIR, see our self-hosted integration runtime documentation.

You can follow our documentation for examples: using the REST API to access the data catalog, and using the REST API to manage access control. We also have a tutorial for using the Python SDK, and Atlas 2.2 APIs.

For network, security, and firewall configuration recommendations, see our security best practices documentation.