Get Data Into Databricks - Delta Sharing - Cross Platform View Sharing

Summary

• Cross-Platform View Sharing: Delta Sharing enables securely sharing filtered rows and masked columns via views—no table replication needed.
• Dynamic Access Control: Use views and the current_recipient function to conditionally show or hide sensitive data based on recipient tier or attributes.
• Multi-Cloud Support: Share from Databricks to Spark, Pandas, Excel, Tableau, and other compatible platforms using the open Delta Sharing protocol.

Get Data Into Databricks - Delta Sharing - Cross Platform View Sharing

Try Databricks today: https://dbricks.co/3EAWLK6
Discover how Databricks enables dynamic view sharing with fine-grained access control. Data providers can securely share filtered rows and masked columns without replicating tables, using views tailored for each recipient. Recipients can query shared views seamlessly across platforms like AWS, Azure, or Google Cloud using popular tools such as Spark, Pandas, Excel, and Tableau, powered by the Delta Sharing protocol. Expand your data-sharing capabilities with flexible access control and cross-platform compatibility.

Chapters

• 0:00 Introduction to delta sharing
• 0:30 Cross-platform view sharing
• 1:12 Dynamic access control
• 2:29 Provider demo
• 3:44 Recipient setup
• 4:14 Consumer demo
• 4:34 Conclusion and resources

FAQs

How does view sharing save storage costs?

This video shows that view sharing eliminates the need to replicate entire tables for different recipients. Providers create views with appropriate filters, saving storage while simplifying data governance.

Can I hide sensitive columns from some recipients?

Yes. The video demonstrates using views with conditional logic and the current_recipient function to mask sensitive columns like customer contact information based on recipient tier (paid vs free).

How do I implement fine-grained access control for data?

This video demonstrates Delta Sharing for cross-platform data sharing on the Databricks Data and AI platform. Delta Sharing enables fine-grained access control at the table, row, and column levels while sharing data across organizational and platform boundaries. The video shows how to set permissions that restrict access to specific columns, apply row-level filters based on user attributes, and audit all data access.

How to share data applications securely across teams?

This video shows how Delta Sharing on the Databricks Data and AI platform enables secure data application sharing. Rather than copying data, Delta Sharing provides live, governed access through REST APIs. Applications from different teams can access shared data with consistent security policies, and all access is audited through Unity Catalog.

How do I control access to serverless resources?

This video demonstrates controlling access to serverless resources on the Databricks Data and AI platform through Delta Sharing and Unity Catalog. Access controls apply not just to data but to compute resources that process that data. The video shows how to restrict which teams can run queries, which models they can serve, and which APIs they can call through comprehensive permissions management.

Full transcript

0:00 Introduction to delta sharing

Hello everyone and welcome to data bricks I'm Kon a Solutions architect here at data bricks and today I'm excited to introduce you to a powerful feature of Delta sharing crossplot from view sharing in this video we'll explore how Delta sharing makes it easier than ever to share data with fine green Access Control to anyone regardless of whether they're inside or outside of your data bricks environment Delta sharing has revolutionized secure data sharing by equipping Enterprises with an open protocol to share live data across platforms clouds and regions with strong security and governance and to do so without replication saving you storage costs and ensuring data consistency now with crossplatform view sharing we're taking it a step further this enhancement enables you to easily and flexibly share data with fine grain Access Control to any recipient allows recipients to discover and query shared views using any connector whether inside or outside of data bricks and Implement Dynamic filtering for each recipient let's dive a little deeper view sharing allows data providers to organize and share data using views eliminating the need to replicate entire tables this not only saves on storage but also simplifies data governance it provides a way to share only the data that each recipient is entitled to see data providers can securely share selected rows and columns to recipients with Dynamic Access Control this is achieved through a powerful combination of views and the current recipient function here's a concrete example imagine a data provider who wants to share a multi-tenant customer table but needs to mass sensitive columns like customer contact based on the recipient's tier for example paid versus free with view sharing this becomes straightforward as you can see the provider creates a view called customer summary within this view the customer contact column is conditionally redacted using the case statement and the current recipient function if the recipient has a tier property set to paid they see the actual contact information otherwise it's redacted this view is then added to a share granting access to the designated recipients now let's see this in action with the live demo let's start with the provider side of our Delta sharing demo in this notebook we'll act as the data provider for this example we are a corporate travel and spend management platform a few of our partner Airlines would like to leverage our daily flight data to enhance their reward and recommendation programs we'll leverage Delta sharing to Grant access to these Partners without data duplication our table contain data for numerous Airlines and we would like to Grant access for each partner airline to only their respective flight data we can do this by creating a view on top of our R flight data a view allows us to share only specific columns and apply any necessary Transformations or filters as you can see in cell 15 we create a view called Flights protected that filters the data in our table by the carrier ID of the recipient in command 16 we see that we've already created a share above using SQL and we add our view to that share this makes the view available to all recipients with access to this share for recipients consuming data outside of data bricks they can directly connect to Shared data using Delta sharing connectors that support various systems like pandas Tableau Apache spark rust or others that support the open Delta sharing protocol alternatively upon creating an open recipient in data bricks a secure one-time activation URL is generated allowing the recipient to download a credential file that contains a Delta sharing endpoint address and a token finally let's step into the shoes of the recipient here the recipient is easily connecting to the shared data using python they can now query the shared View using pandas Spark or other processing tools of their choice as you can see crossplatform view sharing offers a powerful and flexible way to share data securely and efficiently we invite you to explore the resources linked below and start leveraging Delta sharing in your own projects try out the Delta sharing Airlines demo yourself in your own data bricks workspace watch our other video on creating a view in data bricks and sharing it with powerbi using Delta sharing or explore the comprehensive Delta sharing documentation to learn more about all its features and capabilities thank you for watching subscribe and stay tuned for more exciting updates and tutorials on the data bricks YouTube channel

Learn more about the Databricks Data and AI platform.