March 2020

These features and Databricks platform improvements were released in March 2020.

Note

Releases are staged. Your Databricks account may not be updated until up to a week after the initial release date.

Managed MLflow Model Registry collaborative hub available (Public Preview)

March 26-31, 2020: Version 3.16

MLflow Model Registry is a collaborative hub where teams can share machine learning models, work together from experimentation to staging and production, integrate models with approval and governance workflows, and track model deployments. The managed MLflow Model Registry is now available for Public Preview to all Databricks customers. See Manage the Lifecycle of MLflow Models in MLflow Model Registry.

Load data from hundreds of data sources into Delta Lake using Stitch

March 26-31, 2020: Version 3.16

The new Stitch partner integration helps you consolidate all of your business data from different databases and SaaS applications (Salesforce, Hubspot, Marketo, and so on) into Delta Lake. Stitch joins Fivetran, Qlik, Infoworks, StreamSets, and Syncsort in our Partner Integrations gallery. See Stitch integration.

Databricks Runtime 7.0 (Beta) previews Apache Spark 3.0

March 22, 2020

Databricks Runtime 7.0 (Beta) provides a preview of Apache Spark 3.0, with Scala 2.12. Please try it out using non-production workloads and give us your feedback.

For more information, see the complete Databricks Runtime 7.0 (Beta) release notes.

Databricks Runtime 6.5 ML (Beta)

March 20, 2020

Databricks Runtime 6.5 ML (Beta) brings the following library upgrade:

  • MLflow upgraded from 1.5.0 to 1.7.0

For more information, see the complete Databricks Runtime 6.5 ML release notes.

Databricks Runtime 6.5 (Beta)

March 20, 2020

Databricks Runtime 6.5 (Beta) brings many library upgrades and new features, including:

  • Operation metrics for all writes, updates, and deletes on a Delta table now appear in table history
  • You can rate-limit the data processed in Delta Lake streaming micro-batches
  • Snowflake connector is updated to 2.5.9

For more information, see the complete Databricks Runtime 6.5 release notes.

Optimized autoscaling on all-purpose clusters running Databricks Runtime 6.4 and above

March 10-17, 2020: Version 3.15

For clusters running Databricks Runtime 6.4 and above, optimized autoscaling is now used by default on all-purpose clusters if your workspace is on the Premium plan (or, for customers who subscribed to Databricks before March 3, 2020, the Operational Security package).

Single-sign-on (SSO) now available on all pricing plans

March 10-17, 2020: Version 3.15

Single-sign-on (SSO) is now available on all pricing plans.

Develop and test Shiny applications inside RStudio Server

March 10-17, 2020: Version 3.15

You can now develop and test Shiny applications inside RStudio Server hosted in Databricks. See Shiny on Databricks.

Change the default language of a notebook

March 10-17, 2020: Version 3.15

You can now change the default language of a notebook.

Databricks to add anonymized usage analytics

March 4-17, 2020

Starting March 4, Databricks will enable Heap analytics to help us improve our products and services. For details, see Manage third-party analytics tools.

Databricks Connect now supports Databricks Runtime 6.4

March 6, 2020

Databricks Connect now supports Databricks Runtime 6.4.

Databricks Connect now supports Databricks Runtime 6.3

March 3, 2020

Databricks Connect now supports Databricks Runtime 6.3.