Edit

Share via

Fabric Runtime 1.3 (GA)

  • Article

Fabric runtime offers a seamless integration with Azure. It provides a sophisticated environment for both data engineering and data science projects that use Apache Spark. This article provides an overview of the essential features and components of Fabric Runtime 1.3, the newest runtime for big data computations.

Microsoft Fabric Runtime 1.3 is the latest GA runtime version and incorporates the following components and upgrades designed to enhance your data processing capabilities:

  • Apache Spark 3.5
  • Operating System: Mariner 2.0
  • Java: 11
  • Scala: 2.12.17
  • Python: 3.11
  • Delta Lake: 3.2
  • R: 4.4.1

Tip

For up-to-date information, a detailed list of changes, and specific release notes for Fabric runtimes, check and subscribe Spark Runtimes Releases and Updates.

Use the following instructions to integrate runtime 1.3 into your workspace and use its new features:

  1. Navigate to the Workspace settings tab within your Fabric workspace.
  2. Go to Data Engineering/Science tab and select Spark Settings.
  3. Select the Environment tab.
  4. Under the Runtime Versions expand the dropdown.
  5. Select 1.3 (Spark 3.5, Delta 3.2) and save your changes. This action sets 1.3 as the default runtime for your workspace.

Screenshot showing where to select runtime version.

You can now start working with the newest improvements and functionalities introduced in Fabric runtime 1.3 (Spark 3.5 and Delta Lake 3.2).

Key highlights

Apache Spark 3.5

Apache Spark 3.5.0 is the sixth version in the 3.x series. This version is a product of extensive collaboration within the open-source community, addressing more than 1,300 issues as recorded in Jira.

In this version, there's an upgrade in compatibility for structured streaming. Additionally, this release broadens the functionality within PySpark and SQL. It adds features such as the SQL identifier clause, named arguments in SQL function calls, and the inclusion of SQL functions for HyperLogLog approximate aggregations. Other new capabilities also include Python user-defined table functions, the simplification of distributed training via DeepSpeed, and new structured streaming capabilities like watermark propagation and the dropDuplicatesWithinWatermark operation.

You can check the full list and detailed changes here: https://spark.apache.org/releases/spark-release-3-5-0.html.

Delta Spark

Delta Lake 3.2 marks a collective commitment to making Delta Lake interoperable across formats, easier to work with, and more performant. Delta Spark 3.2 is built on top of Apache Spark™ 3.5. The Delta Spark maven artifact has been renamed from delta-core to delta-spark.

You can check the full list and detailed changes here: https://docs.delta.io/3.2.0/index.html.

Related content