2019 年 12 月December 2019

这些功能和 Azure Databricks 平台的改进已于 2019 年 12 月发布。These features and Azure Databricks platform improvements were released in December 2019.

备注

发布分阶段进行。Releases are staged. Azure Databricks 帐户可能要等到初始发布日期后的一周或更长时间才会更新。Your Azure Databricks account may not be updated until a week or more after the initial release date.

Databricks Connect 现在支持 Databricks Runtime 6.2Databricks Connect now supports Databricks Runtime 6.2

2019 年 12 月 17 日December 17, 2019

Databricks Connect 现在支持 Databricks Runtime 6.2。Databricks Connect now supports Databricks Runtime 6.2.

通过 Databricks 容器服务 (GA) 使用自己的容器映像配置群集Configure clusters with your own container image using Databricks Container Services (GA)

2019 年 12 月 16 日 :版本 3.7December 16, 2019: Version 3.7

Databricks 容器服务已在 Databricks Runtime 6.1 和 Azure Databricks 平台版本 3.7 中正式提供,通过该服务你可使用自己的容器映像来配置群集。Generally available in Databricks Runtime 6.1 and Azure Databricks platform version 3.7, Databricks Container Services lets you configure a cluster with your own container image. 你可以将复杂环境预打包在容器中,将其发布到常用容器注册表,如 ACR、ECR 或 Docker Hub,然后让 Azure Databricks 拉取映像以创建群集。You can pre-package complex environments within a container, publish it to a popular container registry such as ACR, ECR, or Docker Hub, and then have Azure Databricks pull the image to create a cluster. 一些示例用例包括:Some example use cases include:

  • 库自定义 - 你可以完全控制你要安装的系统库Library customization - you have full control over the system libraries you want installed
  • 黄金容器环境 - 你的 Docker 映像是锁定的环境,永远不会更改Golden container environment - your Docker image is a locked down environment that will never change
  • Docker CI/CD 集成 - 可以将 Azure Databricks 与 Docker CI/CD 管道集成Docker CI/CD integration - you can integrate Azure Databricks with your Docker CI/CD pipelines

还有许多其他用例,涵盖从“指定配置”到“安装机器学习包”等许多方面。There are many other use cases, ranging from specifying configuration to installing machine learning packages.

有关详细信息,请参阅使用 Databricks 容器服务自定义容器For details, see Customize containers with Databricks Container Services.

用于基因组学的 Databricks Runtime 6.2 正式版Databricks Runtime 6.2 for Genomics GA

2019 年 12 月 3 日December 3, 2019

用于基因组学的 Databricks Runtime 6.2 是基于 Databricks Runtime 6.2 构建的。Databricks Runtime 6.2 for Genomics is built on top of Databricks Runtime 6.2. 它包含用于基因组学的 Databricks Runtime 6.1 的许多改进和升级,包括:It includes many improvements and upgrades from Databricks Runtime 6.1 for Genomics, including:

  • Firth 逻辑回归Firth logistic regression
  • 用户定义的样本质量控制指标User-defined sample quality control metrics
  • 管道转换器性能提升Pipe transformer performance improvement
  • 更可靠的联合基因分型More robust joint genotyping
  • 简化了与 LOFTEE 的集成Simplified integration with LOFTEE
  • Hail 0.26.0Hail 0.26.0
  • Samtools 1.9Samtools 1.9

有关详细信息,请参阅完整的用于基因组学的 Databricks Runtime 6.2(不受支持)发行说明。For more information, see the complete Databricks Runtime 6.2 for Genomics (Unsupported) release notes.

2019 年 12 月 3 日December 3, 2019

Azure Active Directory 应用库中已提供 Azure Databricks SCIM 预配连接器,现在可以更轻松地设置从 Azure Active Directory 到 Azure Databricks 的用户和组的预配。Now that the Azure Databricks SCIM Provisioning Connector is available in the Azure Active Directory app gallery, it is easier to set up provisioning of users and groups from Azure Active Directory to Azure Databricks. 有关详细信息,请参阅为 Microsoft Azure Active Directory 配置 SCIM 预配For details, see Configure SCIM provisioning for Microsoft Azure Active Directory.

Databricks Runtime 5.3 和 5.4 支持结束Databricks Runtime 5.3 and 5.4 support ends

2019 年 12 月 3 日December 3, 2019

对 Databricks Runtime 5.3 和 5.4 的支持于 12 月 3 日结束。Support for 5.3 and 5.4 ended on December 3. 请参阅 Databricks 运行时支持生命周期See Databricks runtime support lifecycle.

Databricks Runtime 6.2 ML 正式版Databricks Runtime 6.2 ML GA

2019 年 12 月 3 日December 3, 2019

Databricks Runtime 6.2 ML 的正式发布引入了许多库升级,其中包括:Databricks Runtime 6.2 ML GA brings many library upgrades, including:

  • TensorFlow 和 TensorBoard:1.14.0 到 1.15.0。TensorFlow and TensorBoard: 1.14.0 to 1.15.0.
  • PyTorch:1.2.0 到 1.3.0。PyTorch: 1.2.0 to 1.3.0.
  • tensorboardX:1.8 到 1.9。tensorboardX: 1.8 to 1.9.
  • MLflow:1.3.0 到 1.4.0。MLflow: 1.3.0 to 1.4.0.
  • Hyperopt:0.2-db1 与 Azure Databricks MLflow 集成。Hyperopt: 0.2-db1 with Azure Databricks MLflow integrations.
  • mleap-databricks-runtime 到 0.15.0,包括 mleap-xgboost-runtime。mleap-databricks-runtime to 0.15.0 and includes mleap-xgboost-runtime.

有关详细信息,请参阅完整的用于机器学习的 Databricks Runtime 6.2(不受支持)发行说明。For more information, see the complete Databricks Runtime 6.2 for Machine Learning (Unsupported) release notes.

Databricks Runtime 6.2 正式版Databricks Runtime 6.2 GA

2019 年 12 月 3 日December 3, 2019

Databricks Runtime 6.2 的正式发布引入了新功能、改进和许多 bug 修复,包括:Databricks Runtime 6.2 GA brings new features, improvements, and many bug fixes, including:

  • 优化的 Delta Lake 仅插入合并Optimized Delta Lake insert-only merge

有关详细信息,请参阅完整的 Databricks Runtime 6.2(不受支持)发行说明。For more information, see the complete Databricks Runtime 6.2 (Unsupported) release notes.

Databricks Connect 现在支持 Databricks Runtime 6.1Databricks Connect now supports Databricks Runtime 6.1

2019 年 12 月 3 日December 3, 2019

Databricks Connect 现在支持 Databricks Runtime 6.1。Databricks Connect now supports Databricks Runtime 6.1. 通过 Databricks Connect,可将喜欢的 IDE(IntelliJ、Eclipse、PyCharm、RStudio 和 Visual Studio)、笔记本服务器(Zeppelin 和 Jupyter)和其他自定义应用程序连接到 Azure Databricks 群集,并运行 Apache Spark 代码。Databricks Connect allows you to connect your favorite IDE (IntelliJ, Eclipse, PyCharm, RStudio, Visual Studio), notebook server (Zeppelin, Jupyter), and other custom applications to Azure Databricks clusters and run Apache Spark code.