替代 Databricks 资产捆绑包中的群集设置

本文介绍如何替代 Databricks 资产捆绑包中 Azure Databricks 群集的设置。 请参阅什么是 Databricks 资产捆绑包?

在 Azure Databricks 捆绑配置文件中,可以将顶级 resources 映射中的群集设置与 targets 映射中的群集设置合并,如以下所示。

对于作业,请使用作业定义中的 job_cluster_key 映射,将顶级 resources 映射中的群集设置与 targets 映射中的群集设置联接,例如(省略号指示省略的内容,为简洁起见):

# ...
resources:
  jobs:
    <some-unique-programmatic-identifier-for-this-job>:
      # ...
      job_clusters:
        - job_cluster_key: <some-unique-programmatic-identifier-for-this-key>
          new_cluster:
            # Cluster settings.

targets:
  <some-unique-programmatic-identifier-for-this-target>:
    resources:
      jobs:
        <the-matching-programmatic-identifier-for-this-job>:
          # ...
          job_clusters:
            - job_cluster_key: <the-matching-programmatic-identifier-for-this-key>
              # Any more cluster settings to join with the settings from the
              # resources mapping for the matching top-level job_cluster_key.
          # ...

如果在顶级 resources 映射和同一 targetsjob_cluster_key 映射中都定义了任何群集设置,则 targets 映射中的设置优先于顶级 resources 映射中的设置。

对于 DLT 管道,请使用 label 管道定义中的映射 cluster 将顶级映射中的 resources 群集设置与映射中的 targets 群集设置联接,例如(省略号指示省略的内容,为简洁起见):

# ...
resources:
  pipelines:
    <some-unique-programmatic-identifier-for-this-pipeline>:
      # ...
      clusters:
        - label: default | maintenance
          # Cluster settings.

targets:
  <some-unique-programmatic-identifier-for-this-target>:
    resources:
      pipelines:
        <the-matching-programmatic-identifier-for-this-pipeline>:
          # ...
          clusters:
            - label: default | maintenance
              # Any more cluster settings to join with the settings from the
              # resources mapping for the matching top-level label.
          # ...

如果在顶级 resources 映射和同一 targetslabel 映射中都定义了任何群集设置,则 targets 映射中的设置优先于顶级 resources 映射中的设置。

示例 1:在多个资源映射中定义的新作业群集设置,且无设置冲突

在此示例中,顶级 spark_version 映射中的 resourcesnode_type_id 映射中 num_workersresourcestargets 中相结合,以定义名为 job_cluster_keymy-cluster 的设置(省略号指示省略的内容,为简洁起见):

# ...
resources:
  jobs:
    my-job:
      name: my-job
      job_clusters:
        - job_cluster_key: my-cluster
          new_cluster:
            spark_version: 13.3.x-scala2.12

targets:
  development:
    resources:
      jobs:
        my-job:
          name: my-job
          job_clusters:
            - job_cluster_key: my-cluster
              new_cluster:
                node_type_id: Standard_DS3_v2
                num_workers: 1
          # ...

为此示例运行 databricks bundle validate 时,生成的图形如下所示(省略号指示省略的内容,为简洁起见):

{
  "...": "...",
  "resources": {
    "jobs": {
      "my-job": {
        "job_clusters": [
          {
            "job_cluster_key": "my-cluster",
            "new_cluster": {
              "node_type_id": "Standard_DS3_v2",
              "num_workers": 1,
              "spark_version": "13.3.x-scala2.12"
            }
          }
        ],
        "...": "..."
      }
    }
  }
}

示例 2:在多个资源映射中定义了冲突的新作业群集设置

在此示例中,spark_versionnum_workers 在顶级 resources 映射和 resourcestargets 映射中定义。 在此示例中,spark_versionnum_workers 映射中的 resourcestargets 优先于顶级 spark_version 映射中的 num_workersresources,以定义命名 job_cluster_keymy-cluster 的设置(省略号指示省略的内容,为简洁起见):

# ...
resources:
  jobs:
    my-job:
      name: my-job
      job_clusters:
        - job_cluster_key: my-cluster
          new_cluster:
            spark_version: 13.3.x-scala2.12
            node_type_id: Standard_DS3_v2
            num_workers: 1

targets:
  development:
    resources:
      jobs:
        my-job:
          name: my-job
          job_clusters:
            - job_cluster_key: my-cluster
              new_cluster:
                spark_version: 12.2.x-scala2.12
                num_workers: 2
          # ...

为此示例运行 databricks bundle validate 时,生成的图形如下所示(省略号指示省略的内容,为简洁起见):

{
  "...": "...",
  "resources": {
    "jobs": {
      "my-job": {
        "job_clusters": [
          {
            "job_cluster_key": "my-cluster",
            "new_cluster": {
              "node_type_id": "Standard_DS3_v2",
              "num_workers": 2,
              "spark_version": "12.2.x-scala2.12"
            }
          }
        ],
        "...": "..."
      }
    }
  }
}

示例 3:在多个资源映射中定义的管道群集设置,且无设置冲突

在此示例中,顶级 node_type_id 映射中的 resources 结合了 num_workersresources 映射中的 targets,以定义名称为 labeldefault 的设置(省略号指示省略的内容,为简洁起见):

# ...
resources:
  pipelines:
    my-pipeline:
      clusters:
        - label: default
          node_type_id: Standard_DS3_v2

targets:
  development:
    resources:
      pipelines:
        my-pipeline:
          clusters:
            - label: default
              num_workers: 1
          # ...

为此示例运行 databricks bundle validate 时,生成的图形如下所示(省略号指示省略的内容,为简洁起见):

{
  "...": "...",
  "resources": {
    "pipelines": {
      "my-pipeline": {
        "clusters": [
          {
            "label": "default",
            "node_type_id": "Standard_DS3_v2",
            "num_workers": 1
          }
        ],
        "...": "..."
      }
    }
  }
}

示例 4:在多个资源映射中定义了冲突的管道群集设置

在此示例中,num_workers 在顶级 resources 映射和 resourcestargets 映射中定义。 num_workersresources 映射中的 targets 优先于顶级 num_workers 映射中的 resources,用于为名为 labeldefault 定义设置(为简洁起见,下面用省略号表示省略的内容):

# ...
resources:
  pipelines:
    my-pipeline:
      clusters:
        - label: default
          node_type_id: Standard_DS3_v2
          num_workers: 1

targets:
  development:
    resources:
      pipelines:
        my-pipeline:
          clusters:
            - label: default
              num_workers: 2
          # ...

为此示例运行 databricks bundle validate 时,生成的图形如下所示(省略号指示省略的内容,为简洁起见):

{
  "...": "...",
  "resources": {
    "pipelines": {
      "my-pipeline": {
        "clusters": [
          {
            "label": "default",
            "node_type_id": "Standard_DS3_v2",
            "num_workers": 2
          }
        ],
        "...": "..."
      }
    }
  }
}