xref: /aosp_15_r20/external/googleapis/google/cloud/dataproc/v1/shared.proto (revision d5c09012810ac0c9f33fe448fb6da8260d444cc9)

// Copyright 2023 Google LLC
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
//     http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.

syntax = "proto3";

package google.cloud.dataproc.v1;

import "google/api/field_behavior.proto";
import "google/api/resource.proto";
import "google/protobuf/duration.proto";
import "google/protobuf/timestamp.proto";

option go_package = "cloud.google.com/go/dataproc/v2/apiv1/dataprocpb;dataprocpb";
option java_multiple_files = true;
option java_outer_classname = "SharedProto";
option java_package = "com.google.cloud.dataproc.v1";
option (google.api.resource_definition) = {
  type: "container.googleapis.com/Cluster"
  pattern: "projects/{project}/locations/{location}/clusters/{cluster}"
};
option (google.api.resource_definition) = {
  type: "metastore.googleapis.com/Service"
  pattern: "projects/{project}/locations/{location}/services/{service}"
};

// Runtime configuration for a workload.
message RuntimeConfig {
  // Optional. Version of the batch runtime.
  string version = 1 [(google.api.field_behavior) = OPTIONAL];

  // Optional. Optional custom container image for the job runtime environment.
  // If not specified, a default container image will be used.
  string container_image = 2 [(google.api.field_behavior) = OPTIONAL];

  // Optional. A mapping of property names to values, which are used to
  // configure workload execution.
  map<string, string> properties = 3 [(google.api.field_behavior) = OPTIONAL];

  // Optional. Dependency repository configuration.
  RepositoryConfig repository_config = 5
      [(google.api.field_behavior) = OPTIONAL];
}

// Environment configuration for a workload.
message EnvironmentConfig {
  // Optional. Execution configuration for a workload.
  ExecutionConfig execution_config = 1 [(google.api.field_behavior) = OPTIONAL];

  // Optional. Peripherals configuration that workload has access to.
  PeripheralsConfig peripherals_config = 2
      [(google.api.field_behavior) = OPTIONAL];
}

// Execution configuration for a workload.
message ExecutionConfig {
  // Optional. Service account that used to execute workload.
  string service_account = 2 [(google.api.field_behavior) = OPTIONAL];

  // Network configuration for workload execution.
  oneof network {
    // Optional. Network URI to connect workload to.
    string network_uri = 4 [(google.api.field_behavior) = OPTIONAL];

    // Optional. Subnetwork URI to connect workload to.
    string subnetwork_uri = 5 [(google.api.field_behavior) = OPTIONAL];
  }

  // Optional. Tags used for network traffic control.
  repeated string network_tags = 6 [(google.api.field_behavior) = OPTIONAL];

  // Optional. The Cloud KMS key to use for encryption.
  string kms_key = 7 [(google.api.field_behavior) = OPTIONAL];

  // Optional. Applies to sessions only. The duration to keep the session alive
  // while it's idling. Exceeding this threshold causes the session to
  // terminate. This field cannot be set on a batch workload. Minimum value is
  // 10 minutes; maximum value is 14 days (see JSON representation of
  // [Duration](https://developers.google.com/protocol-buffers/docs/proto3#json)).
  // Defaults to 1 hour if not set.
  // If both `ttl` and `idle_ttl` are specified for an interactive session,
  // the conditions are treated as `OR` conditions: the workload will be
  // terminated when it has been idle for `idle_ttl` or when `ttl` has been
  // exceeded, whichever occurs first.
  google.protobuf.Duration idle_ttl = 8
      [(google.api.field_behavior) = OPTIONAL];

  // Optional. The duration after which the workload will be terminated,
  // specified as the JSON representation for
  // [Duration](https://protobuf.dev/programming-guides/proto3/#json).
  // When the workload exceeds this duration, it will be unconditionally
  // terminated without waiting for ongoing work to finish. If `ttl` is not
  // specified for a batch workload, the workload will be allowed to run until
  // it exits naturally (or run forever without exiting). If `ttl` is not
  // specified for an interactive session, it defaults to 24 hours. If `ttl` is
  // not specified for a batch that uses 2.1+ runtime version, it defaults to 4
  // hours. Minimum value is 10 minutes; maximum value is 14 days. If both `ttl`
  // and `idle_ttl` are specified (for an interactive session), the conditions
  // are treated as `OR` conditions: the workload will be terminated when it has
  // been idle for `idle_ttl` or when `ttl` has been exceeded, whichever occurs
  // first.
  google.protobuf.Duration ttl = 9 [(google.api.field_behavior) = OPTIONAL];

  // Optional. A Cloud Storage bucket used to stage workload dependencies,
  // config files, and store workload output and other ephemeral data, such as
  // Spark history files. If you do not specify a staging bucket, Cloud Dataproc
  // will determine a Cloud Storage location according to the region where your
  // workload is running, and then create and manage project-level, per-location
  // staging and temporary buckets.
  // **This field requires a Cloud Storage bucket name, not a `gs://...` URI to
  // a Cloud Storage bucket.**
  string staging_bucket = 10 [(google.api.field_behavior) = OPTIONAL];
}

// Spark History Server configuration for the workload.
message SparkHistoryServerConfig {
  // Optional. Resource name of an existing Dataproc Cluster to act as a Spark
  // History Server for the workload.
  //
  // Example:
  //
  // * `projects/[project_id]/regions/[region]/clusters/[cluster_name]`
  string dataproc_cluster = 1 [(google.api.field_behavior) = OPTIONAL];
}

// Auxiliary services configuration for a workload.
message PeripheralsConfig {
  // Optional. Resource name of an existing Dataproc Metastore service.
  //
  // Example:
  //
  // * `projects/[project_id]/locations/[region]/services/[service_id]`
  string metastore_service = 1 [
    (google.api.field_behavior) = OPTIONAL,
    (google.api.resource_reference) = {
      type: "metastore.googleapis.com/Service"
    }
  ];

  // Optional. The Spark History Server configuration for the workload.
  SparkHistoryServerConfig spark_history_server_config = 2
      [(google.api.field_behavior) = OPTIONAL];
}

// Runtime information about workload execution.
message RuntimeInfo {
  // Output only. Map of remote access endpoints (such as web interfaces and
  // APIs) to their URIs.
  map<string, string> endpoints = 1 [(google.api.field_behavior) = OUTPUT_ONLY];

  // Output only. A URI pointing to the location of the stdout and stderr of the
  // workload.
  string output_uri = 2 [(google.api.field_behavior) = OUTPUT_ONLY];

  // Output only. A URI pointing to the location of the diagnostics tarball.
  string diagnostic_output_uri = 3 [(google.api.field_behavior) = OUTPUT_ONLY];

  // Output only. Approximate workload resource usage, calculated when
  // the workload completes (see [Dataproc Serverless pricing]
  // (https://cloud.google.com/dataproc-serverless/pricing)).
  //
  // **Note:** This metric calculation may change in the future, for
  // example, to capture cumulative workload resource
  // consumption during workload execution (see the
  // [Dataproc Serverless release notes]
  // (https://cloud.google.com/dataproc-serverless/docs/release-notes)
  // for announcements, changes, fixes
  // and other Dataproc developments).
  UsageMetrics approximate_usage = 6
      [(google.api.field_behavior) = OUTPUT_ONLY];

  // Output only. Snapshot of current workload resource usage.
  UsageSnapshot current_usage = 7 [(google.api.field_behavior) = OUTPUT_ONLY];
}

// Usage metrics represent approximate total resources consumed by a workload.
message UsageMetrics {
  // Optional. DCU (Dataproc Compute Units) usage in (`milliDCU` x `seconds`)
  // (see [Dataproc Serverless pricing]
  // (https://cloud.google.com/dataproc-serverless/pricing)).
  int64 milli_dcu_seconds = 1 [(google.api.field_behavior) = OPTIONAL];

  // Optional. Shuffle storage usage in (`GB` x `seconds`) (see
  // [Dataproc Serverless pricing]
  // (https://cloud.google.com/dataproc-serverless/pricing)).
  int64 shuffle_storage_gb_seconds = 2 [(google.api.field_behavior) = OPTIONAL];

  // Optional. Accelerator usage in (`milliAccelerator` x `seconds`) (see
  // [Dataproc Serverless pricing]
  // (https://cloud.google.com/dataproc-serverless/pricing)).
  int64 milli_accelerator_seconds = 3 [(google.api.field_behavior) = OPTIONAL];

  // Optional. Accelerator type being used, if any
  string accelerator_type = 4 [(google.api.field_behavior) = OPTIONAL];
}

// The usage snapshot represents the resources consumed by a workload at a
// specified time.
message UsageSnapshot {
  // Optional. Milli (one-thousandth) Dataproc Compute Units (DCUs) (see
  // [Dataproc Serverless pricing]
  // (https://cloud.google.com/dataproc-serverless/pricing)).
  int64 milli_dcu = 1 [(google.api.field_behavior) = OPTIONAL];

  // Optional. Shuffle Storage in gigabytes (GB). (see [Dataproc Serverless
  // pricing] (https://cloud.google.com/dataproc-serverless/pricing))
  int64 shuffle_storage_gb = 2 [(google.api.field_behavior) = OPTIONAL];

  // Optional. Milli (one-thousandth) Dataproc Compute Units (DCUs) charged at
  // premium tier (see [Dataproc Serverless pricing]
  // (https://cloud.google.com/dataproc-serverless/pricing)).
  int64 milli_dcu_premium = 4 [(google.api.field_behavior) = OPTIONAL];

  // Optional. Shuffle Storage in gigabytes (GB) charged at premium tier. (see
  // [Dataproc Serverless pricing]
  // (https://cloud.google.com/dataproc-serverless/pricing))
  int64 shuffle_storage_gb_premium = 5 [(google.api.field_behavior) = OPTIONAL];

  // Optional. Milli (one-thousandth) accelerator. (see [Dataproc
  // Serverless pricing] (https://cloud.google.com/dataproc-serverless/pricing))
  int64 milli_accelerator = 6 [(google.api.field_behavior) = OPTIONAL];

  // Optional. Accelerator type being used, if any
  string accelerator_type = 7 [(google.api.field_behavior) = OPTIONAL];

  // Optional. The timestamp of the usage snapshot.
  google.protobuf.Timestamp snapshot_time = 3
      [(google.api.field_behavior) = OPTIONAL];
}

// The cluster's GKE config.
message GkeClusterConfig {
  // Optional. A target GKE cluster to deploy to. It must be in the same project
  // and region as the Dataproc cluster (the GKE cluster can be zonal or
  // regional). Format:
  // 'projects/{project}/locations/{location}/clusters/{cluster_id}'
  string gke_cluster_target = 2 [
    (google.api.field_behavior) = OPTIONAL,
    (google.api.resource_reference) = {
      type: "container.googleapis.com/Cluster"
    }
  ];

  // Optional. GKE node pools where workloads will be scheduled. At least one
  // node pool must be assigned the `DEFAULT`
  // [GkeNodePoolTarget.Role][google.cloud.dataproc.v1.GkeNodePoolTarget.Role].
  // If a `GkeNodePoolTarget` is not specified, Dataproc constructs a `DEFAULT`
  // `GkeNodePoolTarget`. Each role can be given to only one
  // `GkeNodePoolTarget`. All node pools must have the same location settings.
  repeated GkeNodePoolTarget node_pool_target = 3
      [(google.api.field_behavior) = OPTIONAL];
}

// The configuration for running the Dataproc cluster on Kubernetes.
message KubernetesClusterConfig {
  // Optional. A namespace within the Kubernetes cluster to deploy into. If this
  // namespace does not exist, it is created. If it exists, Dataproc verifies
  // that another Dataproc VirtualCluster is not installed into it. If not
  // specified, the name of the Dataproc Cluster is used.
  string kubernetes_namespace = 1 [(google.api.field_behavior) = OPTIONAL];

  oneof config {
    // Required. The configuration for running the Dataproc cluster on GKE.
    GkeClusterConfig gke_cluster_config = 2
        [(google.api.field_behavior) = REQUIRED];
  }

  // Optional. The software configuration for this Dataproc cluster running on
  // Kubernetes.
  KubernetesSoftwareConfig kubernetes_software_config = 3
      [(google.api.field_behavior) = OPTIONAL];
}

// The software configuration for this Dataproc cluster running on Kubernetes.
message KubernetesSoftwareConfig {
  // The components that should be installed in this Dataproc cluster. The key
  // must be a string from the KubernetesComponent enumeration. The value is
  // the version of the software to be installed.
  // At least one entry must be specified.
  map<string, string> component_version = 1;

  // The properties to set on daemon config files.
  //
  // Property keys are specified in `prefix:property` format, for example
  // `spark:spark.kubernetes.container.image`. The following are supported
  // prefixes and their mappings:
  //
  // * spark:  `spark-defaults.conf`
  //
  // For more information, see [Cluster
  // properties](https://cloud.google.com/dataproc/docs/concepts/cluster-properties).
  map<string, string> properties = 2;
}

// GKE node pools that Dataproc workloads run on.
message GkeNodePoolTarget {
  // `Role` specifies the tasks that will run on the node pool. Roles can be
  // specific to workloads. Exactly one
  // [GkeNodePoolTarget][google.cloud.dataproc.v1.GkeNodePoolTarget] within the
  // virtual cluster must have the `DEFAULT` role, which is used to run all
  // workloads that are not associated with a node pool.
  enum Role {
    // Role is unspecified.
    ROLE_UNSPECIFIED = 0;

    // At least one node pool must have the `DEFAULT` role.
    // Work assigned to a role that is not associated with a node pool
    // is assigned to the node pool with the `DEFAULT` role. For example,
    // work assigned to the `CONTROLLER` role will be assigned to the node pool
    // with the `DEFAULT` role if no node pool has the `CONTROLLER` role.
    DEFAULT = 1;

    // Run work associated with the Dataproc control plane (for example,
    // controllers and webhooks). Very low resource requirements.
    CONTROLLER = 2;

    // Run work associated with a Spark driver of a job.
    SPARK_DRIVER = 3;

    // Run work associated with a Spark executor of a job.
    SPARK_EXECUTOR = 4;
  }

  // Required. The target GKE node pool.
  // Format:
  // 'projects/{project}/locations/{location}/clusters/{cluster}/nodePools/{node_pool}'
  string node_pool = 1 [(google.api.field_behavior) = REQUIRED];

  // Required. The roles associated with the GKE node pool.
  repeated Role roles = 2 [(google.api.field_behavior) = REQUIRED];

  // Input only. The configuration for the GKE node pool.
  //
  // If specified, Dataproc attempts to create a node pool with the
  // specified shape. If one with the same name already exists, it is
  // verified against all specified fields. If a field differs, the
  // virtual cluster creation will fail.
  //
  // If omitted, any node pool with the specified name is used. If a
  // node pool with the specified name does not exist, Dataproc create a
  // node pool with default values.
  //
  // This is an input only field. It will not be returned by the API.
  GkeNodePoolConfig node_pool_config = 3
      [(google.api.field_behavior) = INPUT_ONLY];
}

// The configuration of a GKE node pool used by a [Dataproc-on-GKE
// cluster](https://cloud.google.com/dataproc/docs/concepts/jobs/dataproc-gke#create-a-dataproc-on-gke-cluster).
message GkeNodePoolConfig {
  // Parameters that describe cluster nodes.
  message GkeNodeConfig {
    // Optional. The name of a Compute Engine [machine
    // type](https://cloud.google.com/compute/docs/machine-types).
    string machine_type = 1 [(google.api.field_behavior) = OPTIONAL];

    // Optional. The number of local SSD disks to attach to the node, which is
    // limited by the maximum number of disks allowable per zone (see [Adding
    // Local SSDs](https://cloud.google.com/compute/docs/disks/local-ssd)).
    int32 local_ssd_count = 7 [(google.api.field_behavior) = OPTIONAL];

    // Optional. Whether the nodes are created as legacy [preemptible VM
    // instances] (https://cloud.google.com/compute/docs/instances/preemptible).
    // Also see
    // [Spot][google.cloud.dataproc.v1.GkeNodePoolConfig.GkeNodeConfig.spot]
    // VMs, preemptible VM instances without a maximum lifetime. Legacy and Spot
    // preemptible nodes cannot be used in a node pool with the `CONTROLLER`
    // [role]
    // (/dataproc/docs/reference/rest/v1/projects.regions.clusters#role)
    // or in the DEFAULT node pool if the CONTROLLER role is not assigned (the
    // DEFAULT node pool will assume the CONTROLLER role).
    bool preemptible = 10 [(google.api.field_behavior) = OPTIONAL];

    // Optional. A list of [hardware
    // accelerators](https://cloud.google.com/compute/docs/gpus) to attach to
    // each node.
    repeated GkeNodePoolAcceleratorConfig accelerators = 11
        [(google.api.field_behavior) = OPTIONAL];

    // Optional. [Minimum CPU
    // platform](https://cloud.google.com/compute/docs/instances/specify-min-cpu-platform)
    // to be used by this instance. The instance may be scheduled on the
    // specified or a newer CPU platform. Specify the friendly names of CPU
    // platforms, such as "Intel Haswell"` or Intel Sandy Bridge".
    string min_cpu_platform = 13 [(google.api.field_behavior) = OPTIONAL];

    // Optional. The [Customer Managed Encryption Key (CMEK)]
    // (https://cloud.google.com/kubernetes-engine/docs/how-to/using-cmek)
    // used to encrypt the boot disk attached to each node in the node pool.
    // Specify the key using the following format:
    // <code>projects/<var>KEY_PROJECT_ID</var>/locations/<var>LOCATION</var>/keyRings/<var>RING_NAME</var>/cryptoKeys/<var>KEY_NAME</var></code>.
    string boot_disk_kms_key = 23 [(google.api.field_behavior) = OPTIONAL];

    // Optional. Whether the nodes are created as [Spot VM instances]
    // (https://cloud.google.com/compute/docs/instances/spot).
    // Spot VMs are the latest update to legacy
    // [preemptible
    // VMs][google.cloud.dataproc.v1.GkeNodePoolConfig.GkeNodeConfig.preemptible].
    // Spot VMs do not have a maximum lifetime. Legacy and Spot preemptible
    // nodes cannot be used in a node pool with the `CONTROLLER`
    // [role](/dataproc/docs/reference/rest/v1/projects.regions.clusters#role)
    // or in the DEFAULT node pool if the CONTROLLER role is not assigned (the
    // DEFAULT node pool will assume the CONTROLLER role).
    bool spot = 32 [(google.api.field_behavior) = OPTIONAL];
  }

  // A GkeNodeConfigAcceleratorConfig represents a Hardware Accelerator request
  // for a node pool.
  message GkeNodePoolAcceleratorConfig {
    // The number of accelerator cards exposed to an instance.
    int64 accelerator_count = 1;

    // The accelerator type resource namename (see GPUs on Compute Engine).
    string accelerator_type = 2;

    // Size of partitions to create on the GPU. Valid values are described in
    // the NVIDIA [mig user
    // guide](https://docs.nvidia.com/datacenter/tesla/mig-user-guide/#partitioning).
    string gpu_partition_size = 3;
  }

  // GkeNodePoolAutoscaling contains information the cluster autoscaler needs to
  // adjust the size of the node pool to the current cluster usage.
  message GkeNodePoolAutoscalingConfig {
    // The minimum number of nodes in the node pool. Must be >= 0 and <=
    // max_node_count.
    int32 min_node_count = 2;

    // The maximum number of nodes in the node pool. Must be >= min_node_count,
    // and must be > 0.
    // **Note:** Quota must be sufficient to scale up the cluster.
    int32 max_node_count = 3;
  }

  // Optional. The node pool configuration.
  GkeNodeConfig config = 2 [(google.api.field_behavior) = OPTIONAL];

  // Optional. The list of Compute Engine
  // [zones](https://cloud.google.com/compute/docs/zones#available) where
  // node pool nodes associated with a Dataproc on GKE virtual cluster
  // will be located.
  //
  // **Note:** All node pools associated with a virtual cluster
  // must be located in the same region as the virtual cluster, and they must
  // be located in the same zone within that region.
  //
  // If a location is not specified during node pool creation, Dataproc on GKE
  // will choose the zone.
  repeated string locations = 13 [(google.api.field_behavior) = OPTIONAL];

  // Optional. The autoscaler configuration for this node pool. The autoscaler
  // is enabled only when a valid configuration is present.
  GkeNodePoolAutoscalingConfig autoscaling = 4
      [(google.api.field_behavior) = OPTIONAL];
}

// Configuration for dependency repositories
message RepositoryConfig {
  // Optional. Configuration for PyPi repository.
  PyPiRepositoryConfig pypi_repository_config = 1
      [(google.api.field_behavior) = OPTIONAL];
}

// Configuration for PyPi repository
message PyPiRepositoryConfig {
  // Optional. PyPi repository address
  string pypi_repository = 1 [(google.api.field_behavior) = OPTIONAL];
}

// Cluster components that can be activated.
enum Component {
  // Unspecified component. Specifying this will cause Cluster creation to fail.
  COMPONENT_UNSPECIFIED = 0;

  // The Anaconda python distribution. The Anaconda component is not supported
  // in the Dataproc [2.0 image]
  // (/https://cloud.google.com/dataproc/docs/concepts/versioning/dataproc-release-2.0).
  // The 2.0 image is pre-installed with Miniconda.
  ANACONDA = 5;

  // Docker
  DOCKER = 13;

  // The Druid query engine. (alpha)
  DRUID = 9;

  // Flink
  FLINK = 14;

  // HBase. (beta)
  HBASE = 11;

  // The Hive Web HCatalog (the REST service for accessing HCatalog).
  HIVE_WEBHCAT = 3;

  // Hudi.
  HUDI = 18;

  // The Jupyter Notebook.
  JUPYTER = 1;

  // The Presto query engine.
  PRESTO = 6;

  // The Trino query engine.
  TRINO = 17;

  // The Ranger service.
  RANGER = 12;

  // The Solr service.
  SOLR = 10;

  // The Zeppelin notebook.
  ZEPPELIN = 4;

  // The Zookeeper service.
  ZOOKEEPER = 8;
}

// Actions in response to failure of a resource associated with a cluster.
enum FailureAction {
  // When FailureAction is unspecified, failure action defaults to NO_ACTION.
  FAILURE_ACTION_UNSPECIFIED = 0;

  // Take no action on failure to create a cluster resource. NO_ACTION is the
  // default.
  NO_ACTION = 1;

  // Delete the failed cluster resource.
  DELETE = 2;
}