REST Resource: projects.locations.pipelineJobs

Resource: PipelineJob

An instance of a machine learning PipelineJob.

Fields
name string

Output only. The resource name of the PipelineJob.

displayName string

The display name of the Pipeline. The name can be up to 128 characters long and can consist of any UTF-8 characters.

createTime string (Timestamp format)

Output only. Pipeline creation time.

A timestamp in RFC3339 UTC "Zulu" format, with nanosecond resolution and up to nine fractional digits. Examples: "2014-10-02T15:01:23Z" and "2014-10-02T15:01:23.045123456Z".

startTime string (Timestamp format)

Output only. Pipeline start time.

A timestamp in RFC3339 UTC "Zulu" format, with nanosecond resolution and up to nine fractional digits. Examples: "2014-10-02T15:01:23Z" and "2014-10-02T15:01:23.045123456Z".

endTime string (Timestamp format)

Output only. Pipeline end time.

A timestamp in RFC3339 UTC "Zulu" format, with nanosecond resolution and up to nine fractional digits. Examples: "2014-10-02T15:01:23Z" and "2014-10-02T15:01:23.045123456Z".

updateTime string (Timestamp format)

Output only. timestamp when this PipelineJob was most recently updated.

A timestamp in RFC3339 UTC "Zulu" format, with nanosecond resolution and up to nine fractional digits. Examples: "2014-10-02T15:01:23Z" and "2014-10-02T15:01:23.045123456Z".

pipelineSpec object (Struct format)

The spec of the pipeline.

state enum (PipelineState)

Output only. The detailed state of the job.

jobDetail object (PipelineJobDetail)

Output only. The details of pipeline run. Not available in the list view.

error object (Status)

Output only. The error that occurred during pipeline execution. Only populated when the pipeline's state is FAILED or CANCELLED.

labels map (key: string, value: string)

The labels with user-defined metadata to organize PipelineJob.

label keys and values can be no longer than 64 characters (Unicode codepoints), can only contain lowercase letters, numeric characters, underscores and dashes. International characters are allowed.

See https://goo.gl/xmQnxf for more information and examples of labels.

Note there is some reserved label key for Vertex AI Pipelines. - vertex-ai-pipelines-run-billing-id, user set value will get overrided.

runtimeConfig object (RuntimeConfig)

Runtime config of the pipeline.

encryptionSpec object (EncryptionSpec)

Customer-managed encryption key spec for a pipelineJob. If set, this PipelineJob and all of its sub-resources will be secured by this key.

serviceAccount string

The service account that the pipeline workload runs as. If not specified, the Compute Engine default service account in the project will be used. See https://cloud.google.com/compute/docs/access/service-accounts#default_service_account

Users starting the pipeline must have the iam.serviceAccounts.actAs permission on this service account.

network string

The full name of the Compute Engine network to which the Pipeline Job's workload should be peered. For example, projects/12345/global/networks/myVPC. Format is of the form projects/{project}/global/networks/{network}. Where {project} is a project number, as in 12345, and {network} is a network name.

Private services access must already be configured for the network. Pipeline job will apply the network configuration to the Google Cloud resources being launched, if applied, such as Vertex AI Training or Dataflow job. If left unspecified, the workload is not peered with any network.

reservedIpRanges[] string

A list of names for the reserved ip ranges under the VPC network that can be used for this Pipeline Job's workload.

If set, we will deploy the Pipeline Job's workload within the provided ip ranges. Otherwise, the job will be deployed to any ip ranges under the provided VPC network.

Example: ['vertex-ai-ip-range'].

pscInterfaceConfig object (PscInterfaceConfig)

Optional. Configuration for PSC-I for PipelineJob.

templateUri string

A template uri from where the PipelineJob.pipeline_spec, if empty, will be downloaded. Currently, only uri from Vertex Template Registry & Gallery is supported. Reference to https://cloud.google.com/vertex-ai/docs/pipelines/create-pipeline-template.

templateMetadata object (PipelineTemplateMetadata)

Output only. Pipeline template metadata. Will fill up fields if PipelineJob.template_uri is from supported template registry.

scheduleName string

Output only. The schedule resource name. Only returned if the Pipeline is created by Schedule API.

preflightValidations boolean

Optional. Whether to do component level validations before job creation.

satisfiesPzs boolean

Output only. reserved for future use.

satisfiesPzi boolean

Output only. reserved for future use.

originalPipelineJobId string (int64 format)

Optional. The original pipeline job id if this pipeline job is a rerun of a previous pipeline job.

pipelineTaskRerunConfigs[] object (PipelineTaskRerunConfig)

Optional. The rerun configs for each task in the pipeline job. By default, the rerun will: 1. Use the same input artifacts as the original run. 2. Use the same input parameters as the original run. 3. Skip all the tasks that are already succeeded in the original run. 4. Rerun all the tasks that are not succeeded in the original run. By providing this field, users can override the default behavior and specify the rerun config for each task.

JSON representation 
{
  "name": string,
  "displayName": string,
  "createTime": string,
  "startTime": string,
  "endTime": string,
  "updateTime": string,
  "pipelineSpec": {
    object
  },
  "state": enum (PipelineState),
  "jobDetail": {
    object (PipelineJobDetail)
  },
  "error": {
    object (Status)
  },
  "labels": {
    string: string,
    ...
  },
  "runtimeConfig": {
    object (RuntimeConfig)
  },
  "encryptionSpec": {
    object (EncryptionSpec)
  },
  "serviceAccount": string,
  "network": string,
  "reservedIpRanges": [
    string
  ],
  "pscInterfaceConfig": {
    object (PscInterfaceConfig)
  },
  "templateUri": string,
  "templateMetadata": {
    object (PipelineTemplateMetadata)
  },
  "scheduleName": string,
  "preflightValidations": boolean,
  "satisfiesPzs": boolean,
  "satisfiesPzi": boolean,
  "originalPipelineJobId": string,
  "pipelineTaskRerunConfigs": [
    {
      object (PipelineTaskRerunConfig)
    }
  ]
}

PipelineJobDetail

The runtime detail of PipelineJob.

Fields
pipelineContext object (Context)

Output only. The context of the pipeline.

pipelineRunContext object (Context)

Output only. The context of the current pipeline run.

taskDetails[] object (PipelineTaskDetail)

Output only. The runtime details of the tasks under the pipeline.

JSON representation 
{
  "pipelineContext": {
    object (Context)
  },
  "pipelineRunContext": {
    object (Context)
  },
  "taskDetails": [
    {
      object (PipelineTaskDetail)
    }
  ]
}

PipelineTaskDetail

The runtime detail of a task execution.

Fields
taskId string (int64 format)

Output only. The system generated id of the task.

parentTaskId string (int64 format)

Output only. The id of the parent task if the task is within a component scope. Empty if the task is at the root level.

taskName string

Output only. The user specified name of the task that is defined in pipelineSpec.

createTime string (Timestamp format)

Output only. Task create time.

A timestamp in RFC3339 UTC "Zulu" format, with nanosecond resolution and up to nine fractional digits. Examples: "2014-10-02T15:01:23Z" and "2014-10-02T15:01:23.045123456Z".

startTime string (Timestamp format)

Output only. Task start time.

A timestamp in RFC3339 UTC "Zulu" format, with nanosecond resolution and up to nine fractional digits. Examples: "2014-10-02T15:01:23Z" and "2014-10-02T15:01:23.045123456Z".

endTime string (Timestamp format)

Output only. Task end time.

A timestamp in RFC3339 UTC "Zulu" format, with nanosecond resolution and up to nine fractional digits. Examples: "2014-10-02T15:01:23Z" and "2014-10-02T15:01:23.045123456Z".

executorDetail object (PipelineTaskExecutorDetail)

Output only. The detailed execution info.

state enum (State)

Output only. state of the task.

execution object (Execution)

Output only. The execution metadata of the task.

error object (Status)

Output only. The error that occurred during task execution. Only populated when the task's state is FAILED or CANCELLED.

pipelineTaskStatus[] object (PipelineTaskStatus)

Output only. A list of task status. This field keeps a record of task status evolving over time.

inputs map (key: string, value: object (ArtifactList))

Output only. The runtime input artifacts of the task.

outputs map (key: string, value: object (ArtifactList))

Output only. The runtime output artifacts of the task.

JSON representation 
{
  "taskId": string,
  "parentTaskId": string,
  "taskName": string,
  "createTime": string,
  "startTime": string,
  "endTime": string,
  "executorDetail": {
    object (PipelineTaskExecutorDetail)
  },
  "state": enum (State),
  "execution": {
    object (Execution)
  },
  "error": {
    object (Status)
  },
  "pipelineTaskStatus": [
    {
      object (PipelineTaskStatus)
    }
  ],
  "inputs": {
    string: {
      object (ArtifactList)
    },
    ...
  },
  "outputs": {
    string: {
      object (ArtifactList)
    },
    ...
  }
}

PipelineTaskExecutorDetail

The runtime detail of a pipeline executor.

Fields
details Union type
details can be only one of the following:
containerDetail object (ContainerDetail)

Output only. The detailed info for a container executor.

customJobDetail object (CustomJobDetail)

Output only. The detailed info for a custom job executor.

JSON representation 
{

  // details
  "containerDetail": {
    object (ContainerDetail)
  },
  "customJobDetail": {
    object (CustomJobDetail)
  }
  // Union type
}

ContainerDetail

The detail of a container execution. It contains the job names of the lifecycle of a container execution.

Fields
mainJob string

Output only. The name of the CustomJob for the main container execution.

preCachingCheckJob string

Output only. The name of the CustomJob for the pre-caching-check container execution. This job will be available if the PipelineJob.pipeline_spec specifies the pre_caching_check hook in the lifecycle events.

failedMainJobs[] string

Output only. The names of the previously failed CustomJob for the main container executions. The list includes the all attempts in chronological order.

failedPreCachingCheckJobs[] string

Output only. The names of the previously failed CustomJob for the pre-caching-check container executions. This job will be available if the PipelineJob.pipeline_spec specifies the pre_caching_check hook in the lifecycle events. The list includes the all attempts in chronological order.

JSON representation 
{
  "mainJob": string,
  "preCachingCheckJob": string,
  "failedMainJobs": [
    string
  ],
  "failedPreCachingCheckJobs": [
    string
  ]
}

CustomJobDetail

The detailed info for a custom job executor.

Fields
job string

Output only. The name of the CustomJob.

failedJobs[] string

Output only. The names of the previously failed CustomJob. The list includes the all attempts in chronological order.

JSON representation 
{
  "job": string,
  "failedJobs": [
    string
  ]
}

State

Specifies state of TaskExecution

Enums
STATE_UNSPECIFIED Unspecified.
PENDING Specifies pending state for the task.
RUNNING Specifies task is being executed.
SUCCEEDED Specifies task completed successfully.
CANCEL_PENDING Specifies Task cancel is in pending state.
CANCELLING Specifies task is being cancelled.
CANCELLED Specifies task was cancelled.
FAILED Specifies task failed.
SKIPPED Specifies task was skipped due to cache hit.
NOT_TRIGGERED Specifies that the task was not triggered because the task's trigger policy is not satisfied. The trigger policy is specified in the condition field of PipelineJob.pipeline_spec.

PipelineTaskStatus

A single record of the task status.

Fields
updateTime string (Timestamp format)

Output only. Update time of this status.

A timestamp in RFC3339 UTC "Zulu" format, with nanosecond resolution and up to nine fractional digits. Examples: "2014-10-02T15:01:23Z" and "2014-10-02T15:01:23.045123456Z".

state enum (State)

Output only. The state of the task.

error object (Status)

Output only. The error that occurred during the state. May be set when the state is any of the non-final state (PENDING/RUNNING/CANCELLING) or FAILED state. If the state is FAILED, the error here is final and not going to be retried. If the state is a non-final state, the error indicates a system-error being retried.

JSON representation 
{
  "updateTime": string,
  "state": enum (State),
  "error": {
    object (Status)
  }
}

ArtifactList

A list of artifact metadata.

Fields
artifacts[] object (Artifact)

Output only. A list of artifact metadata.

JSON representation 
{
  "artifacts": [
    {
      object (Artifact)
    }
  ]
}

RuntimeConfig

The runtime config of a PipelineJob.

Fields
parameters
(deprecated) map (key: string, value: object (Value))

Deprecated. Use RuntimeConfig.parameter_values instead. The runtime parameters of the PipelineJob. The parameters will be passed into PipelineJob.pipeline_spec to replace the placeholders at runtime. This field is used by pipelines built using PipelineJob.pipeline_spec.schema_version 2.0.0 or lower, such as pipelines built using Kubeflow Pipelines SDK 1.8 or lower.

gcsOutputDirectory string

Required. A path in a Cloud Storage bucket, which will be treated as the root output directory of the pipeline. It is used by the system to generate the paths of output artifacts. The artifact paths are generated with a sub-path pattern {job_id}/{taskId}/{outputKey} under the specified output directory. The service account specified in this pipeline must have the storage.objects.get and storage.objects.create permissions for this bucket.

parameterValues map (key: string, value: value (Value format))

The runtime parameters of the PipelineJob. The parameters will be passed into PipelineJob.pipeline_spec to replace the placeholders at runtime. This field is used by pipelines built using PipelineJob.pipeline_spec.schema_version 2.1.0, such as pipelines built using Kubeflow Pipelines SDK 1.9 or higher and the v2 DSL.

failurePolicy enum (PipelineFailurePolicy)

Represents the failure policy of a pipeline. Currently, the default of a pipeline is that the pipeline will continue to run until no more tasks can be executed, also known as PIPELINE_FAILURE_POLICY_FAIL_SLOW. However, if a pipeline is set to PIPELINE_FAILURE_POLICY_FAIL_FAST, it will stop scheduling any new tasks when a task has failed. Any scheduled tasks will continue to completion.

inputArtifacts map (key: string, value: object (InputArtifact))

The runtime artifacts of the PipelineJob. The key will be the input artifact name and the value would be one of the InputArtifact.

defaultRuntime object (DefaultRuntime)

Optional. The default runtime for the PipelineJob. If not set, Standard Vertex Custom Job(https://cloud.google.com/vertex-ai/docs/training/overview) is used as the runtime. If set, all pipeline tasks will run on the default runtime unless a task is a GCPC custom job component (https://cloud.google.com/vertex-ai/docs/pipelines/customjob-component) based task. If the task is based on a GCPC custom job component, it runs solely according to the component's configuration.

JSON representation 
{
  "parameters": {
    string: {
      object (Value)
    },
    ...
  },
  "gcsOutputDirectory": string,
  "parameterValues": {
    string: value,
    ...
  },
  "failurePolicy": enum (PipelineFailurePolicy),
  "inputArtifacts": {
    string: {
      object (InputArtifact)
    },
    ...
  },
  "defaultRuntime": {
    object (DefaultRuntime)
  }
}

Value

value is the value of the field.

Fields
value Union type
value can be only one of the following:
intValue string (int64 format)

An integer value.

doubleValue number

A double value.

stringValue string

A string value.

JSON representation 
{

  // value
  "intValue": string,
  "doubleValue": number,
  "stringValue": string
  // Union type
}

PipelineFailurePolicy

Represents the failure policy of a pipeline. Currently, the default of a pipeline is that the pipeline will continue to run until no more tasks can be executed, also known as PIPELINE_FAILURE_POLICY_FAIL_SLOW. However, if a pipeline is set to PIPELINE_FAILURE_POLICY_FAIL_FAST, it will stop scheduling any new tasks when a task has failed. Any scheduled tasks will continue to completion.

Enums
PIPELINE_FAILURE_POLICY_UNSPECIFIED Default value, and follows fail slow behavior.
PIPELINE_FAILURE_POLICY_FAIL_SLOW Indicates that the pipeline should continue to run until all possible tasks have been scheduled and completed.
PIPELINE_FAILURE_POLICY_FAIL_FAST Indicates that the pipeline should stop scheduling new tasks after a task has failed.

InputArtifact

The type of an input artifact.

Fields
kind Union type
kind can be only one of the following:
artifactId string

Artifact resource id from MLMD. Which is the last portion of an artifact resource name: projects/{project}/locations/{location}/metadataStores/default/artifacts/{artifactId}. The artifact must stay within the same project, location and default metadatastore as the pipeline.

JSON representation 
{

  // kind
  "artifactId": string
  // Union type
}

DefaultRuntime

The default runtime for the PipelineJob.

Fields
runtime_detail Union type
runtime_detail can be only one of the following:
persistentResourceRuntimeDetail object (PersistentResourceRuntimeDetail)

Persistent resource based runtime detail.

JSON representation 
{

  // runtime_detail
  "persistentResourceRuntimeDetail": {
    object (PersistentResourceRuntimeDetail)
  }
  // Union type
}

PersistentResourceRuntimeDetail

Persistent resource based runtime detail. For more information about persistent resource, refer to https://cloud.google.com/vertex-ai/docs/training/persistent-resource-overview

Fields
persistentResourceName string

Persistent resource name. Format: projects/{project}/locations/{location}/persistentResources/{persistentResource}

taskResourceUnavailableWaitTimeMs string (int64 format)

The max time a pipeline task waits for the required CPU, memory, or accelerator resource to become available from the specified persistent resource. Default wait time is 0.

taskResourceUnavailableTimeoutBehavior enum (TaskResourceUnavailableTimeoutBehavior)

Specifies the behavior to take if the timeout is reached.

JSON representation 
{
  "persistentResourceName": string,
  "taskResourceUnavailableWaitTimeMs": string,
  "taskResourceUnavailableTimeoutBehavior": enum (TaskResourceUnavailableTimeoutBehavior)
}

TaskResourceUnavailableTimeoutBehavior

An enum that specifies the behavior to take if the timeout is reached.

Enums
TASK_RESOURCE_UNAVAILABLE_TIMEOUT_BEHAVIOR_UNSPECIFIED Unspecified. Behavior is same as FAIL.
FAIL Fail the task if the timeout is reached.
FALL_BACK_TO_ON_DEMAND Fall back to on-demand execution if the timeout is reached.

PscInterfaceConfig

Configuration for PSC-I.

Fields
networkAttachment string

Optional. The name of the Compute Engine network attachment to attach to the resource within the region and user project. To specify this field, you must have already created a network attachment. This field is only used for resources using PSC-I.

JSON representation 
{
  "networkAttachment": string
}

PipelineTemplateMetadata

Pipeline template metadata if PipelineJob.template_uri is from supported template registry. Currently, the only supported registry is Artifact Registry.

Fields
version string

The versionName in artifact registry.

Will always be presented in output if the PipelineJob.template_uri is from supported template registry.

Format is "sha256:abcdef123456...".

JSON representation 
{
  "version": string
}

PipelineTaskRerunConfig

user provided rerun config to submit a rerun pipelinejob. This includes 1. Which task to rerun 2. user override input parameters and artifacts.

Fields
taskId string (int64 format)

Optional. The system generated id of the task. Retrieved from original run.

taskName string

Optional. The name of the task.

inputs object (Inputs)

Optional. The runtime input of the task overridden by the user.

skipTask boolean

Optional. Whether to skip this task. Default value is False.

skipDownstreamTasks boolean

Optional. Whether to skip downstream tasks. Default value is False.

JSON representation 
{
  "taskId": string,
  "taskName": string,
  "inputs": {
    object (Inputs)
  },
  "skipTask": boolean,
  "skipDownstreamTasks": boolean
}

Inputs

Runtime inputs data of the task.

Fields
artifacts map (key: string, value: object (ArtifactList))

Optional. Input artifacts.

parameterValues map (key: string, value: value (Value format))

Optional. Input parameters.

JSON representation 
{
  "artifacts": {
    string: {
      object (ArtifactList)
    },
    ...
  },
  "parameterValues": {
    string: value,
    ...
  }
}

ArtifactList

A list of artifact metadata.

Fields
artifacts[] object (RuntimeArtifact)

Optional. A list of artifact metadata.

JSON representation 
{
  "artifacts": [
    {
      object (RuntimeArtifact)
    }
  ]
}

RuntimeArtifact

The definition of a runtime artifact.

Fields
name string

The name of an artifact.

type object (ArtifactTypeSchema)

The type of the artifact.

uri string

The URI of the artifact.

properties
(deprecated) map (key: string, value: object (Value))

The properties of the artifact. Deprecated. Use RuntimeArtifact.metadata instead.

customProperties
(deprecated) map (key: string, value: object (Value))

The custom properties of the artifact. Deprecated. Use RuntimeArtifact.metadata instead.

metadata object (Struct format)

Properties of the Artifact.

JSON representation 
{
  "name": string,
  "type": {
    object (ArtifactTypeSchema)
  },
  "uri": string,
  "properties": {
    string: {
      object (Value)
    },
    ...
  },
  "customProperties": {
    string: {
      object (Value)
    },
    ...
  },
  "metadata": {
    object
  }
}

ArtifactTypeSchema

The definition of a artifact type in MLMD.

Fields
schemaVersion string

The schema version of the artifact. If the value is not set, it defaults to the latest version in the system.

kind Union type
kind can be only one of the following:
schemaTitle string

The name of the type. The format of the title must be: <namespace>.<title>. Examples: - aiplatform.Model - acme.CustomModel When this field is set, the type must be pre-registered in the MLMD store.

schemaUri
(deprecated) string

Points to a YAML file stored on Cloud Storage describing the format. Deprecated. Use [PipelineArtifactTypeSchema.schema_title][] or [PipelineArtifactTypeSchema.instance_schema][] instead.

instanceSchema string

Contains a raw YAML string, describing the format of the properties of the type.

JSON representation 
{
  "schemaVersion": string,

  // kind
  "schemaTitle": string,
  "schemaUri": string,
  "instanceSchema": string
  // Union type
}

Methods

batchCancel

 Batch cancel PipelineJobs.

batchDelete

 Batch deletes PipelineJobs The Operation is atomic.

cancel

 Cancels a PipelineJob.

create

 Creates a PipelineJob.

delete

 Deletes a PipelineJob.

get

 Gets a PipelineJob.

list

 Lists PipelineJobs in a Location.