Schedule a pipeline run with scheduler API

Console

Use the following instructions to create a schedule using the Google Cloud console. If a schedule already exists for the project and region, use the instructions in Create a pipeline run .

Use the following instructions to create a pipeline schedule:

1. In the Google Cloud console, in the Vertex AI section, go to the Schedulestab on the Pipelinespage.

   Go to Schedules

2. Click Create scheduled runto open the Create pipeline runpane.

3. Specify the following Run detailsby selecting one of the following options:

   • To create a pipeline run based on an existing pipeline template, click Select from existing pipelinesand enter the following details:

     1. Select the Repositorycontaining the pipeline or component definition file.

     2. Select the Pipeline or componentand Version.

   • To upload a compiled pipeline definition, click Upload fileand enter the following details:

     1. Click Browseto open the file selector. Navigate to the compiled pipeline YAML file that you want to run, select the pipeline, and click Open.

     2. The Pipeline or component nameshows the name specified in the pipeline definition, by default. Optionally, specify a different Pipeline name.

   • To import a pipeline definition file from Cloud Storage, click Import from Cloud Storageand enter the following details:

     1. Click Browseto navigate to the Cloud Storage bucket containing the pipeline definition object, select the file, and then click Select.

     2. Specify the Pipeline or component name.

4. Specify a Run nameto uniquely identify the pipeline run.

5. Specify the Run schedule, as follows:

   1. Select Recurring.

   2. Under Start time, specify the when the schedule becomes active.

      • To schedule the first run to occur immediately upon schedule creation, select Immediately.

      • To schedule the first run to occur at a specific time and date, select On.

   3. In the Frequencyfield, specify the frequency to schedule and execute the pipeline runs, using a cron schedule expression based on unix-cron .

   4. Under Ends, specify when the schedule ends.

      • To indicate that the schedule creates pipeline runs indefinitely, select Never.

      • To indicate that the schedule ends on a specific date and time, select On, and specify the end date and time for the schedule.

6. Optional: To specify a custom service account, a customer-managed encryption key (CMEK), or a peered VPC network, click Advanced optionsand specify a service account, CMEK, or peered VPC network name.

7. Click Continueand specify the Runtime configurationfor the pipeline.

8. Click Submitto create your pipeline run schedule.

REST

To create a pipeline run schedule, send a POST request by using the projects.locations.schedules.create method.

Before using any of the request data, make the following replacements:

• LOCATION : The region where you want to run the pipeline. For more information about the regions where Vertex AI Pipelines is available, see the Vertex AI locations guide .
• PROJECT_ID : The Google Cloud project where you want to run the pipeline.
• DISPLAY_NAME : The name of the pipeline schedule. You can specify a name having a maximum length of 128 UTF-8 characters.
• START_TIME : Timestamp after which the first run can be scheduled, for example, 2045-07-26T00:00:00Z . If you don't specify this parameter, the timestamp corresponding to the date and time when you create the schedule is used as the default value.
• END_TIME : Timestamp after which pipeline runs are no longer scheduled scheduled. After the END_TIME is reached, the state of the schedule changes to COMPLETED . If you don't specify this parameter, then the schedule continues to run new pipeline jobs indefinitely until you pause or delete the schedule.
• CRON_EXPRESSION : Cron schedule expression representing the frequency to schedule and execute pipeline runs. For more information, see cron .
• MAX_CONCURRENT_RUN_COUNT : The maximum number of concurrent runs for the schedule.
• API_REQUEST_TEMPLATE : PipelineService.CreatePipelineJob API request template used to execute the scheduled pipeline runs. For more information about the parameters in the API request template, see the documentation for pipelineJobs.create . Note that you can't specify the pipelineJobId parameter in this template, as the scheduler API doesn't support this parameter.

  Click here to view a sample API request template, where gs://gcs_directory is the Cloud Storage bucket for pipeline output artifacts.

  {
    "parent":"projects/ /locations/us-central1",
    "pipelineJob": {
      "displayName": "hello-world",
      "pipelineSpec": {
        "deploymentConfig": {
          "@type": "type.googleapis.com/ml_pipelines.PipelineDeploymentConfig",
          "executors": {
            "HelloWorld_executor": {
              "container": {
                "command": ["sleep", "1"],
                "image": "google/cloud-sdk:latest"
              }
            }
          }
        },
        "pipelineInfo": {
          "name": "hello-world"
        },
        "root": {
          "dag": {
            "tasks": {
              "task-test-hello-world": {
                "taskInfo": {"name": "HelloWorld"},
                "componentRef": {"name": "HelloWorld"},
                "cachingOptions": {}
              }
            }
          }
        },
        "components": {
          "HelloWorld": {
            "executorLabel": "HelloWorld_executor"
          }
        },
        "schemaVersion": "2.0.0"
      },
      "runtimeConfig": {
        "gcsOutputDirectory": "gs://gcs_directory"
      }
    }
  } 
  

HTTP method and URL:

POST https:// LOCATION 
-aiplatform.googleapis.com/v1/projects/ PROJECT_ID 
/locations/ LOCATION 
/schedules

Request JSON body:

{
  "display_name":" DISPLAY_NAME 
",
  "start_time": " START_TIME 
",
  "end_time": " END_TIME 
",
  "cron": " CRON_EXPRESSION 
",
  "max_concurrent_run_count": " MAX_CONCURRENT_RUN_COUNT 
",
  "create_pipeline_job_request": API_REQUEST_TEMPLATE 
}

To send your request, choose one of these options:

curl -X POST \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "Content-Type: application/json; charset=utf-8" \
-d @request.json \
"https:// LOCATION 
-aiplatform.googleapis.com/v1/projects/ PROJECT_ID 
/locations/ LOCATION 
/schedules"

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred" }

Invoke-WebRequest `
-Method POST `
-Headers $headers `
-ContentType: "application/json; charset=utf-8" `
-InFile request.json `
-Uri "https:// LOCATION 
-aiplatform.googleapis.com/v1/projects/ PROJECT_ID 
/locations/ LOCATION 
/schedules" | Select-Object -Expand Content

You should see output similar to the following. You can use the SCHEDULE_ID from the response to retrieve, pause, resume, or delete the schedule. PIPELINE_JOB_CREATION_REQUEST represents the API request to create the pipeline job.

{
  "name": "projects/ PROJECT_ID 
/locations/ LOCATION 
/schedules/ SCHEDULE_ID 
",
  "displayName": " DISPLAY_NAME 
",
  "startTime": " START_TIME 
",
  "state": "ACTIVE",
  "createTime": "2025-01-01T00:00:00.000000Z",
  "nextRunTime": "2045-08-01T00:00:00Z",
  "cron": " CRON_EXPRESSION 
",
  "maxConcurrentRunCount": " MAX_CONCURRENT_RUN_COUNT 
",
  "createPipelineJobRequest": PIPELINE_JOB_CREATION_REQUEST 
}

Python

You can create a pipeline run schedule in the following ways:

• Create a schedule based on a PipelineJob using the PipelineJob.create_schedule method.

• Creating a schedule using the PipelineJobSchedule.create method.

While creating a pipeline run schedule, you can also pass the following placeholders supported by the KFP SDK as inputs:

• {{$.pipeline_job_name_placeholder}}

• {{$.pipeline_job_resource_name_placeholder}}

• {{$.pipeline_job_id_placeholder}}

• {{$.pipeline_task_name_placeholder}}

• {{$.pipeline_task_id_placeholder}}

• {{$.pipeline_job_create_time_utc_placeholder}}

• {{$.pipeline_job_schedule_time_utc_placeholder}}

• {{$.pipeline_root_placeholder}}

For more information, see Special input types in the Kubeflow Pipelines v2 documentation .

Create a schedule from a PipelineJob

Use the following sample to schedule pipeline runs using the PipelineJob.create_schedule method:

  from 
  
 google.cloud 
  
 import 
 aiplatform 
 pipeline_job 
 = 
 aiplatform 
 . 
 PipelineJob 
 ( 
 template_path 
 = 
 " COMPILED_PIPELINE_PATH 
" 
 , 
 pipeline_root 
 = 
 " PIPELINE_ROOT_PATH 
" 
 , 
 display_name 
 = 
 " DISPLAY_NAME 
" 
 , 
 ) 
 pipeline_job_schedule 
 = 
 pipeline_job 
 . 
 create_schedule 
 ( 
 display_name 
 = 
 " SCHEDULE_NAME 
" 
 , 
 cron 
 = 
 "TZ= CRON 
" 
 , 
 max_concurrent_run_count 
 = 
  MAX_CONCURRENT_RUN_COUNT 
 
 , 
 max_run_count 
 = 
  MAX_RUN_COUNT 
 
 , 
 ) 
 

• COMPILED_PIPELINE_PATH : The path to your compiled pipeline YAML file. It can be a local path or a Cloud Storage URI.

  Optional: To specify a particular version of a template, include the version tag along with the path in any one of the following formats:

  • COMPILED_PIPELINE_PATH : TAG , where TAG is the version tag.

  • COMPILED_PIPELINE_PATH @ SHA256_TAG , where SHA256_TAG is the sha256 hash value of the pipeline version.

• PIPELINE_ROOT_PATH : ( optional ) To override the pipeline root path specified in the pipeline definition, specify a path that your pipeline job can access, such as a Cloud Storage bucket URI.

• DISPLAY_NAME : The name of the pipeline. This will show up in the Google Cloud console.

• SCHEDULE_NAME : The name of the pipeline schedule. You can specify a name having a maximum length of 128 UTF-8 characters.

• CRON : Cron schedule expression representing the frequency to schedule and execute pipeline runs. For more information, see Cron .

• MAX_CONCURRENT_RUN_COUNT : The maximum number of concurrent runs for the schedule.

• MAX_RUN_COUNT : The maximum number of pipeline runs that the schedule creates after which it's completed.

Create a schedule using PipelineJobSchedule.create

Use the following sample to schedule pipeline runs using the PipelineJobSchedule.create method:

  from 
  
 google.cloud 
  
 import 
 aiplatform 
 pipeline_job 
 = 
 aiplatform 
 . 
 PipelineJob 
 ( 
 template_path 
 = 
 " COMPILED_PIPELINE_PATH 
" 
 , 
 pipeline_root 
 = 
 " PIPELINE_ROOT_PATH 
" 
 , 
 display_name 
 = 
 " DISPLAY_NAME 
" 
 , 
 ) 
 pipeline_job_schedule 
 = 
 aiplatform 
 . 
 PipelineJobSchedule 
 ( 
 pipeline_job 
 = 
 pipeline_job 
 , 
 display_name 
 = 
 " SCHEDULE_NAME 
" 
 ) 
 pipeline_job_schedule 
 . 
 create 
 ( 
 cron 
 = 
 "TZ= CRON 
" 
 , 
 max_concurrent_run_count 
 = 
  MAX_CONCURRENT_RUN_COUNT 
 
 , 
 max_run_count 
 = 
  MAX_RUN_COUNT 
 
 , 
 ) 
 

• COMPILED_PIPELINE_PATH : The path to your compiled pipeline YAML file. It can be a local path or a Cloud Storage URI.

  Optional: To specify a particular version of a template, include the version tag along with the path in any one of the following formats:

  • COMPILED_PIPELINE_PATH : TAG , where TAG is the version tag.

  • COMPILED_PIPELINE_PATH @ SHA256_TAG , where SHA256_TAG is the sha256 hash value of the pipeline version.

• PIPELINE_ROOT_PATH : ( optional ) To override the pipeline root path specified in the pipeline definition, specify a path that your pipeline job can access, such as a Cloud Storage bucket URI.

• DISPLAY_NAME : The name of the pipeline. This will show up in the Google Cloud console.

• SCHEDULE_NAME : The name of the pipeline schedule. You can specify a name having a maximum length of 128 UTF-8 characters.

• CRON : Cron schedule expression representing the frequency to schedule and execute pipeline runs. For more information, see Cron .

• MAX_CONCURRENT_RUN_COUNT : The maximum number of concurrent runs for the schedule.

• MAX_RUN_COUNT : The maximum number of pipeline runs that the schedule creates after which it's completed.