Using the Python SDK

The IBM Analytics Engine SDK can be installed by installing the library iaesdk from the Python Package Index.

Type the following command into a command line:

pip install --upgrade "iaesdk>=1.1.1"

Source code can be found at GitHub. The iaesdk library provides complete access to the IBM Analytics Engine API.

You need to provide the service endpoints and the API key when you create a IBM Analytics Engine service resource or a low-level client.

The service instance ID is also referred to as a instance GUID. You can retrieve the service instance ID when you create service credentials or through the CLI. See Retrieving service endpoints.

To use the iaesdk library, you need the following values:

IAM_API_KEY: The API key generated when creating the service credentials. You can retrieve by viewing the service credentials on the IBM Cloud dashboard.
instance_guid: The value in resource_instance_id generated when the service credentials are created. You can retrieve by viewing the service credentials on the IBM Cloud dashboard.
IAE_ENDPOINT_URL: The service endpoint URL including the https:// protocol. See Service endpoints.

Code samples using `iaesdk`

Getting started with the Python SDK after you have installed it, involves sourcing credentials to the IBM Analytics Engine service, invoking the service and then issuing different cluster commands as shown in the following sample code snippets. The code examples are written for Python 3.7.

In addition to the sample code snippets in this topic, you can work with Python code samples from the IBM Analytics Engine V3 API reference.

The following code samples show you how to:

Authenticate to the IBM Analytics Engine service and build a service client:

from iaesdk import IbmAnalyticsEngineApiV3
from ibm_cloud_sdk_core.authenticators import IAMAuthenticator
import json

# Constants for IBM Analytics Engine values
IAM_API_KEY = "{apikey}" # eg "W00YiRnLW4a3fTjMB-odB-2ySfTrFBIQQWanc--P3byk"
IAE_ENDPOINT_URL = "{url}" # Current list avaiable at https://cloud.ibm.com/apidocs/ibm-analytics-engine#service-endpoints
const API_AUTH_URL = "{api auth url}"  # "https://iam.cloud.ibm.com/identity/token"

# Create an IAM authenticator.
authenticator = IAMAuthenticator(apikey=IAM_API_KEY, url=API_AUTH_URL)

# Construct the service client.
ibm_analytics_engine_api_service = IbmAnalyticsEngineApiV3(authenticator=authenticator)

# Set our custom service URL
ibm_analytics_engine_api_service.set_service_url(IAE_ENDPOINT_URL)

# Service operations can now be invoked using the "ibm_analytics_engine_api_service" variable.

Retrieve the details of a single instance:

get_instance(self,
    instance_id: str,
    **kwargs
) -> DetailedResponse

Example request:

instance = ibm_analytics_engine_api_service.get_instance(
instance_id='dc0e9889-eab2-4t9e-9441-566209499546'
).get_result()
print(json.dumps(instance, indent=2))

Deploy a Spark application on a given serverless Spark instance:

create_application(self,
    instance_id: str,
    *,
    application_details: 'ApplicationRequestApplicationDetails' = None,
    **kwargs
) -> DetailedResponse

Example request:

application_request_application_details_model = {
  'application': '/opt/ibm/spark/examples/src/main/python/wordcount.py',
  'arguments': ['/opt/ibm/spark/examples/src/main/resources/people.txt']
  }
create_application_response = ibm_analytics_engine_api_service.create_application(
  instance_id='dc0e9889-eab2-4t9e-9441-566209499546',
  application_details=application_request_application_details_model
).get_result()
print(json.dumps(create_application_response, indent=2))

Retrieve all Spark applications run on a given instance:

list_applications(self,
    instance_id: str,
    **kwargs
) -> DetailedResponse

Example request:

list_applications_response = ibm_analytics_engine_api_service.list_applications(
  instance_id='dc0e9889-eab2-4t9e-9441-566209499546'
  ).get_result()
print(json.dumps(list_applications_response, indent=2))

Retrieve the details of a given Spark application:

get_application(self,
    instance_id: str,
    application_id: str,
    **kwargs
) -> DetailedResponse

Example request:

get_application_response = ibm_analytics_engine_api_service.get_application(
  instance_id='dc0e9889-eab2-4t9e-9441-566209499546',
  application_id='db933645-0b68-4dcb-80d8-7b71a6c8e542'
  ).get_result()
print(json.dumps(get_application_response, indent=2))

Stop a running application identified by the app_id identifier. This is an idempotent operation. Performs no action if the requested application is already stopped or completed.

delete_application(self,
    instance_id: str,
    application_id: str,
    **kwargs
) -> DetailedResponse

Example request:

delete_application_by_id_response = ibm_analytics_engine_api_service.delete_application(
  instance_id='dc0e9889-eab2-4t9e-9441-566209499546',
  application_id='db933645-0b68-4dcb-80d8-7b71a6c8e542'
  ).get_result()
print(json.dumps(delete_application_by_id_response, indent=2))

Return the state of the application identified by the app_id identifier:

get_application_state(self,
    instance_id: str,
    application_id: str,
    **kwargs
) -> DetailedResponse

Example request:

get_application_state_response = ibm_analytics_engine_api_service.get_application_state(
  instance_id='dc0e9889-eab2-4t9e-9441-566209499546',
  application_id='db933645-0b68-4dcb-80d8-7b71a6c8e542'
  ).get_result()
print(json.dumps(get_application_state_response, indent=2))

Using the Python SDK

Code samples using iaesdk

Code samples using `iaesdk`