Create datastores

APPLIES TO: Azure CLI ml extension v2 (current) Python SDK azure-ai-ml v2 (current)

In this article, you learn how to connect to Azure data storage services with Azure Machine Learning datastores.

Prerequisites

• An Azure subscription. If you don't have an Azure subscription, create a free account before you begin. Try the free or paid version of Azure Machine Learning.
• The Azure Machine Learning SDK for Python.
• A Machine Learning workspace.

Note

Machine Learning datastores do not create the underlying storage account resources. Instead, they link an existing storage account for Machine Learning use. Machine Learning datastores aren't required. If you have access to the underlying data, you can use storage URIs directly.

Create an Azure Blob datastore

Python SDK: Identity-based access

from azure.ai.ml.entities import AzureBlobDatastore
from azure.ai.ml import MLClient

ml_client = MLClient.from_config()

store = AzureBlobDatastore(
    name="",
    description="",
    account_name="",
    container_name=""
)

ml_client.create_or_update(store)

Python SDK: Account key

from azure.ai.ml.entities import AzureBlobDatastore
from azure.ai.ml.entities import AccountKeyConfiguration
from azure.ai.ml import MLClient

ml_client = MLClient.from_config()

store = AzureBlobDatastore(
    name="blob_protocol_example",
    description="Datastore pointing to a blob container using https protocol.",
    account_name="mytestblobstore",
    container_name="data-container",
    protocol="https",
    credentials=AccountKeyConfiguration(
        account_key="XXXxxxXXXxXXXXxxXXXXXxXXXXXxXxxXxXXXxXXXxXXxxxXXxxXXXxXxXXXxxXxxXXXXxxxxxXXxxxxxxXXXxXXX"
    ),
)

ml_client.create_or_update(store)

Python SDK: SAS

from azure.ai.ml.entities import AzureBlobDatastore
from azure.ai.ml.entities import SasTokenConfiguration
from azure.ai.ml import MLClient

ml_client = MLClient.from_config()

store = AzureBlobDatastore(
    name="blob_sas_example",
    description="Datastore pointing to a blob container using SAS token.",
    account_name="mytestblobstore",
    container_name="data-container",
    credentials=SasTokenConfiguration(
        sas_token= "?xx=XXXX-XX-XX&xx=xxxx&xxx=xxx&xx=xxxxxxxxxxx&xx=XXXX-XX-XXXXX:XX:XXX&xx=XXXX-XX-XXXXX:XX:XXX&xxx=xxxxx&xxx=XXxXXXxxxxxXXXXXXXxXxxxXXXXXxxXXXXXxXXXXxXXXxXXxXX"
    ),
)

ml_client.create_or_update(store)

CLI: Identity-based access

Create the following YAML file (update the appropriate values):

# my_blob_datastore.yml
$schema: https://azuremlschemas.azureedge.net/latest/azureBlob.schema.json
name: my_blob_ds # add your datastore name here
type: azure_blob
description: here is a description # add a datastore description here
account_name: my_account_name # add the storage account name here
container_name: my_container_name # add the storage container name here

Create the Machine Learning datastore in the Azure CLI:

az ml datastore create --file my_blob_datastore.yml

CLI: Account key

Create this YAML file (update the appropriate values):

# my_blob_datastore.yml
$schema: https://azuremlschemas.azureedge.net/latest/azureBlob.schema.json
name: blob_example
type: azure_blob
description: Datastore pointing to a blob container.
account_name: mytestblobstore
container_name: data-container
credentials:
  account_key: XXXxxxXXXxXXXXxxXXXXXxXXXXXxXxxXxXXXxXXXxXXxxxXXxxXXXxXxXXXxxXxxXXXXxxxxxXXxxxxxxXXXxXXX

Create the Machine Learning datastore in the CLI:

az ml datastore create --file my_blob_datastore.yml

CLI: SAS

Create this YAML file (update the appropriate values):

# my_blob_datastore.yml
$schema: https://azuremlschemas.azureedge.net/latest/azureBlob.schema.json
name: blob_sas_example
type: azure_blob
description: Datastore pointing to a blob container using SAS token.
account_name: mytestblobstore
container_name: data-container
credentials:
  sas_token: ?xx=XXXX-XX-XX&xx=xxxx&xxx=xxx&xx=xxxxxxxxxxx&xx=XXXX-XX-XXXXX:XX:XXX&xx=XXXX-XX-XXXXX:XX:XXX&xxx=xxxxx&xxx=XXxXXXxxxxxXXXXXXXxXxxxXXXXXxxXXXXXxXXXXxXXXxXXxXX

Create the Machine Learning datastore in the CLI:

az ml datastore create --file my_blob_datastore.yml

Create an Azure Data Lake Storage Gen2 datastore

Python SDK: Identity-based access

from azure.ai.ml.entities import AzureDataLakeGen2Datastore
from azure.ai.ml import MLClient

ml_client = MLClient.from_config()

store = AzureDataLakeGen2Datastore(
    name="",
    description="",
    account_name="",
    filesystem=""
)

ml_client.create_or_update(store)

Python SDK: Service principal

from azure.ai.ml.entities import AzureDataLakeGen2Datastore
from azure.ai.ml.entities._datastore.credentials import ServicePrincipalCredentials

from azure.ai.ml import MLClient

ml_client = MLClient.from_config()

store = AzureDataLakeGen2Datastore(
    name="adls_gen2_example",
    description="Datastore pointing to an Azure Data Lake Storage Gen2.",
    account_name="mytestdatalakegen2",
    filesystem="my-gen2-container",
     credentials=ServicePrincipalCredentials(
        tenant_id= "XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX",
        client_id= "XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX",
        client_secret= "XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX",
    ),
)

ml_client.create_or_update(store)

CLI: Identity-based access

Create this YAML file (update the values):

# my_adls_datastore.yml
$schema: https://azuremlschemas.azureedge.net/latest/azureDataLakeGen2.schema.json
name: adls_gen2_credless_example
type: azure_data_lake_gen2
description: Credential-less datastore pointing to an Azure Data Lake Storage Gen2 instance.
account_name: mytestdatalakegen2
filesystem: my-gen2-container

Create the Machine Learning datastore in the CLI:

az ml datastore create --file my_adls_datastore.yml

CLI: Service principal

Create this YAML file (update the values):

# my_adls_datastore.yml
$schema: https://azuremlschemas.azureedge.net/latest/azureDataLakeGen2.schema.json
name: adls_gen2_example
type: azure_data_lake_gen2
description: Datastore pointing to an Azure Data Lake Storage Gen2 instance.
account_name: mytestdatalakegen2
filesystem: my-gen2-container
credentials:
  tenant_id: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
  client_id: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
  client_secret: XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

Create the Machine Learning datastore in the CLI:

az ml datastore create --file my_adls_datastore.yml

Create an Azure Files datastore

Python SDK: Account key

from azure.ai.ml.entities import AzureFileDatastore
from azure.ai.ml.entities import AccountKeyConfiguration
from azure.ai.ml import MLClient

ml_client = MLClient.from_config()

store = AzureFileDatastore(
    name="file_example",
    description="Datastore pointing to an Azure File Share.",
    account_name="mytestfilestore",
    file_share_name="my-share",
    credentials=AccountKeyConfiguration(
        account_key= "XXXxxxXXXxXXXXxxXXXXXxXXXXXxXxxXxXXXxXXXxXXxxxXXxxXXXxXxXXXxxXxxXXXXxxxxxXXxxxxxxXXXxXXX"
    ),
)

ml_client.create_or_update(store)

Python SDK: SAS

from azure.ai.ml.entities import AzureFileDatastore
from azure.ai.ml.entities import SasTokenConfiguration
from azure.ai.ml import MLClient

ml_client = MLClient.from_config()

store = AzureFileDatastore(
    name="file_sas_example",
    description="Datastore pointing to an Azure File Share using SAS token.",
    account_name="mytestfilestore",
    file_share_name="my-share",
    credentials=SasTokenConfiguration(
        sas_token="?xx=XXXX-XX-XX&xx=xxxx&xxx=xxx&xx=xxxxxxxxxxx&xx=XXXX-XX-XXXXX:XX:XXX&xx=XXXX-XX-XXXXX:XX:XXX&xxx=xxxxx&xxx=XXxXXXxxxxxXXXXXXXxXxxxXXXXXxxXXXXXxXXXXxXXXxXXxXX"
    ),
)

ml_client.create_or_update(store)

CLI: Account key

Create this YAML file (update the values):

# my_files_datastore.yml
$schema: https://azuremlschemas.azureedge.net/latest/azureFile.schema.json
name: file_example
type: azure_file
description: Datastore pointing to an Azure File Share.
account_name: mytestfilestore
file_share_name: my-share
credentials:
  account_key: XxXxXxXXXXXXXxXxXxxXxxXXXXXXXXxXxxXXxXXXXXXXxxxXxXXxXXXXXxXXxXXXxXxXxxxXXxXXxXXXXXxXxxXX

Create the Machine Learning datastore in the CLI:

az ml datastore create --file my_files_datastore.yml

CLI: SAS

Create this YAML file (update the values):

# my_files_datastore.yml
$schema: https://azuremlschemas.azureedge.net/latest/azureFile.schema.json
name: file_sas_example
type: azure_file
description: Datastore pointing to an Azure File Share using an SAS token.
account_name: mytestfilestore
file_share_name: my-share
credentials:
  sas_token: ?xx=XXXX-XX-XX&xx=xxxx&xxx=xxx&xx=xxxxxxxxxxx&xx=XXXX-XX-XXXXX:XX:XXX&xx=XXXX-XX-XXXXX:XX:XXX&xxx=xxxxx&xxx=XXxXXXxxxxxXXXXXXXxXxxxXXXXXxxXXXXXxXXXXxXXXxXXxXX

Create the Machine Learning datastore in the CLI:

az ml datastore create --file my_files_datastore.yml

Create an Azure Data Lake Storage Gen1 datastore

Python SDK: Identity-based access

from azure.ai.ml.entities import AzureDataLakeGen1Datastore
from azure.ai.ml import MLClient

ml_client = MLClient.from_config()

store = AzureDataLakeGen1Datastore(
    name="",
    store_name="",
    description="",
)

ml_client.create_or_update(store)

Python SDK: Service principal

from azure.ai.ml.entities import AzureDataLakeGen1Datastore
from azure.ai.ml.entities._datastore.credentials import ServicePrincipalCredentials
from azure.ai.ml import MLClient

ml_client = MLClient.from_config()

store = AzureDataLakeGen1Datastore(
    name="adls_gen1_example",
    description="Datastore pointing to an Azure Data Lake Storage Gen1.",
    store_name="mytestdatalakegen1",
    credentials=ServicePrincipalCredentials(
        tenant_id= "XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX",
        client_id= "XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX",
        client_secret= "XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX",
    ),
)

ml_client.create_or_update(store)

CLI: Identity-based access

Create this YAML file (update the values):

# my_adls_datastore.yml
$schema: https://azuremlschemas.azureedge.net/latest/azureDataLakeGen1.schema.json
name: alds_gen1_credless_example
type: azure_data_lake_gen1
description: Credential-less datastore pointing to an Azure Data Lake Storage Gen1 instance.
store_name: mytestdatalakegen1

Create the Machine Learning datastore in the CLI:

az ml datastore create --file my_adls_datastore.yml

CLI: Service principal

Create this YAML file (update the values):

# my_adls_datastore.yml
$schema: https://azuremlschemas.azureedge.net/latest/azureDataLakeGen1.schema.json
name: adls_gen1_example
type: azure_data_lake_gen1
description: Datastore pointing to an Azure Data Lake Storage Gen1 instance.
store_name: mytestdatalakegen1
credentials:
  tenant_id: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
  client_id: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
  client_secret: XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

Create the Machine Learning datastore in the CLI:

az ml datastore create --file my_adls_datastore.yml

Create a OneLake (Microsoft Fabric) datastore (preview)

This section describes various options to create a OneLake datastore. The OneLake datastore is part of Microsoft Fabric. At this time, Machine Learning supports connection to Microsoft Fabric lakehouse artifacts in "Files" folder that include folders or files and Amazon S3 shortcuts. For more information about lakehouses, see What is a lakehouse in Microsoft Fabric?.

OneLake datastore creation requires the following information from your Microsoft Fabric instance:

• Endpoint
• Workspace GUID
• Artifact GUID

The following screenshots describe the retrieval of these required information resources from your Microsoft Fabric instance.

You will then find "Endpoint", "Workspace GUID" and "Artifact GUID" in "URL" and "ABFS path" from the "Properties" page:

• URL format: https://{your_one_lake_endpoint}/{your_one_lake_workspace_guid}/{your_one_lake_artifact_guid}/Files
• ABFS path format: abfss://{your_one_lake_workspace_guid}@{your_one_lake_endpoint}/{your_one_lake_artifact_guid}/Files

Create a OneLake datastore

Python SDK: Identity-based access

from azure.ai.ml.entities import OneLakeDatastore, OneLakeArtifact
from azure.ai.ml import MLClient

ml_client = MLClient.from_config()

store = OneLakeDatastore(
    name="onelake_example_id",
    description="Datastore pointing to an Microsoft fabric artifact.",
    one_lake_workspace_name="XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX", #{your_one_lake_workspace_guid}
    endpoint="msit-onelake.dfs.fabric.microsoft.com" #{your_one_lake_endpoint}
    artifact = OneLakeArtifact(
        name="XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX/Files", #{your_one_lake_artifact_guid}/Files
        type="lake_house"
    )
)

ml_client.create_or_update(store)

Python SDK: Service principal

from azure.ai.ml.entities import AzureDataLakeGen1Datastore
from azure.ai.ml.entities._datastore.credentials import ServicePrincipalCredentials
from azure.ai.ml import MLClient

ml_client = MLClient.from_config()

rom azure.ai.ml.entities import OneLakeDatastore, OneLakeArtifact
from azure.ai.ml import MLClient

ml_client = MLClient.from_config()

store = OneLakeDatastore(
    name="onelake_example_sp",
    description="Datastore pointing to an Microsoft fabric artifact.",
    one_lake_workspace_name="XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX", #{your_one_lake_workspace_guid}
    endpoint="msit-onelake.dfs.fabric.microsoft.com" #{your_one_lake_endpoint}
    artifact = OneLakeArtifact(
    name="XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX/Files", #{your_one_lake_artifact_guid}/Files
    type="lake_house"
    )
    credentials=ServicePrincipalCredentials(
        tenant_id= "XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX",
        client_id= "XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX",
        client_secret= "XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX",
    ),
)

ml_client.create_or_update(store)

CLI: Identity-based access

Create the following YAML file (update the values):

# my_onelake_datastore.yml
$schema: http://azureml/sdk-2-0/OneLakeDatastore.json
name: onelake_example_id
type: one_lake
description: Credential-less datastore pointing to a OneLake lakehouse.
one_lake_workspace_name: "XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX"
endpoint: "msit-onelake.dfs.fabric.microsoft.com"
artifact:
  type: lake_house
  name: "XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX/Files"

Create the Machine Learning datastore in the CLI:

az ml datastore create --file my_onelake_datastore.yml

CLI: Service principal

Create the following YAML file (update the values):

# my_onelakesp_datastore.yml
$schema: http://azureml/sdk-2-0/OneLakeDatastore.json
name: onelake_example_id
type: one_lake
description: Credential-less datastore pointing to a OneLake lakehouse.
one_lake_workspace_name: "XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX"
endpoint: "msit-onelake.dfs.fabric.microsoft.com"
artifact:
  type: lake_house
  name: "XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX/Files"
credentials:
  tenant_id: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
  client_id: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
  client_secret: XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

Create the Machine Learning datastore in the CLI:

az ml datastore create --file my_onelakesp_datastore.yml

Next steps

• Access data in a job
• Create and manage data assets
• Import data assets (preview)
• Data administration