The pattern used to make changes to Azure storage resources has been deprecated and no replacement or desired pattern has been described.

Greetings,

There are several other issues related to this topic but none seem to drive home the main point. Dataset.File.upload_directory() is not fully a replacement for Datastore.upload() because it does not allow data to propagate from inside an ML Studio workspace to Azure storage resources (like Blob storage). I understand if this is now an anti-pattern, but it seems odd that this functionality has been removed with no apparent public discussion of what the desired pattern for propagating changes upstream from an ML environment is. Lacking documentation about the desired pattern leads to unproductive (and wasteful) speculation about what is the appropriate method to make changes in Azure storage resources from within ML Studio.

For example:
* Is this a temporary change and this functionality will return? https://github.com/Azure/MachineLearningNotebooks/issues/1672
* Are we supposed to use the AZ cli to make changes to these resources? https://github.com/MicrosoftDocs/azure-docs/issues/84419
* Will writes to db (which I haven't even verified still work) or exports to storage resources via pipelines (I'm guessing through OutputFileDatasetConfig) continue to be supported or are these also anti-patterns?

Apologies if I have missed something, and I won't argue why I think this functionality should be supported, but it would be very helpful if you could describe the supported pattern for migrating data changes, additions, and deletions to storage resources from within Azure ML Studio.

---
#### Document Details

⚠ *Do not edit this section. It is required for docs.microsoft.com ➟ GitHub issue linking.*

* ID: b43f5598-ed69-4aad-9620-5b663902e1e7
* Version Independent ID: 93ee43dd-326c-2ef5-b027-e86a1bf30d49
* Content: [azureml.data.azure_storage_datastore.AzureBlobDatastore class - Azure Machine Learning Python](https://docs.microsoft.com/en-us/python/api/azureml-core/azureml.data.azure_storage_datastore.azureblobdatastore?view=azure-ml-py)
* Content Source: [AzureML-Docset/stable/docs-ref-autogen/azureml-core/azureml.data.azure_storage_datastore.AzureBlobDatastore.yml](https://github.com/MicrosoftDocs/MachineLearning-Python-pr/blob/live/AzureML-Docset/stable/docs-ref-autogen/azureml-core/azureml.data.azure_storage_datastore.AzureBlobDatastore.yml)
* Service: **machine-learning**
* Sub-service: **core**
* GitHub Login: @DebFro
* Microsoft Alias: **debfro**

Comece por AzureML-Docset/stable/docs-ref-autogen/azureml-core/azureml.data.azure_storage_datastore.AzureBlobDatastore.yml e pela documentação da classe AzureBlobDatastore. Revise o comportamento documentado de Dataset.File.upload_directory(), Datastore.upload() e OutputFileDatasetConfig, juntamente com as issues vinculadas. A tarefa estará concluída quando o padrão compatível para propagar alterações de dados do Azure ML Studio para o Azure storage estiver claramente documentado.

Escrita pelo modelo de indexação a partir do texto da issue.