Microsoft has made available a preview of an Azure SDK for Python (Conda) for data scientists working with the Python language and Azure cloud services.
Available now in the Microsoft channel, the packages are a collection of open-source libraries to simplify provisioning, managing, and use of Azure resources from a developer’s Python application code. They are currently available as a preview release, but Microsoft says they will be stable in the near future.
Announcing the release on its developer blog, Microsoft reckons Conda is the most popular platform for scientific computing used by data scientists. It uses a separate packaging system from the standard Python Package Index (PyPI) optimised for self-contained packages, guarantees dependency enforcement, and provides built-in execution environment isolation.
Developers need to be running either the Miniconda build, which delivers minimal installation of Conda, or the Anaconda build.
Azure SDK for Python (Conda) packages are divided into a number of client libraries that serve different purposes, organised by services such as storage. Choosing the storage package will install all storage-related libraries including azure-storage-blob, azure-storage-queue, azure-storage-file-share and azure-storage-file-datalake.
Other client libraries available include the Azure App Configuration library, a managed service that is designed to let developers centralise their application configurations in a single place, a much needed tool for distributed applications running in the cloud.
The Azure Cosmos client library provides developers with access to Microsoft’s scale-out NoSQL database service, enabling them to create and manage databases and the JSON documents they contain, as well as query documents in a database using SQL-like syntax.
Another library allows developers to access Azure Event Hubs, Microsoft’s data streaming and event ingestion service for Azure. The Azure Service Bus client library provides access to Azure’s messaging service for real-time communication between distributed application and service components.
Developers and data scientists interested in using Conda with Azure services can check out full details of the Azure SDK for Python (Conda) Preview on the Microsoft developer blog.