Encryption of output blobs in Azure ML Studio RRS feed

  • Question

  • Hey all,

      Azure doesn't appear to have support for transparent (ie. where Azure manages the keys) encryption of data at rest with Blob storage . Most of the articles I've read point to encrypting it yourself before saving. That's fine for a regular app but it's really annoying in an Azure ML Studio experiment.

    Given there isn't an encrypt with key option of the Writer Module, it appears I'd have to do it in Python or R cell by cell first.

    Another option is to not use blob storage but store the data back into an Azure SQL database with TDE enabled but that is a problematic for a number of reasons; not the least of which is that the table structure is highly volatile during experiment design stages. 

    Any ideas for better options?

    ----- Ed

    Monday, August 3, 2015 7:05 PM


  • To enable our enterprise customers, encryption is certainly something that we are working on. I don't have a timeline, but it is on the backlog.

    Dan Manrique Senior Program Manager - Azure Machine Learning

    • Proposed as answer by neerajkh_MSFT Monday, August 17, 2015 6:26 AM
    • Marked as answer by DigitalEdSC Wednesday, August 19, 2015 7:48 PM
    Wednesday, August 12, 2015 8:58 PM