Using Blob storage

If you want to upload large files (>10 GB) to the workspace, it is advised to use Blob storage rather than the File system. Blob storage is optimized for storing massive amounts of unstructured data such as large images, videos, executable files or genomic data. Blob storage is ideal for raw data analysis and information extraction tasks.

Access constraints

Before you choose the best storage type for your files, you should be aware of some Blob storage access constraints.

Files uploaded through Blob storage are mainly accessed and analysed within the Virtual Machine. Currently, access to workspace Blobs from a Virtual Machine is not enabled by default. If you require this, please contact the Service Desk.

It is also possible to use standard tools for performing actions on files or folders stored in Blobs, like converting CSV files to data tables, using Airlock, Data Table Editor and Medical Statistics Core functions.

However, files stored in Blobs cannot be accessed from the built-in R console and Mini-apps. We are currently working to resolve this but in the meantime, you can use R in the Virtual Machine for analysing these files.

Uploading files to Blobs

The file upload to Blob storage follows the process described in uploading files via the workspace article. Note that due to the nature of Blob storage, folder hierarchies cannot exist without content. This means that you won't be able to create empty folders, and after refreshing the page the empty folders will be gone from your Blob storage. There is a workaround: you can create an empty folder, and without closing the window, add or upload a new file to the folder.