Migrating file shares


We considered two main options for cloud-hosting the data on the file shares. Generally, cloud hosting the data comes at a cost. Usually, providers charge by a sum per GB rate, but often it is differentiated between data at rest and data which is moved, accessed, or changed frequently.

If we look back at our original goal - to migrate our historically grown on-prem file structure spanning over multiple locations to the cloud and integrating, it into a cloud-first target structure – we certainly made good progress by finishing the migration of content lying on SharePoint 2019 to SharePoint Online. But as mentioned at the beginning of the series, there were still files and documents left on file servers, which we wanted to move to the cloud.

We considered two main options for cloud-hosting the data on the file shares. Generally, cloud hosting the data comes at a cost. Usually, providers charge by a sum per GB rate, but often it is differentiated between data at rest and data which is moved, accessed, or changed frequently.

 First, the data can be uploaded to SharePoint Online. This holds the advantages that all the data is stored in the same system in the cloud. Storing file share data on SPO involves careful planning and preparation, since a site and library architecture must be created, and access rights must be properly assigned. Additionally, SharePoint Online does not allow mapping a document library to the network drive.

Second, the data can be stored on Azure Files, which behaves as an online file server or file share. When set up correctly, users can map the file share to the file explorer as before with the on-prem file share and they do not notice a difference if they have an internet connection. Cost-wise, Azure files is the cheaper solution considering the prices per gigabyte.

In our case, we decided to employ both methods by uploading the documents from our on-prem file share to SPO and storing the rest of the data on an Azure files file share. This approach requires sorting through the data beforehand, which is generally a good advice, since old files can simultaneously be sorted out to shrink the overall amount of data.