Zum Hauptinhalt wechseln

 Subscribe

This post is authored by Bret Grinslade, Principal Program Manager and Josh Caplan, Senior Program Manager, Azure Analysis Services.

We have gotten good feedback from customers and partners starting to adopt Azure Analysis Services in production. Based on this feedback, this week we are releasing improvements around pricing options, support for backup and restore, and improved Azure Active Directory support. Please try them out and let us know how they work for you.

New Basic Tier

The new Basic Tier is designed to support smaller workloads with simpler refresh and processing needs. While you can put multiple models in one Standard instance, this new tier enables you to create models that are more targeted at less cost. The key differences between Standard and Basic is that the Basic tier does not support some specific enterprise features. Standard supports larger sizes and higher QPUs for concurrent queries and adds data partitioning for improved processing, translations, perspectives, and Direct Query. If your solution doesn’t need these capabilities, you can start with Basic. You can also scale up from Basic to Standard at any time. However, once you scale up to the higher tier you can’t scale back down to Basic. As an example, you can scale from B1 to S0 and then from S0 to S1 and back to S0, but you cannot scale from S0 to either the Basic or Develop tier.

Backup & Restore

We have added backup and restore. At a high level, you configure a backup storage location from your subscription for use with your Azure Analysis Services instance. If you do not have a storage account, you will need to create one. You can do this from the Azure Analysis Services blade for backup configuration or you can create it separately. Once you have associated a storage location, you can backup and restore from that location using TMSL commands or a tool like SQL Server Management Studio (SSMS) which will support this shortly. The documentation has more details on Backing Up and Restoring Azure Analysis Services models. One note, to restore a 1200 tabular model you have created with an on-premises version of SQL Server Analysis Services, you will need to copy it up to the storage account before it can be restored to Azure Analysis Services. The Microsoft Azure Storage Explorer or the AzCopy command-line utility are useful tools for moving large files in to Azure. In addition, if you restore a model from an on-premises server, the on-premises domain users will not have access to the model.  You will need to remove all of the on-premises users from the model roles and then you can add Azure Active Directory users to roles. The roles will be the same. Azure Analysis Services Server Admins will still have access as these are AAD based members. The setting on restore for “SkipMembership” will honored in a future service update to make managing cloud based role membership easier.

Improved Azure Active Directory integration

We have also done some work to improve the way Azure Analysis Services works with Azure Active Directory. Starting now, any newly created Azure AS server will be tied to the Azure AD tenant for which your Azure subscription is associated with and only users within that directory will be able to use your Azure AS server if granted access. This means that if a server is created in a subscription that is owned by Contoso.com than only users within the Contoso.com directory will be able to use those servers. In order to use that server, users must still be granted access to a role within the model. Azure AD supports a few options for allowing users outside of your organization to get access to resources within your tenant. One of these upcoming options will be Azure AD B2B. With B2B, you will be able to add guest access to users outside of your organization to your models through Azure Active Directory. We are hard at work enabling B2B for Azure Analysis Services end-to-end and will post an update when it is fully available in SSMS in SSDT as well as client tools.

  • Explore

     

    Let us know what you think of Azure and what you would like to see in the future.

     

    Provide feedback

  • Build your cloud computing and Azure skills with free courses by Microsoft Learn.

     

    Explore Azure learning


Join the conversation