We live in the era of hybrid clouds – hybrid of public & private clouds. More than 90% of Azure customers either already have or plan to have a hybrid strategy for their IT infrastructure. Their future IT plans include keeping part of their workload in Azure & part on the edge infrastructure. They are modernizing their Apps to run across these infrastructures. And our focus is to empower such customers to seamlessly operate their compute infrastructure in the hybrid world. Our commitment is to provide tools that allow transparent & common operations in Azure and on their edge deployments.
Microsoft is making sure that customers have the best technology for their hybrid needs with Azure Stack HCI and Azure Arc:
Today we are announcing the Public Preview of management of virtual machines on Azure Stack HCI through Azure Arc. With this, Azure customers get consistent experience in provisioning Linux or Windows virtual machines on Azure Stack HCI or in Azure from the Azure portal. The integration provides ARM templates for VM CRUD to provide maximum level of infrastructure deployment automation for your edge infrastructure running Azure Stack HCI.
In addition, Azure Arc integration also makes Azure Stack HCI and the virtual machines on it, first class entities in Azure with ARM projections. That means, these entities can have role-based access natively through ARM which allows for self-service in VM provisioning. That’s right – with Azure Stack HCI, you can now delegate VM CRUD permissions to other users & groups in the organization to truly bring cloud like management on the edge infrastructure.
Some other benefits of ARM projections for virtual machines are that the projected entities can have Locks & Tags and can be put under organization policies. These entities can be globally searched across all resource groups & subscriptions.
In short, you can now leverage the same skills and processes across all your environments and manage Azure Stack HCI resources just like you would manage your Azure cloud resources.
Over the next few months, we will also introduce support for VM extensions for virtual machines on Azure Stack HCI. You will be able to discover, add and use extensions that allow you to run post deployment configuration or provide easy integration with your favorite service.
To bring Arc VM management to Azure Stack HCI cluster, you need the following:
Setting up an Arc Resource Bridge and creating a custom location for the Azure Stack HCI can be done through Windows Admin Center or command line. You will need the latest extensions on your Windows Admin Center to get the new functionality.
You can find detailed documentation on deployment of Arc Resource Bridge and creating a custom location. The next step is to project cluster resources such as VM Images and VM network to Azure for provisioning VMs from Azure portal.
And with that, you are now ready to provision your first VM from Azure portal on Azure Stack HCI cluster. Navigate to www.portal.azure.com and create Azure Arc Virtual Machine using the Custom Location for your Azure Stack HCI cluster.
As an infrastructure admin, you can also manage all the VMs on the cluster from the Azure Stack HCI resource page in Azure portal:
Users or groups with Contributor roles in the subscription, can also create and manage virtual machines on the cluster even though they do not have any direct access to the underlying Azure Stack HCI cluster. In future, we will bring more finer grained access controls for self-service VM management.
We can’t wait for you to get started with Arc VM management on your Azure Stack HCI clusters. For any issues, reach out to the Microsoft Support team. We would love to hear your feedback and suggestions on this form.
Today’s announcements bring us one important step closer toward realizing our vision with Azure Stack HCI - to bring together the familiarity and flexibility of on-premises virtualization with the Azure workloads, Azure management, and Azure infrastructure innovation you want, as a subscription service that keeps getting better over time.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.