The cloud offers a model of simplicity for on-premises workloads
Customer IT environments have greatly evolved over the past decade, with a growing number of organizations choosing to host resources in the public cloud to gain benefits such as simplified management, lower total cost of ownership (TCO), and unlimited scale. At the same time, teams have also continued to acquire infrastructure in their own datacenters to host workloads on premises when this arrangement best suits their needs. Data sovereignty rules, government regulations, and corporate policies, for example, often keep data from being stored in the public cloud—and data gravity naturally encourages teams to deploy applications close to where data already resides. Internet of Things (IoT) applications are also working against the overall trend toward the public cloud by creating a new center of data gravity at the intelligent edge.
The net result of these developments is that, even though customers have come to expect the level of automation, portability, consistency, and at-scale management available through public cloud services, the cloud still remains out of reach for many workloads. In addition, customer infrastructure is becoming less manageable than ever, dispersed not only among different locations in different clouds, datacenters, and the edge, but also diversified among different technologies, including physical infrastructure, virtual infrastructure, and, more recently, container-based infrastructure.
What’s sorely needed to improve manageability is to take the managed services and self-service administration workflows that are available through public clouds such as Microsoft Azure and bring them to any environment, including the customer datacenter. A single control plane is needed both to unify management and to deliver platform-as-a-service (PaaS)-like capabilities outside the cloud, so as to simplify the deployment, management, scaling, governance, and patching of databases and applications. Bringing PaaS-like services outside the cloud would also help protect workloads by easing the configuration of high availability (HA), disaster recovery (DR), and security for these resources.
Microsoft has been a leader in recognizing and supporting businesses’ needs for hybrid infrastructure, offering a portfolio of Azure Stack products that combine hardware and software to extend Azure services and capabilities to datacenter, remote office, and edge locations. Now, Microsoft is introducing a way to run Azure services on customer premises and to extend Azure management across all customer environments.
Introducing Azure Arc
Azure Arc is a set of software technologies that is designed to help companies meet their governance challenges and make Azure services available everywhere across their diverse and sometimes sprawling infrastructures. Azure Arc also enables users to centrally deploy and manage servers, services, and applications in any environment where Azure Arc has been enabled. With Azure Arc, familiar Azure tools, technologies, and methods can be used across environments, including environments with limited or no connectivity to Azure.
At the most basic level, Azure Arc delivers two broad features:
- Azure Arc extends the management capabilities of Azure to customer infrastructure outside of Azure—across datacenter, multicloud, and edge environments. Through this feature, administrators can register resources in Azure Resource Manager that are hosted outside of Azure, and then manage these resources centrally in Azure. This capability is delivered through Azure Arc enabled servers and Azure Arc enabled Kubernetes.
- Azure Arc takes PaaS data services previously available only in Azure and brings them to customer premises and other public clouds. This feature is delivered through Azure Arc enabled data services.
We are planning to publish several articles on Azure Arc and its use cases in various customers scenarios in the coming weeks and months. Stay tuned!
To read more about Azure Arc, visit the product page.
Microsoft
