Azure Arc – Bringing the best of public cloud to hybrid and multi-cloud environments
Microsoft continues to bring Azure-based managed services to Azure Arc, the hybrid and multicloud platform. At BUILD 2021, which took place last week, the company announced the availability of Platform as a Service (PaaS) features on Azure Arc in preview.
What is Azure Arc?
Azure Arc is Microsoft’s hybrid and multicloud platform that seamlessly extends the capabilities of Azure public cloud to on-premises data centers, edge locations, and any public cloud. Customers who invest in Azure can centrally manage and manage infrastructure that runs in a variety of environments.
With Azure Arc, Microsoft has extended the core resource management functionality of Azure to support external resources such as Linux and Windows servers, Kubernetes clusters, SQL Server instances, PostgreSQL Hyperscale servers, and more.
Azure Resource Manager (ARM), the control plane responsible for provisioning and managing cloud services such as Azure VMs and Azure SQL Database, is expanding its capabilities to support non-Azure resources running outside of the public cloud.
Azure Arc acts as the glue connecting external resources to Azure Resource Manager.
Kubernetes – The foundation of Azure Arc
At the heart of Azure Arc is Kubernetes, the open source container orchestration engine that powers modern infrastructure. Microsoft relies on Kubernetes to provide a cohesive environment to run its hybrid and multicloud services.
It doesn’t matter if the Kubernetes cluster is running on Amazon EC2 or Google Compute Engine or a vSphere VM farm. Microsoft can push an agent and start running their managed services. The Azure engineering team is squarely focused on porting some of its best-managed services to Kubernetes without worrying about the underlying compute infrastructure.
Microsoft is one of the first companies to leverage Kubernetes as the basis of its hybrid and multi-cloud strategy. In just two years, he’s ported a dozen managed services to Kubernetes, calling them Arc-compatible services.
Although Anthos, Azure Arc’s closest competitor, launched long before, Google has been slow to integrate cloud-based managed services into its hybrid platform. In contrast, Microsoft has been aggressive in adding services to Azure Arc.
Azure Arc as an end-to-end hybrid platform
A business needs a reliable infrastructure to run the registration system and the engagement system. The recording system manages data management through relational databases, while the engagement system acts as an interface to the data. More recently, companies have started investing in the intelligence system, which provides predictive analytics and services based on machine learning.
Azure Arc is the only hybrid and multi-cloud platform capable of running all three layers: the registration system, the intelligence system, and the engagement system.
Azure Arc Enabled Data Services as a registration system
Microsoft has a history of managing the business registration system through SQL Server and Azure SQL DB.
Using Azure Arc, Microsoft introduced SQL Managed Instance and PostgreSQL Hyperscale in hybrid and multi-cloud environments. These two types of databases cover a wide range of use cases and scenarios that allow customers to run modern applications with the ease of consuming DB as a Service (DBaaS) offerings.
Azure Arc Enabled data services are becoming the gold standard for modern businesses.
Azure Arc machine learning as an intelligence system
With Azure Arc Enabled Machine Learning, Microsoft integrated Azure ML with Arc. Customers can run the MLOps pipeline in their environment by ingesting data from on-premises and cloud data sources, performing training tasks, and even hosting models for inference.
Customers can run their ML training on any Kubernetes target cluster in Azure Cloud, Google Cloud, AWS, edge devices, and on-premises through Azure Arc-enabled Kubernetes. With just a few clicks, they can get the Azure Machine Learning agent running on any open source Kubernetes cluster supported by Azure Arc.
The Azure Arc enabled machine learning service provides the intelligence system.
Application services on Azure become the engagement system
During BUILD 2021, Microsoft announced the availability of application services on Azure Arc. With this integration, customers can run App Service, Functions, and Logic Apps on any Azure Arc-enabled Kubernetes cluster.
Interestingly, clients can run different classes of applications ranging from web interfaces, API services, serverless applications, and event-based applications.
An application deployed on Azure Arc can communicate with the database hosted on Arc Compatible Data Services while leveraging machine learning models for predictions without ever leaving the corporate data center.
The Azure Arc App service provides the engagement system to end users.
Key points to remember
Microsoft is rapidly evolving with Azure Arc by porting key managed services to the hybrid and multicloud platform.
Every Arc-enabled service running on Kubernetes takes advantage of automation, centralized logging, GitOps-based configuration, and role-based access control and security. Customers can take advantage of centralized Azure tools and frameworks to manage distributed infrastructure and platforms.
Azure Arc builds a solid bridge between Microsoft’s public cloud and the cloud platforms of its competitors. Imagine running SQL Managed Instance as part of Arc Compatible Data Services on Amazon EKS while training a machine learning model through Arc Compatible ML on Google Kubernetes Engine.
With industry best practices built into Azure Arc, customers only focus on workloads and applications.
Azure Arc is one of Microsoft’s best initiatives, helping the company become the leader in hybrid cloud and multi-cloud offerings.