What is the expected throughput for that resource at that layer? How much scale is provided by the resource to fit that demand? What's the dependency on other resources, globally or in other regions? Is the resource required to be globally distributed? Can the resource communicate with other resources, globally or in regions? What impact will the persisted state at this layer have on reliability or manageability? What is the expected lifetime of resource, relative to other resources in the solution? Should the resource outlive or share the lifetime with the entire system or region, or should it be temporary? For more information, see Architecture pattern of a typical mission-critical workload. The resources in each layer have distinct characteristics. Each stamp equates to a scale unit and, in case it becomes unhealthy, can be entirely replaced. The regional resources are provisioned as part of a deployment stamp. In this architecture, the application platform consists of global, deployment stamp, and regional resources. Use your capacity model and expected performance to define the boundaries of, number of, and the baseline scale of each unit. Scale units allow you to logically group resources and a unit can be scaled independent of other units or services in the architecture. Use a scale units-based architecture to handle increased load. As you choose Azure services for your application's platform, consider their Availability Zones support and deployment and operational patterns to use multiple Azure regions. Building zonal and regional redundancy in an active-active configuration is the main strategy. This approach ultimately increases the reliability of the system.Ī mission-critical application must be highly reliable and resistant to datacenter and regional failures. It's useful in defining roles and functions, and assigning appropriate privileges, and deployment strategies. This layered approach helps in creating logical and physical segmentation. Choose the right set of services, their configuration, and the application-specific dependencies.
Here are some overarching recommendations.ĭesign in layers.
Platform refers to the infrastructure components and Azure services that must be provisioned to support the application.
A key design area of any mission critical architecture is the application platform.
0 kommentar(er)
