Cloud-native greenfield projects will often be serverless, abstracting away the concept of traditional servers. Alternatively, they might leverage containerization, for example, using Azure Kubernetes Service, if dictated by requirements such as confidential computing. (Confidential computing is a concept where data is safeguarded not only during storage and transit but also when being actively processed within the system. That is, “data in use,” in addition to the more familiar concepts of “data at rest” and “data in transit.”)
But sometimes, it’s necessary to “lift and shift” on-prem servers to the cloud as part of a migration. When doing so, don’t let the fear of costs prevent you from taking full advantage of the opportunities for high availability. With Azure Availability Zones, each zone is a discrete data center with independent power, cooling, and infrastructure, connected by a low-latency network. Whether you’re using Availability Sets or Virtual Machine Scale Sets (VMSS), you’ll get better availability at a lower cost by diversifying your workload across at least three zones instead of just two.