If we look at Azure Stack and compare it to Azure, the huge difference is that Azure Stack is based on containers with hundreds and thousands of servers running a kind of Windows Server with some kind of Hyper-V, specially designed for Azure data centers. As Azure Stack works starting with four servers and as first step moves to 12 servers, there needs to be a kind of difference in the underlying technology. Microsoft always talks of a cloud inspired infrastructure with Azure Stack. This means the technology is somehow similar to Azure, but it is only similar and not the same.

In this chapter, we will talk about the technical basis of Azure Stack as it makes it easier for you to troubleshoot an environment that you understand from the technical perspective. Azure Stack is running based on Windows Server 2016 and the Hyper-V technology.

The most important technologies we will look at are as follows:

• Windows Server 2016 Storage Spaces Direct
• Windows Server 2016 Software Defined Networks

The reason for this is simple: starting at the upper level going down, we have Azure Stack Services that rely on Azure Stack Resource Providers (RPs), managed by Azure Resource Manager (ARM). In general, this is the same as with Azure. Going down from ARM is when the difference begins. With Azure Stack Deployment Toolkit, we have a single server Storage Space cluster with Hyper-V VMs for the Azure Stack management services on top. If we take a look at the multi-node deployments, we will see four to 12 servers running in a single Window Server 2016 clustered environment with at least two management VMs per service with a software load balancer that is experienced by Azure public sitting in front of it. Now let's look at how it goes by taking a deep dive into it.