It’s Microsoft’s cloud computing platform equipped with services you can use. It supports all the cloud service types (IaaS, PaaS and SaaS) with services such as VMs, website and DB hosting, event driven apps and AI, ML and IoT.
It’s the account that allow you to create and use Azure Services through subscriptions, after you have them, then you can create resources within each subscription.
It has datacenters around the world, which are facilities with resources arranged in racks, with dedicated power, cooling and networking infrastructure. Azure datacenters are grouped into Azure Regions or Azure Availability Zones, to help achieve resiliency and reliability.
Microsoft has created boundaries called geographies, which oftentimes are the border of a country. Within those geographies, there are regions (miles apart from each other), which contain at least one but multiple datacenters that are nearby and networked together (with very low latency), that contain the hardware that Azure uses.
These are also within each geography and contain two regions (separated at least 300 miles). For these you should make sure to deploy resources redundantly to each regional pair (replicate the resources), so when updates are performed or any interruption, the services there are not impacted by it.
These are instances of Azure isolated from the main instance of Azure, and are used for compliance or legal purposes, for government agencies and so on.
These are physically separated datacenters within an Azure Region, equipped with independent power, cooling and networking. There are at least three availability zones within an enabled Azure Region.
<aside> 💡 These provide high availability and fault tolerance (in localized disasters, like something happened in one zone, the other ones continue working), but they may not help with disaster recovery (like a large-scale natural disaster, because it affects the whole region)
</aside>
Categories of Azure Services that support them: