/
Azure Global Infrastructure

Azure Global Infrastructure

Owned by thierry sinassamy
Last updated: Nov 21, 2023
2 min read

1- Azure Regions

  • Azure Regions leverage a low-latency interconnect to keep them all running together smoothly. IT is important to remember this point from a solution architecture perspective.

  • We can have multiple regions and something referred to as region pair set up → Cross-Region Replication for disaster recovery.

  • Azure Region is referred as a geography → data has to remain within a specific geogpraphy or regarding other compliance requirements in terms of where services and data are allowed to reside.

2- Availability Zones

  • This is physically separate location within a region

  • We can pin some services to a specific availability zone.

  • Not all services can be set in an availbility zone 1 and in an availibility zone 2, for example, storage accounts are actually referred to as zone-redundant services and that means we can configure them in such a way that we pick the region and that’s it → We just pick 1 region. If we choose to make it Zone Redundant, Microsoft is going to replicate that data accross the availability zones with a region.

3- Availability Sets

  • We have the possibility that our services (for example, 2 VMs for Web App) could both go offline at the same time if we we do not use availability sets. So, availibity set is to tell Microsoft that we have the solution that is going to be highly available.

  • WM Host Infra : Microsoft ensures that they are not going to sit on the same infrastructure → Fault Domain = Hardware that shares common sources of failure.

  • VM Host Infra : Platform services updated / restarted by Microsoft → Update Domain.