Microsoft has begun adding Availability Zones to its Azure cloud platform, providing customers with an additional tool to make sure their applications stay operational.

The zones are sets of data centers within a single cloud region that are geographically isolated from one another. Users can replicate workloads across zones that need higher redundancy but will be hurt by latency in the event of a failure.

Starting today, Availability Zones are in preview in two cloud regions: East US 2 in Virginia and West Europe in the Netherlands. Microsoft Azure CTO Mark Russinovich said in an interview that the company plans to roll out more such zones in the future, with the forthcoming Paris region one of the key expansion points.

This is one part of Microsoft’s overall redundancy strategy with Azure. The company also offers Availability Sets, which are sets of virtual machines within the same data center that can help isolate workloads from a single hardware failure. Region Pairs let customers match up virtual machines across two regions in the same data residency boundary for protection against large-scale disasters like tornadoes and earthquakes.

Microsoft is far from the first company to use Availability Zones for its cloud platform. Amazon Web Services and Google Cloud Platform both offer their own versions, as do others.

Adding Availability Zones to Azure is going to be a long process. While one of the cloud platform’s regions is often made up of multiple data centers, those locations may not be isolated enough to meet the fault tolerance guarantees that Microsoft is making, Russinovich said.

Looking toward the future, Microsoft will focus on building out Availability Zones where they’re needed most, so it’s possible that not all of the company’s 42 announced and built regions will support the capability going forward.