Microsoft commits to expanding Azure Availability Zones in 2021 and beyond

Microsoft is growing its Azure Availability Zone footprint significantly in this calendar year and beyond to provide broader datacenter failure protection.
Written by Mary Jo Foley, Senior Contributing Editor

Last year, Microsoft was slowly starting to ramp up its Azure Availability Zone coverage. Today, March 17, officials publicly committed to offering Azure Availability Zones in every county in which Microsoft publicly operates by the end of calendar 2021, plus offering Availability Zones in all new Azure regions Microsoft launches going forward.

Microsoft announced Azure Availability Zones for datacenter-failure protection in March 2018, starting with US Central and France Central. Availability Zones are located inside Azure regions and offer independent power source, networking, and cooling. Microsoft officials have said there is a minimum of three separate zone locations in enabled regions, which offer 99.99% uptime service-level agreements on covered services.   

Last fall, Gartner researchers dinged Microsoft for having what it said was the lowest ratio of availability zones to regions of any vendor in its category, and added that only a limited set of Microsoft services supported the availability-zone model. Since then, Microsoft has been expanding its availability-zone footprint, adding more support in Canada, Australia, and, as of this week, Brazil. Officials also said today that Microsoft would have Availability Zone support for all foundational and mainstream services by the end of 2021.

Availability Zones, in addition to being a key capability for users concerned with disaster recovery and business continuity, also play into how Microsoft and AWS have differentiated and defined their clouds. Microsoft officials often tout the "fact" that Azure has the biggest global footprint of all the public cloud vendors, with more than 60 available and announced Azure regions. (Microsoft defines an Azure region as "a set of data centers, deployed within a latency-defined perimeter and connected through a dedicated regional low-latency network".) But AWS defines a region as being comprised of at least two Availability Zones. 

Editorial standards