Following closely on the heels of two recent announcements about new high performance capabilities on Microsoft Azure (HPC Team blog, Azure blog), we are happy to announce  that we are adding support for A8 and A9 compute intensive instances in 2 new regions: US South Central and Europe North.

These 2 new regions complement the 5 regions already supporting A8 and A9 compute intensive instances.

At the same time, we are making the US North Central and US South Central regions available to all customers (previously, these regions were only available to existing subscriptions).

The following is a list of all 7 regions where support for A8 and A9 compute intensive instances is currently available, grouped by geographical area:

United States

  • US East
  • US West
  • US North Central
  • US South Central (new!)

Europe

  • Europe West
  • Europe North (new!)

Japan

  • Japan East

Finally, it is important to mention the following:

  • All 7 regions where A8 and A9 compute intensive instances are available include support for Cloud Services (PaaS) and for Virtual Machines (IaaS).
  • A8 and A9 compute intensive instances in all 7 regions include a 32Gbps backend network that supports remote direct memory access (RDMA) for low-latency and high-throughput communications between compute nodes.
  • RDMA capabilities on A8 and A9 compute intensive instances are supported in Microsoft MPI (MS-MPI), and we are working with partners to broaden the support both in other MPI stacks, and in Linux Virtual Machines.

You can learn more about the A8 and A9 compute intensive instances, and about Big Compute in Azure, in the Azure Technical Documentation Library.

Note: some of our documentation is still being updated to reflect the new regions, as well as the availability of Virtual Machines (IaaS) in all regions where A8 and A9 compute intensive instances are available.

As always, we want to know about your experience doing compute-intensive work on Microsoft Azure. Send us an email, with any feedback or suggestions.