Extra Resiliency Options, a Name Change and a Price Drop for Microsoft Azure
Out with the old and in with the new
First up, you may notice the change in name I use in the title from 'Windows Azure' to 'Microsoft Azure'. This is a change we will be implementing over the next few months as it will take quite a bit of time to rebrand everything. Personally I think this is a great move as it hopefully will change any perceptions out there that Microsoft Azure is great for running Windows workloads but not Linux. Of course, Microsoft Azure, is a platform that can run Windows, Linux and Oracle workloads in a fully supported way, so the change is certainly welcomed by me. You can see the full announcement here:
The Price War
Recently AWS announced price changes so we have responded by matching and you can get full details below. As usual we will match prices at a minimum but under cut where we can.
Having built infrastructures on both AWS and Azure I still think there are a couple of points around cost that help us differentiate and are often forgotten:
- We bill VMs by the minute and not by the hour. Now this may not matter if you are running a production system 24x7, but if you are making best use of cloud infrastructure and scaling in and out or bursting to the cloud from on-premise for HPC reasons then this can make a real difference.
- We truly try to only charge for what you use. For example attach 16Tb of data disks to your SQL Server deployment and we won't charge you for 16Tb but only what you use. This allows you to build the infrastructure you need now and into the future but pay for what you actually use along the way.
- We are introducing a new type of instance that will not support load balancing or auto-scaling, i.e. single instance VMs. These instances (basic instances) will be priced even lower than our standard instances.
Resilience, Resilience, Resilience
As a cloud provider it is essential that we offer resilience within our platform and infrastructure. To do this we replicate data that is stored within Azure so that we can survive hardware failures within our data centres. When you create a storage container you choose which data centre to use, e.g. Europe North which is Dublin. We then replicate the data three times within Dublin for resilience. This is called "locally Redundant" storage. We also have the option to geo replicate data to another data centre within the same region, in our case Holland. This gives an extra layer of resilience such that if Dublin was to disappear we have a mechanism to get your data back. Today we have introduced a new option which is Zone Redundant storage. Each data centre is actually made up of a number of physically isolated data centres or Zones. This new redundancy option allows users to keep their data in one country but still have a good redundancy option. This will be of particular interest to companies that want to keep their data within a single country.
NOTE: This is not a backup solution. Like with any system you should always backup data that makes sense for your company e.g. 7 days retention and incremental backups. You can of course use our backup tools or third party tools in conjunction with Azure Backup Services or Blob storage and the resilience options above to create a robust disaster recovery process.