One of the many benefits of moving your in-house servers up into the cloud is to off load the responsibility of the uptime.
But what does Microsoft offer its Azure customers in its Service Level Agreement?
A quote from the SLA:
“Windows Azure has separate SLA’s for compute and storage. For compute, we guarantee that when you deploy two or more role instances in different fault and upgrade domains your Internet facing roles will have external connectivity at least 99.95% of the time.”
1 Compute Instance
So basically if you only have 1 compute instance you won’t get any SLA. Which sounds a bit rubbish but the reasoning makes sense. If they need to repair the hardware your instance is on or even do an upgrade to the Azure framework, which we all want, they would need to take you down. Now realistically this is a fast process and they would bring you up on another piece of hardware, but that would incur downtime.
2+ Compute Instances
Now the SLA kicks in and you’re at 99.95% (which is pretty damn good).
When you deploy your Azure site package up to the cloud with only 1 instance you are greeted with the following warning/error, which explains about the SLA to you.
A warning has been encountered
A warning has been encountered. Click “See more details” to see the warning. Would you like to override and submit?
Warning: This deployment has at least one role with only one instance. We recommend that you deploy at least two instances per role to ensure high availability in case one of the instances becomes unavailable. Doing so also enables coverage of the Windows Azure Compute SLA (http://go.microsoft.com/fwlink/?LinkID=159704&clcid=0x409), which guarantees 99.95% uptime. For more information please visit http://go.microsoft.com/fwlink/?LinkID=202707 (http://go.microsoft.com/fwlink/?LinkID=202707). Find more solutions in the Windows Azure support forum (http://go.microsoft.com/fwlink/?LinkID=206235).