EDIT: I'm assuming you're talking about Virtual Machines (IaaS), and not Cloud Services (PaaS).
Virtual Machines can, and will, restart, for several reasons. For example:
- Hardware failure, where your Virtual Machine will then be restarted on another server.
- Host OS refresh. This is the operating system running the physical server.
- Some type of OS crash
Also keep in mind: Virtual Machines are in Preview with no SLA today. So there wouldn't be any information readily available to you for determining why your Virtual Machine became unavailable.
If it was unavailable for 30 minutes, then this hints at something akin to a host OS update or your virtual machine being moved. If it was down for, say, 5 minutes, then I'd guess it was an OS crash.
UPDATE I just looked at the Azure Dashboard which is showing degraded Compute with Virtual Machines (see RSS feed with problem description). Perhaps this is the root cause of your particular outage...