This Monday, opening the PDC 2008 conference in Las Vegas, the Microsoft Chief Architect Ray Ozzie introduced Windows Azure, a new flavor of Windows made for cloud computing.
During his keynote Ozzie briefly mentioned that Azure is built on top of a hypervisor, without adding further details. It was easy to support that such hypervisor is the upcoming Hyper-V 2.0.
In an interview with InfoWorld today, the Microsoft Corporate Vice President Amitabh Srivistava unveiled a surprising detail:
InfoWorld: So that’s all dynamic provisioning? It’s all using Hyper-V?
Srivistava: No, we have our own hypervisor, which we have designed specifically for the design point. We just call it Windows Azure hypervisor, but a lot of the advancements we are making are moving into Hyper-V, [so] the same thing will be available to the on-premises customer. The virtualization support that you get in hardware like NPT and EPT, the nested page tables that are there. We are taking a lot of advantage of that, so that’s what [Microsoft engineer] Dave Cutler has done in really optimizing the thing out, so the overhead of the hypervisor is extremely low. And when the hypervisor overhead is low, then we can get maximum utilization by dividing the machine.
It’s very interesting that Microsoft develops two hypervisors at the same time, and that the Azure one is more efficient than Hyper-V at the point that the former’s features are incorporated to the latter and not vice versa.