Distributed Hypervisor State Management for Virtual Machine Fault Tolerance

The patent segments the hypervisor control functions across multiple network devices rather than consolidating them in a single device. Each network device maintains a portion of the hypervisor functions and state information, creating distributed virtual machine management. This segmentation allows the system to maintain high resource utilization (efficiency) while ensuring that failure of one device does not affect the entire system (fault tolerance).

The patent implements local quality by storing state information locally at each network device while maintaining distributed coordination. Each network device maintains its own local state information for the virtual machines it manages, allowing independent operation and recovery. This local quality enables each device to continue serving its virtual machines even when other devices fail, thus improving fault tolerance while maintaining overall system efficiency.

The patent implements universality by creating a distributed hypervisor system where multiple network devices collectively provide virtual machine management services. Instead of requiring separate dedicated operating systems for each service, the distributed hypervisor allows any network device to potentially manage any virtual machine, providing multi-functionality across the system. This maintains fault tolerance through distribution while improving resource utilization through flexible allocation.

The patent transitions from a single-dimension approach (single hypervisor on one device) to a multi-dimensional distributed architecture where hypervisor functions are spread across multiple network devices. This dimensional change allows the system to simultaneously achieve the protection benefits of multiple systems while maintaining the resource efficiency of shared virtualization, as resources can be dynamically allocated across the distributed network devices.