Floating Node Boundaries for Multi-Hypervisor Resource Allocation

The patent segments the multi-node server resources at the socket level rather than at the physical node level. Each socket within a node can be independently assigned to different hypervisors, creating finer-grained partitions that maintain electrical isolation where needed while allowing flexible resource allocation across hypervisors. This socket-level segmentation enables multiple hypervisors to coexist on the same physical node without requiring complete node partitioning.

The patent applies local quality by allowing different isolation and sharing characteristics at different levels of the hardware hierarchy. At the socket level, resources can be locally assigned to specific hypervisors with appropriate isolation, while at the node level, multiple hypervisors can share access to common resources like memory and I/O devices. This localized application of isolation principles maintains security where required while enabling flexibility at higher levels.

The patent further segments resources at the socket level within each cell, providing finer granularity than traditional cell-level partitioning. This allows individual sockets or groups of sockets to be allocated to different virtual machines or hypervisors within the same cell, enabling more precise resource allocation while maintaining the security boundaries of the cell structure.

The patent introduces dynamic resource allocation capabilities that allow resource boundaries to be adjusted based on workload requirements. The system can dynamically reassign sockets and resources between hypervisors and virtual machines without requiring physical reconfiguration, enabling floating node boundaries that adapt to changing resource demands while maintaining security constraints.

The patent implements a universal service processor that can manage multiple hypervisors and their associated virtual machines across the entire multi-node server. This single service processor provides multi-functional capabilities including resource allocation, monitoring, and control for all hypervisors, eliminating the need for separate management processors for each hypervisor and reducing overall system complexity.

The service processor acts as an intermediary between the hardware resources and multiple hypervisors, providing a centralized management interface. It mediates resource allocation requests, enforces security policies, and coordinates operations across multiple hypervisors, simplifying the management architecture by providing a single point of control rather than requiring direct peer-to-peer management between multiple hypervisors.