Domain Management Processor for Blade Partition Configuration

The patent replaces expensive proprietary switch I/O fabrics with standard, low-cost PCI Express switches and standard PCI Express blades. These standard components can be readily substituted and upgraded without proprietary constraints, achieving high-performance communication while dramatically reducing system cost through the use of commodity hardware instead of custom ASICs and proprietary fabrics.

The patent creates a universal interconnect architecture using standard PCI Express switches that can aggregate multiple blades across different enclosures and support various blade types. This universal approach replaces proprietary I/O fabrics with a standardized interface that provides the same high-performance communication capabilities while enabling greater flexibility and lower costs through component interchangeability.

The patent implements dynamic blade aggregation where standard PCI Express switches can be configured to aggregate multiple blades into a unified computing resource pool. This dynamic configuration allows the system to achieve high-performance symmetric multiprocessing capabilities by logically combining multiple standard blades, enabling flexible resource allocation and parallel processing that overcomes the limitations of individual standard blades while maintaining cost-effectiveness.

The patent merges multiple standard PCI Express blades into a unified computing system through the PCI Express switch fabric. By combining the processing power, memory, and I/O resources of multiple standard blades, the system achieves high-performance symmetric multiprocessing capabilities that would be impossible with a single standard blade, while maintaining the cost advantages of using standard components throughout.