Blade Power State Management via Enclosure Manager
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing power management techniques for blade computer systems, such as OSPM/ACPI, are not applicable directly due to the absence of human user attendance and the presence of individual operating systems in each blade, leading to issues with independent power state transitions that can result in blades being misinterpreted as malfunctioning when placed in sleeping states.
Innovation Solution
An allocation server and enclosure manager system that dynamically manages blade power states by maintaining records of allocated and unallocated blades, using network messages to transition blades between working and sleeping states, and employing ACPI-compliant mechanisms to control power mode bits, ensuring efficient power conservation without impacting latency or functionality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If OSPM/ACPI power management techniques are applied to blade computer systems, then power consumption is reduced, but blade status may be misinterpreted as malfunctioning due to independent power state transitions
Solution Approach 1:
The patent introduces an enclosure manager as an intermediary between the allocation server and individual blades. This mediator coordinates power state transitions across all blades, preventing independent transitions that cause status misinterpretation. The enclosure manager receives allocation status information and manages power states centrally, resolving the contradiction between power savings and reliable status interpretation.
2Use of energy by moving object
If blades are placed in sleeping states to conserve power, then power consumption and heat generation are reduced, but resource availability latency increases
Solution Approach 1:
The system dynamically adjusts blade power states based on real-time allocation status. Blades transition to sleeping states when unallocated to conserve power, and wake up when allocation is needed. This dynamic state management resolves the contradiction by adapting power consumption levels to actual resource demand, minimizing both waste and latency.
Solution Approach 2:
The enclosure manager proactively manages power state transitions before allocation needs arise. When blades become unallocated, the system preemptively transitions them to sleeping states, preparing for future power savings while maintaining the ability to quickly restore service when needed.
3Adaptability or versatility
If individual operating systems control power states in each blade, then local autonomy is maintained, but coordinated power management across the enclosure becomes difficult
Solution Approach 1:
The patent extracts power management coordination functionality from individual blade operating systems and places it in the enclosure manager. This separation allows blades to maintain their local operating system autonomy while the enclosure manager handles coordinated power state management across the entire enclosure, resolving the contradiction between local autonomy and system-wide coordination.
Data Source
AI summary
A blade computer system includes a plurality of client devices, a blade enclosure having a plurality of blades therein, and an allocation server configured to allocate and deallocate the blades to and from the client devices. The blade enclosure is configured to place individual ones of the blades into or out of a sleeping state responsive to network messages received from the allocation server.