Blade Server Dual CMC Redundancy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional blade servers with a single chassis management controller lack redundancy, resulting in server board failure when the controller fails, leading to unavailable resources.
Innovation Solution
A blade server design featuring dual chassis management controllers (CMCs) where the second CMC detects the status of the first CMC and takes control if it malfunctions, ensuring continuous operation by switching control signals and maintaining communication capabilities through redundant switches and inter-connected blade server module boards.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single chassis management controller is used in a blade server chassis, then the device complexity is reduced and manufacturing cost is lowered, but the reliability deteriorates because the server board becomes unavailable when the controller fails
Solution Approach 1:
The system performs preliminary configuration where the first CMC is set as the primary controller and the second CMC as the standby controller before any failure occurs. The second CMC is pre-configured to detect the status of the first CMC and automatically take over control when the first CMC fails, ensuring seamless redundancy without requiring complex real-time decision-making logic
Solution Approach 2:
The patent implements a standby CMC that replicates the functionality of the primary CMC. The second CMC serves as a copy of the first CMC, ready to assume the same control responsibilities. This copying approach provides reliability through redundancy while maintaining relatively simple device architecture, as the standby controller uses the same hardware and software configuration as the primary controller
2Adaptability or versatility
If multiple chassis management controllers are deployed to control different server boards, then the coverage and control capability are improved, but the reliability deteriorates because the controllers cannot mutually support each other
Solution Approach 1:
The patent merges the functionality of multiple controllers into a unified redundant control system. Instead of having independent controllers that cannot support each other, the first and second CMCs are combined into a single logical control unit where one serves as primary and the other as standby. They share the same control responsibilities and can seamlessly transfer control, providing both comprehensive coverage and mutual support capability
Solution Approach 2:
The system establishes preliminary control relationships between the first and second CMCs, where the second CMC is pre-configured to detect and respond to failures of the first CMC. This preliminary arrangement ensures that when controllers need to work together for redundancy, the support relationship is already in place, eliminating the need for complex runtime coordination
Data Source
AI summary
A blade server has a casing, a plurality of blade server module boards, a first chassis management controller (CMC) and a second CMC. There are a plurality of main board slots at a first side of the casing and at least a first CMC slot and a second CMC slot at a second side of the casing opposite to the first side. The plurality of main board slots are all electrically connected to both of the first CMC slot and the second CMC slot, wherein the first CMC slot and the second CMC slot are electrically connected to each other. The plurality of blade server module boards are plugged into the plurality of main board slots. The first CMC and the second CMC are redundant to each other. When one of the two CMCs malfunctions, the blade server module boards are controlled by the other one CMC functioning normally.