Blade Server Management via Distributed Subsystems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing blade server management systems have complex designs, leading to long development cycles, high design costs, and poor reliability due to the inability to maintain remote management when both system management modules fail, along with difficulties in heat dissipation and wiring complexity.
Innovation Solution
A blade server management system with management subsystems on each blade connected via a local area network, a communication subsystem on the backboard for external networking, and a monitoring subsystem for parameter monitoring, featuring a master management subsystem for redundancy and self-management, eliminating the need for a KVM switching module and simplifying design.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If two system management modules with master-backup architecture are used, then remote control management is implemented, but reliability deteriorates when both modules fail
Solution Approach 1:
The management system is segmented into multiple independent management subsystems, each capable of autonomous operation. Each blade server has its own management subsystem that can independently provide remote control management functions, eliminating the single point of failure in master-backup architecture.
Solution Approach 2:
Each blade is equipped with its own local management subsystem that can independently handle management tasks for that specific blade. This distributed local management capability ensures that failure of one management subsystem does not affect the manageability of other blades.
2Reliability
If multiple management subsystems are deployed on each blade, then redundancy and reliability are improved, but system design complexity increases
Solution Approach 1:
Each management subsystem is designed with universal functionality to perform multiple management tasks including remote control, monitoring, and configuration. This multi-functional design reduces the need for specialized components and simplifies the overall system architecture while maintaining high redundancy.
3Ease of manufacture
If traditional backboard routing design is used, then system management is achieved, but heat dissipation becomes difficult and wiring becomes complex
Solution Approach 1:
The management subsystems communicate through a network interface rather than traditional backboard routing, transitioning from a planar wiring approach to a multi-dimensional networked architecture. This allows for more flexible heat dissipation paths and simplified wiring while maintaining full management functionality.
Data Source
AI summary
The present invention discloses a blade server management system comprising a plurality of management subsystems provided on each blade in a blade server and connected to each other via a local area network, each of which implements remote control and self-management for the corresponding blade, and said management subsystems including at least one master management subsystem for monitoring the operation parameters of the blade server through a monitoring subsystem; a communication subsystem provided on a backboard for connecting a external network and said local area network made up of said management subsystems; and a monitoring subsystem provided on the backboard and connected to said management subsystems for monitoring the operation parameters of the blade server based on a command from the master management subsystem. The blade server management system of the present invention has high system reliability and stability as well as a simple routing design while reducing the complexity of system design.