Dual Node Controller Architecture for Hot-Swap Multi-CPU Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multi-CPU systems with Non-Uniform Memory Access Architecture, the limited number of CPU interconnect ports restricts scalability, and the process of replacing a node controller (NC) for reliability causes system delays and affects cross-domain access, especially when NC reliability decreases.
Innovation Solution
Implementing a dual NC group architecture where the main CPU manages the system to switch cross-domain access routes from one NC group to another, allowing for hot removal and addition of NCs without disrupting system performance by modifying route configurations and initiating system silence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single NC is used for cross-domain access in a multi-CPU system, then the system structure is simple, but the system delay increases and reliability decreases when NC replacement is needed
Solution Approach 1:
The single NC is segmented into multiple independent NC groups (first NC group and second NC group), each capable of handling cross-domain access independently. This segmentation allows the system to switch between groups without complete system interruption, reducing delay while maintaining reliability through redundancy.
Solution Approach 2:
The system changes the operational parameter from single-NC mode to multi-NC group mode dynamically. By modifying route configurations and initiating system silence protocols, the system can switch between different NC groups based on reliability requirements, thereby reducing system delay during NC replacement while maintaining high availability.
2Reliability
If an NC is removed for reliability reasons, then system reliability improves, but cross-domain access is affected and system performance deteriorates
Solution Approach 1:
The system introduces an intermediary switching mechanism that redirects cross-domain access from the removed NC to an alternative NC group. This intermediary routing layer ensures that reliability is improved through NC replacement while productivity is maintained through seamless traffic redirection without performance deterioration.
Solution Approach 2:
The NC group configuration is made dynamic, allowing the system to adaptively switch between different NC groups based on operational needs and reliability requirements. This dynamic reconfiguration enables hot removal and addition of NCs without disrupting cross-domain access performance, maintaining both reliability and productivity.
3Adaptability or versatility
If dual NC groups are implemented for hot-swap capability, then system scalability and reliability improve, but device complexity increases
Solution Approach 1:
Both NC groups are designed with universal functionality to handle identical cross-domain access tasks. This multi-functionality allows either group to take over completely, simplifying the management complexity despite having dual groups. The standardized interface and routing logic reduce the operational complexity while maintaining hot-swap capability and scalability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention provides a multi-CPU system, where the multi-CPU system includes: at least two Quick-Path Interconnect QPI domains, a first node controller NC group, and a second node controller NC group; according to a CPU route configuration, there is at least one CPU that can access a CPU in another QPI domain by using the first NC group; and there is at least one CPU that can access a CPU in another QPI domain by using the second NC group. According to this topology, hot swap of an NC can be implemented while the system is relatively slightly affected.