Rack Management Controller Node Type Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional rack management systems struggle to manage power consumption and temperature control in real-time, especially when high-power nodes are added, leading to unstable power supply and heat dissipation failures.

Innovation Solution

A method is introduced where a detecting unit connected to the Rack Management Controller (RMC) detects status changes in the rack system, determines the type of newly inserted nodes, and performs sensor readings to provide power management and temperature control, ensuring power distribution and heat regulation within the system's limits.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the RMC actively inquires messages of each node periodically, then the rack management can be maintained, but the RMC cannot obtain the type and power consumption information of newly inserted nodes in real time

Engineering Contradiction:
Improverack management stabilityVSAvoidreal-time information acquisition delay
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

Instead of the RMC actively inquiring node information periodically, the invention inverts the approach by having the detecting unit passively detect chassis status changes and actively acquire node information only when changes occur. This reverses the traditional active-inquiry model to a change-triggered model, achieving real-time information updates without continuous periodic polling.

Inventive Principle:
Principle #13The other way round (Inversion)

Solution Approach 2:

The detecting unit autonomously detects chassis status changes and determines whether nodes have been inserted or removed without requiring RMC intervention. The system serves itself by automatically identifying configuration changes and triggering information acquisition only when necessary, eliminating the need for continuous RMC polling.

Inventive Principle:
Principle #25Self-service

2Productivity

If multiple motherboards with high power consumption are turned on simultaneously, then the rack system functionality is enhanced, but the total power consumption exceeds the maximal load of the power supply

Engineering Contradiction:
Improverack system functionalityVSAvoidtotal power consumption
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The RMC receives real-time power consumption information from the detecting unit about each node type and its actual consumption. This feedback mechanism allows the RMC to monitor total power consumption continuously and make informed decisions about node activation, preventing the system from exceeding power supply capacity while maximizing functional utilization.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system dynamically adjusts operational parameters based on real-time conditions. The RMC changes the operational state of nodes (activation/deactivation) based on power consumption thresholds and available capacity, optimizing the balance between system functionality and power consumption without fixed pre-configuration.

Inventive Principle:
Principle #35Parameter changes

3Productivity

If high-power nodes are added to the rack system, then the system capabilities are improved, but heat dissipation becomes insufficient

Engineering Contradiction:
Improvesystem capabilitiesVSAvoidheat dissipation capacity
Core Design Contradiction:
ProductivityVSTemperature

Solution Approach 1:

The detecting unit provides real-time information about node types and their power consumption characteristics to the RMC. This feedback enables the RMC to predict heat generation from high-power nodes and adjust thermal management strategies accordingly, balancing system capabilities with heat dissipation capacity.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The RMC determines node types and power consumption characteristics before activating nodes. This preliminary action allows the system to pre-plan power distribution and thermal management strategies, preventing overheating issues before they occur by anticipating the thermal load of newly inserted high-power nodes.

Inventive Principle:
Principle #10Preliminary action

4Device complexity

If the RMC does not have real-time power consumption information, then the system complexity is reduced, but precise power control and temperature management become difficult

Engineering Contradiction:
Improvecontrol system complexityVSAvoidpower consumption measurement
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The detecting unit serves as an intermediary between the physical node configuration and the RMC. It automatically detects chassis status changes, determines node types, queries power consumption information from nodes, and presents this information to the RMC. This intermediary handles the complexity of real-time monitoring, keeping the RMC's control logic simple while achieving precise power consumption measurement.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS8656003B2Method for controlling rack system using RMC to determine type of node based on FRU's message when status of chassis is changed
Publication Date: 2014.02.18 INVENTEC CORP
  • US8656003B2 patent drawing
  • US8656003B2 patent drawing
  • US8656003B2 patent drawing

AI summary

A method for controlling a rack system including a plurality of detachable chassis, where at lease one node is disposed in the chassis and a rack management controller (RMC) is disposed in the rack system. First, at least one detecting unit connected to the RMC and the node of the chassis in the rack system is provided. Next, a status message of the chassis is detected for determining whether the status of the chassis is changed. When the status is changed, the detecting unit determines whether the node corresponding to the chassis exists in the rack system. When the node exists, the detecting unit acquires a message of a field replaceable unit (FRU) of the node. Thereafter, the detecting unit transmits the message of the FRU to the RMC. Then, the RMC determines a type of the node according to the message of the FRU.