BMC Soft Reset via Master-Slave PMC Bus Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing liquid cooling systems face inefficiencies in managing communication abnormalities of the baseboard management controller (BMC) within the power management controller (PMC), leading to tedious hard resets when the BMC hangs, which disrupts real-time monitoring of servers.
Innovation Solution
A method for resetting the BMC using a master device to identify and transmit a reset command to a target slave device through a specific communication bus, ensuring a soft reset is performed via the programmable logic unit, thereby maintaining continuous monitoring of server health.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a hard reset is performed on the BMC when communication abnormalities occur, then the BMC can be recovered from hanging states, but the process becomes tedious and inefficient
Solution Approach 1:
The patent introduces a master device as an intermediary between the PMC and the BMC. When communication abnormalities occur, the master device receives the reset instruction and transmits it to the BMC through the communication bus, enabling soft reset without manual intervention. This mediator approach resolves the contradiction by automating the reset process while maintaining reliable BMC recovery capability.
2Reliability
If a hard reset is performed on the BMC, then the BMC can be recovered from hanging states, but real-time monitoring of server health is disrupted
Solution Approach 1:
The patent replaces the mechanical hard reset process with a software-based soft reset mechanism. The master device transmits reset instructions through the communication bus to the BMC, allowing the BMC to reset gracefully while maintaining the monitoring connection. This substitution eliminates the need to physically interrupt the monitoring system, thus resolving the contradiction between BMC recovery and monitoring continuity.
3Reliability
If manual hard reset operations are performed on the BMC, then communication abnormalities can be resolved, but the process is tedious and requires human intervention
Solution Approach 1:
The patent implements a self-service mechanism where the master device automatically detects communication abnormalities between the PMC and BMC, and autonomously transmits reset instructions to the BMC without requiring human intervention. The system monitors itself and performs corrective actions automatically, resolving the contradiction by making the system self-maintaining while ensuring communication stability.
Data Source
AI summary
A method for resetting a baseboard management controller (BMC) and a power management system are provided. The power management system includes two or more power management controllers (PMCs). Each of the PMCs includes a BMC. One of the two or more PMCs severs as a master device and remaining of the two or more PMCs sever as one or more slave devices communicated with the master device. The method includes: generating a reset command by the master device when a first condition is met. Then the master device determines a target bus and transmits the reset command to the target slave device through the target bus. The target slave device obtains the reset command through the target bus, and resets the BMC of the target slave device, in response to the reset command.