Method and system for monitoring and managing sensor through BMC based on I3C bus

Abstract

The invention discloses a method and system for monitoring and managing a sensor through a BMC based on an I3C bus, and the method comprises the steps: S1, enabling the BMC and a to-be-monitored sensor to be connected through the I3C bus, and embedding a preset soft interrupt in a protocol; s2, acquiring a current temperature value of the sensor, judging whether the current temperature value exceeds a threshold range or not, and if yes, sending an interrupt request to the BMC, and increasing the polling frequency of the sensor after the temperature value of the sensor is greater than the maximum value of the threshold range or reducing the polling frequency of the sensor after the temperature value of the sensor is less than the minimum value of the threshold range. The BMC and the sensor are connected through the I3C bus, soft interruption is started for the I3C bus, polling processing is conducted on the I3C sensor, when the temperature of the sensor exceeds or reduces the set threshold value, the BMC is informed of the need of increasing or reducing the polling frequency through the soft interruption, the query frequency of the sensor is adjusted in time, resources are saved, and the heat dissipation problem is avoided.

Description

technical field [0001] The invention relates to the technical field of BMC management, in particular to a method and system for monitoring and managing sensors of a BMC based on an I3C bus. Background technique [0002] With the rapid development of the Internet, it has also brought prosperity to the server market. BMC assumes the role of monitoring and management Sensor in the server. In the traditional sensor model, the I2C bus is generally used, the sensor is used as the slave device, and the BMC is generally used as the master device to continuously query the value of the sensor. [0003] There are two main types of management schemes available: [0004] One is that the BMC queries at a fixed frequency. Due to the large number of sensors, when the sensor value is lower than the critical value, too high query frequency will consume a lot of BMC resources, which is not only not energy-saving enough, but also occupies the BMC processing performance. On the contrary, BMC ...

AI Technical Summary

Problems solved by technology

Due to the large number of sensors, when the sensor value is lower than the critical value, too high query frequency will consume a lot of BMC resources, which is not only not energy-saving enough, but also occupies the BMC processing performance. On the contrary, BMC queries at too low frequency, because There are many sensors, which may delay the timing of the alarm, and the heat dissipation cannot keep up, causing the CPU frequency to drop, causing serious problems

If the sensor as a slave device uses the I2C bus to actively send interrupt signals for BMC to process, this situation can save the processing performance of the BMC, but at the same time it will consume a lot of physical pins, because if the I2C bus uses interrupts, a real The cable to send out the interrupt signal will increase the difficulty of wiring the motherboard, and the cost will increase geometrically

[0005] The other is that the BMC has been querying at a low frequency, and then use a high frequency when it reaches a critical point. The problem is also the risk of low-frequency query. The temperature may rise rapidly within a few seconds, causing the heat dissipation to fail to keep up and causing other more serious problems. catastrophic problem

Images