Protection device and method for power failure of multi-node server and readable storage medium

Abstract

The invention discloses a protection device for power failure of a multi-node server. The protection device comprises N power supplies used for providing electric energy for the multi-node server, and a CPLD connected with the N power supplies and used for acquiring failure warning information of each power supply through the hard wire circuit and performing frequency reduction control on the multi-node server according to a preset frequency reduction rule when a frequency reduction condition is determined to be satisfied based on the acquired failure warning information of each power supply; wherein N is a positive integer. By applying the scheme provided by the invention, the power failure protection of the multi-node server is effectively realized, the real-time performance of the scheme is ensured, and the situation that the frequency reduction is not timely is avoided. The invention further provides a protection method for power failure of the multi-node server and a readable storage medium. The protection method has the corresponding technical effects.

Description

technical field [0001] The invention relates to the technical field of servers, in particular to a protection device, method and readable storage medium for multi-node server power failure protection. Background technique [0002] With the improvement of server capabilities, the power consumption is also increasing. Especially in multi-node servers, the computing density is high and the requirements for power consumption are very high. It is easy for the power supply to fail to meet the requirements of redundant functions. At this time, if When single or multiple power supplies fail, the system may exceed the power supply load, resulting in machine damage or data loss. The current countermeasure is to use CMC (Chassis Management Controller, Chassis Management Controller) to poll and detect the failure alarm status of the power supply in each working state. Power consumption requirements, so that the multi-node server as a whole will not exceed the power load. [0003] Howe...

AI Technical Summary

Problems solved by technology

[0003] However, in the current solution, it is required that the time-consuming from power failure alarm to node frequency reduction is relatively short, but in practical applications, the software response time of the CMC controller is relatively long, which cannot better meet the real-time requirements. That is to say, it often happens that the CMC controls the frequency reduction slowly, which leads to the power consumption of the whole machine cannot support, which has caused the consequences of burning the power supply, power failure and data loss.

Images