Hardware monitor device for server

Abstract

The invention discloses a hardware monitor device for a server. The device comprises a first acquisition module which is used for acquiring CPU state data, memory state data and hard-disk state data through the operation system of a server, a second acquisition module which is used for acquiring mainboard state data, fan state data, power supply state data and temperature state data through BMC, and a processing and alarming module which is used for monitoring the CPU state data, the memory state data, the hard-disk state data, the mainboard state data, the fan state data, the power supply state data and the temperature state data according to requirements, and alarming when an arbitrary device is failed. The hardware monitor device can be used for acquiring the hardware state of a server in a remote mode to effectively reduce the intensity of labor routing inspection, instantly finding out a sever failure and accurately positioning the failed hardware so as to improve the maintenance efficiency.

Description

technical field [0001] The present invention relates to the technical field of servers, in particular to a server hardware monitoring device. Background technique [0002] In a modern data center (Internet DataCenter, IDC), the number of server nodes is increasing, and the corresponding operation and maintenance workload is becoming heavier and heavier. How to detect hardware problems in the server earlier and more accurately has become a guarantee The primary issue for the normal operation of the data center business. [0003] The existing method of monitoring the server is that the operation and maintenance personnel of the IDC need to go to the computer room for inspection every once in a while, so as to find problems in time. However, the inspection of the computer room can only be observed through the indicator light of the server The state of the hardware, some hidden problems (for example, memory correctable errors causing CPU performance degradation) can not be foun...

AI Technical Summary

Problems solved by technology

[0003] The existing method of monitoring the server is that the operation and maintenance personnel of the IDC need to go to the computer room for inspection every once in a while, so as to find problems in time. However, the inspection of the computer room can only be observed through the indicator light of the server The state of the hardware, some hidden problems (for example, memory correctable errors causing CPU performance degradation) are not found

[0004] Another method of monitoring the server in the prior art is to provide the server with a unified BMC (Baseboard Management Controller, baseboard management controller) mechanism to obtain the hardware health status, but the BMC cannot obtain the status of all hardware of the server

[0005] For the above-mentioned problems in related technologies, no effective solution has been proposed yet

Images