Detection And Application Of Server Signal LAMP Based On Micro Service Architecture And DBN

In today's society,each industry has put forward higher requirements for the long and stable operation of the software system.As the infrastructure of the software system,the server of the computer room is of great significance in the operation control of the whole software system and the storage and transmission of the information.In order to ensure the quality of software service,the background server system must also have strong reliability.If the external environment such as the power system,network system and other failures,or the aging of the server hardware failure,the server will run interruption,causing a bad impact on the continuous operation of the whole software system,resulting in huge economic losses.So it is necessary to monitor the running condition of the server in real time.It is very important to detect the running state of the server in time,find out the abnormal fault and send out the corresponding alarm signal.The health of the server is reflected by the signal lamp on the front panel of the server.In general,the green light represents the normal state,the yellow light indicates the system demotion state,and the red light indicates the serious system failure.The monitoring of early server state is conducted by manual timing inspection.This method is not only inefficient,but also difficult to guarantee the timeliness of feedback.Later,a monitoring method based on sensor networks was developed,but this method requires high economic cost,and it needs to be fully planned at the beginning of the construction of the machine room.Some scholars at home and abroad can apply to the automatic identification of traffic lights,which can be applied to the status detection of the server.The general idea is to extract the shape of the signal traffic lights and the HSV characteristics and combine some deep learning algorithms,because the server's signal lights have no obvious shape characteristic and small size,so this way is also the same There is a certain non applicability.This paper makes a deep research on the server state automatic detection system.In terms of algorithm,a feature extraction algorithm of signal light is proposed and evaluated by DBN network.After that,we compare the algorithm with the related algorithm,and prove that the algorithm has higher recognition rate.In engineering,the real-time video of the front panel of the server collected by industrial camera is used as the data source,and the architecture concept of micro service is adopted.The whole business is divided into a control subsystem,a video/image acquisition subsystem,an analysis subsystem and other modules,andThe expansibility and scalability of the system are considered.Finally,the system has been tested and tested with satisfactory results.