Design And Implementation Of SSD Monitor System

With the rapid development of information technology in recent years,the business system has higher and higher requirements for processing data volume,concurrent requests,and throughput.Based on the traditional magnetic hard disk database system,disk 10 has always been a big bottleneck.Most engineers are beginning to shift their focus from traditional mechanical hard drive solutions to better SSD solutions.SSD(Solid State Drive)is considered a new generation of HHD(hard disk drive)replacement,SSD will be an important part of system planning.In order to ensure business stability,monitoring of SSD solid state drives is essential.It is very meaningful to realize a monitoring system that is low-cost,predicts the remaining life of the SSD,and can monitor the storage,reading and writing rate,and dynamic bad block number of the solid-state hard disk.This paper is derived from the actual engineering project-SSD disk monitoring system.By installing the Telegraf plugin,the plugin can execute the python acquisition program,obtain the collection interval of each monitoring item and the corresponding data source by reading the configuration file,complete the data collection task,and integrate the data.Analyze abnormal data,and send detailed and detailed alarm information to relevant responsible persons in time,and display the integrated data on the Web platform for user's access.At the same time,in order to reduce the number of alarms and ensure the quality of the alarms,this project also makes some measures.From the source,the monitoring platform supports multi-point judgment during configuration monitoring.The user can set consecutive n-point anomalies or n points in the historical m-points.When an exception occurs,an alarm is triggered to reduce the alarm misjudgment caused by data jitter.After the alarm is generated,the system controls the alarm interval and the number of alarms,and combines multiple alarm information intelligently in combination with the machine learning method.The alarm entry is reduced while ensuring the quality of the alarm information.In solving the alarm,the system combines statistical knowledge,performs root cause analysis on a certain alarn information,and assists the user in troubleshooting.First,the author participated in the need analysis of the entire project and the prototype design of the Web platform.Secondly,in the design and development stage,using the python Web development framework Flask,the author has independently completed the design and development of the monitoring data acquisition module,alarm module,alarm intelligent merge module and Web platform module.In the test phase,the author completed the user management module,alarm module,host management module,monitoring information management module,the test case design of the alarm module and carried out a series of test work,and verified the system compatibility and performance.Ensure that the system meets the established requirements.At present,the SSD disk monitoring system completed in the paper is in the trial operation stage.The data of the Web monitoring platform is accurate,and the alarms can be detected and reported in time,which greatly releases the manpower.