Operation and maintenance management system for intelligent detection equipment

Abstract

The invention discloses an operation and maintenance management system for intelligent detection equipment, which relates to the technical field of operation and maintenance management of the intelligent detection equipment and comprises a detection acquisition module, an operation and maintenance analysis module, an operation and maintenance management module and a management execution module. Problems existing in an existing system include incomplete data acquisition, delayed monitoring, insufficient fault diagnosis accuracy and relatively slow response speed to equipment state change. According to the invention, the detection acquisition module is used for acquiring the state data of the company server, the operation and maintenance analysis module analyzes the state data, and the operation and maintenance management module analyzes the state data of the company server and then performs scheduling management; potential problems can be quickly responded, an alarm is triggered or a strategy is automatically adjusted, so that the downtime and the maintenance cost are reduced, and the overall operation and maintenance efficiency is improved; the utilization rate of resources is improved, the capability of dealing with emergencies is enhanced, and more reliable IT infrastructure support is provided for companies.

Description

Technical Field

[0001] The present invention relates to the technical field of operation and maintenance management of intelligent detection equipment, and in particular to an operation and maintenance management system for intelligent detection equipment. Background Art

[0002] With the rapid development of information technology, enterprises are increasingly dependent on IT infrastructure. Intelligent detection equipment plays a key role in ensuring system stability and efficient operation. The background technology of the operation and maintenance management system of intelligent detection equipment mainly includes functions such as equipment monitoring, status data collection and fault warning. By analyzing the operating status of the equipment in real time, the reliability and operating efficiency of the equipment can be improved. However, problems with the existing system include incomplete data collection, delayed monitoring, insufficient fault diagnosis accuracy, slow response to changes in equipment status, low task scheduling efficiency, lack of flexibility in task priority scheduling, and inability to reasonably allocate resources; insufficient ability to respond to emergencies, and the system is prone to instability under high load conditions; these problems lead to inefficient operation and maintenance management, increase the risk of equipment failure, and affect the overall operational results. Summary of the invention

[0003] The purpose of the present invention is to solve the above-mentioned problem and to provide an operation and maintenance management system for intelligent detection equipment.

[0004] The object of the present invention can be achieved by the following technical solutions: an operation and maintenance management system for intelligent detection equipment, including a detection acquisition module, an operation and maintenance analysis module and an operation and maintenance management module; The operation and maintenance analysis module is used to analyze the status data, specifically: Step 1: By setting the status data threshold of the corresponding server, the formula Output the energy factor YS, s is the number of servers in the company, x is the performance parameter value of a certain server, e is a natural number, and a1 and a2 are preset weight factors; multiply the status data threshold of the corresponding server by the energy factor YS to obtain the step-one initial threshold; then substitute the statistical changes in the time of the corresponding parameters of each status data into the curve graph, and obtain the maximum value exceeding the corresponding parameter step-one initial threshold within a fixed time interval from the data in the graph, as well as the maximum slope and minimum slope exceeding the corresponding parameter step-one initial threshold, shade the curve area exceeding the corresponding parameter step-one initial threshold, and count all the shadows within the fixed time interval. The sum of the areas is used to calculate the abnormal value of the state; the second step: by setting the minimum threshold of the state data of the corresponding server, and through the historical data collected over a long period of time, the peak period and the off-peak period are obtained; the corresponding time adjustment factor is set by the peak period and the off-peak period to obtain the initial threshold of step two; the third step: by combining the initial threshold of step one and the initial threshold of step two respectively; the initial threshold of step one is multiplied by the time adjustment factor Ti in the second step to obtain the threshold of step one; the initial threshold of step two is multiplied by the energy factor YS in the first step to obtain the threshold of step two; then set the initial threshold interval, normal interval, warning interval and critical interval; The operation and maintenance management module is used to perform scheduling management after analyzing the status data of the company's server. The specific process is: analyzing the priority scheduling of the company's server tasks to obtain the task priority value; then coordinating the task edge computing and the center to obtain the total time for task processing; analyzing the adaptive resource prediction of the task to obtain the resource demand value.

[0005] As a preferred embodiment of the present invention, it also includes a detection and collection module and a management execution module; The detection and collection module is used to collect the status data of the company's servers, including CPU status, memory status, disk status, network status, system temperature and power status, process and service status, and database status; The detection and acquisition module is also used to obtain the number of company servers and the performance parameters of different servers, identify and distinguish the location of each server according to the ID, and establish a 3D computer room model for display; The management execution module is mainly used to execute and adjust the output results of the operation and maintenance analysis module and the operation and maintenance management module.

[0006] As a preferred implementation of the present invention, the specific process of setting the corresponding time adjustment factor based on the obtained peak period and off-peak period to obtain the initial threshold value of step 2 is: The obtained peak period and off-peak period are set with corresponding time adjustment factors. If there is a time period of peak period, the time adjustment factors are T1, T2, T3, and T1<T2<T3. The size of the time adjustment factor is determined by the peak time period in the peak period. The time in the peak period is divided into 3 time periods, and the state data in the 3 time periods are compared. The maximum state data is matched with T3, the minimum state data is matched with T1, and the intermediate state data is matched with T2. Similarly, if there is a time period of off-peak period, the time adjustment factors are T4, T5, T6, and T4<T5<T6. The time in the off-peak period is divided into 3 time periods, and the state data in the 3 time periods are compared. The maximum state data is matched with T4, the minimum state data is matched with T6, and the intermediate state data is matched with T5. Then the minimum threshold of the state data of the corresponding server is multiplied by the time adjustment factor Ti to obtain the initial threshold of step 2, i=1 or 2 or 3 or 4 or 5 or 6.

[0007] As a preferred implementation of the present invention, the specific process of analyzing the company server task priority scheduling to obtain the task priority value is as follows: Determine the real-time requirements of the task and calculate it using the real-time requirements evaluation formula: Output the real-time coefficient Sr, tr is the deadline of the task, td is the current time, tc' is the preset time correction factor, if tr is before td, then tc1 is the time between the deadline of the task and the current time, if tr is after td, then tc2 is the time between the deadline of the task and the current time, if tr=td, the real-time coefficient Sr takes a fixed value of 0.5; The task complexity is determined by the comprehensive complexity evaluation formula: Output complexity coefficient F, Ot is time complexity, Os is space complexity, R is required resources, α1, α2 and α3 are preset weight coefficients; Determine the importance of tasks based on their impact on the overall system performance or business objectives, and give each task a critical business score, system health score, user experience score, and resource dependency score. By formula: Output the task importance coefficient P, where p1 is the business importance score, p2 is the system health impact score, p3 is the user experience score, and p4 is the resource consumption dependency score; β1, β2, β3, and β4 are all preset weight coefficients; Finally, through comprehensive judgment and classification, the priority of each task is comprehensively determined according to the importance, real-time and computational complexity coefficients of the task; tasks can be classified into high, medium and low priorities, and the specific calculation formula is: Output the task priority value RWZ, b1, b2 and b3 are the preset weights of each indicator; and substitute the data into the formula: , Pj: priority of the jth task, Tj: processing time of task j, Rj: resource requirement of the task, Cj: computing power of the server, the processing time of task A is δ unit time; compared with task B, its processing time is τ unit time, δ>τ>0; and the management execution module will give priority to task B before task A.

[0008] As a preferred implementation of the present invention, the specific process of obtaining the total time of task processing by coordinating task edge computing and center is as follows: When allocating tasks, edge servers are used to handle delay-sensitive tasks, and central servers are used to handle the delays of large-scale computing tasks. The delays of each task on the edge and central servers are calculated, and the path with the minimum delay is selected: The total time Tz for output task processing, Tb, Tw and Tx are the processing time on the edge server, network transmission time and processing time on the central server respectively. It should be noted that Tb, Tw and Tx are all obtained through the detection and acquisition module; if the task requires ηms of processing time, the central server processing requires ωms, the network transmission delay is φms, ω>φ>η>0, the management execution module will give priority to allocating the task to the edge server to reduce the overall delay.

[0009] As a preferred implementation of the present invention, the specific process of analyzing the adaptive resource prediction of the task to obtain the resource demand value is: Collect the current resource demand Dc, the resource demand Dp of the previous cycle, and the historical trend data Tt; and obtain the corresponding preset weight parameters according to the historical data for matching, which are μ1, μ2, and μ3, respectively, and 1=μ1+μ2+μ3; then calculate the predicted resource demand, and use the formula: Output resource demand value DZ, and the management execution module performs prediction and scheduling.

[0010] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention realizes efficient server status monitoring and management through the collaborative work of four modules. The detection and acquisition module collects key status data such as CPU, memory, disk, network, etc. in real time, and adopts active, passive and streaming data collection methods to ensure the accuracy and timeliness of information. The operation and maintenance analysis module automatically identifies abnormal conditions by setting dynamic thresholds and combining historical data analysis, and sets adaptive adjustment factors for different loads to ensure the stability of the system under various workloads; at the same time, through real-time comparison of status data, the system can quickly respond to potential problems, trigger alarms or automatically adjust strategies, thereby reducing downtime and maintenance costs, and improving overall operation and maintenance efficiency; it not only improves resource utilization, but also enhances the ability to respond to emergencies, providing the company with more reliable IT infrastructure support.

[0011] 2. The present invention significantly improves the resource utilization efficiency and task processing capabilities of the company's servers through the systematic task priority scheduling analysis of the operation and maintenance management module; the module comprehensively considers the real-time, complexity and importance of the task, and adopts a dynamic evaluation formula to assign a reasonable priority to each task, thereby ensuring that tasks related to key business, system health and user experience are given priority; in addition, through the coordinated scheduling of edge computing and central servers, delay-sensitive tasks can respond quickly, optimizing the overall processing time; adaptive resource prediction analysis is based on historical data and current demand, identifying resource requirements in advance to ensure stable operation of the system under high load; it not only improves the efficiency of task execution, but also reduces potential system risks, thereby providing the company with more efficient and reliable IT operation and maintenance support. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0013] Figure 1 It is a principle block diagram of the present invention; Figure 2 It is a 3D machine room model diagram of the present invention; Figure 3 This is the initial threshold curve diagram of the present invention. DETAILED DESCRIPTION

[0014] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0015] It should be understood that the terms "include" and "comprising" used in the specification and claims of the present disclosure indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0016] It should also be understood that the terms used in this disclosure are only for the purpose of describing specific embodiments and are not intended to limit the disclosure. As used in this disclosure and claims, the singular forms of "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" used in this disclosure and claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations.

[0017] See also Figure 1 As shown, an operation and maintenance management system for intelligent detection equipment includes a detection acquisition module, an operation and maintenance analysis module, an operation and maintenance management module and a management execution module; The detection and collection module collects the status data of the company's servers, including CPU status (CPU utilization, CPU load), memory status (memory usage, available memory), disk status (disk space usage, disk I / O operation rate, disk read and write latency), network status (network bandwidth usage), system temperature (CPU and motherboard temperature) and power status (power usage and voltage fluctuation), process and service status (number of active processes, operation status of specific services, service response time) and database status (number of database connections, index usage), including the following methods: actively obtain server status data by sending requests regularly through active polling method; then by passive monitoring method: collect server operation status and abnormal information in real time through server log files or event management systems (such as Syslog); then by streaming data collection: obtain and analyze server status data in real time through stream processing technology, which is suitable for environments that require timely response; See also Figure 2 As shown, the detection and acquisition module is also used to obtain the number of company servers and the performance parameters of different servers, and identify each server according to its ID (such as MAC address, host name or IP address), distinguish its location, and display it by establishing a 3D computer room model; See also Figure 3 As shown, the operation and maintenance analysis module is used to analyze the status data, specifically: Step 1: Set the status data threshold of the corresponding server. The threshold value is set according to the number of company servers and the performance parameters of different servers. Output the energy factor YS, where s is the number of servers in the company, and x is the performance parameter value of a server (by selecting and obtaining key indicators of server performance, including CPU performance (processing speed, number of cores, number of threads, etc., using benchmark tools to measure actual performance), memory performance (total memory size, memory frequency, memory bandwidth, etc., using tools such as MemTest86 or AIDA64 to test memory performance), disk performance (storage capacity, read and write speed, IOPS, etc., using operating system built-in commands (such as Linux hdparm or dd commands) to view disk performance), network performance (bandwidth, throughput, latency, etc., using network monitoring tools (such as Wireshark, Ping command) to measure network performance), system load (CPU utilization, memory utilization, disk utilization, etc., monitored in real time by system load monitoring tools such as Nagios, Zabbix, Prometheus, etc.), temperature and power consumption indicators (thermal design power consumption, energy efficiency, obtaining real-time power consumption and temperature data through server BMC), and then assign weights to each performance indicator ω o , o=1,2,3,...,f, the weight is determined according to the degree of influence of the indicator on the overall performance of the server; then, the key indicators of server performance are normalized and then Output the performance parameter value x of the server, where f is the number of performance indicators involved in the calculation, and x o ' is the normalized value pair of the oth performance indicator, the larger the x is, the higher the performance of the server is), e is a natural number, a1 and a2 are preset weight factors; the state data threshold of the corresponding server is multiplied by the energy factor YS to obtain the initial threshold of step one; Then, by statistically analyzing the time changes of the corresponding parameters of each state data and substituting them into the curve graph, we can get the initial threshold curve graph. Then, we can get the maximum value exceeding the initial threshold of the corresponding parameter step 1 within a fixed time interval and the maximum slope and minimum slope exceeding the initial threshold of the corresponding parameter step 1 from the data in the graph. Then, we can mark the curve area exceeding the initial threshold of the corresponding parameter step 1 with a shadow, and count all the shadow areas and the sum within the fixed time interval. By the formula Output state abnormal value YD, Z max is the maximum value within the initial threshold of the corresponding parameter step 1, M is the value of the sum of all shadow areas within the initial threshold of the corresponding parameter step 1, k max is the maximum slope exceeding the initial threshold of the corresponding parameter step 1, k min is the minimum slope exceeding the initial threshold of the corresponding parameter step 1, a3, a4 and a5 are preset weight factors; Step 2: By setting the minimum threshold of the status data of the corresponding server (it is abnormal if the status data is lower than the low threshold within a fixed time), and through the historical data collected over a long period of time, obtain data covering peak periods and off-peak periods, such as data when the load changes or the task is heavy; set the corresponding time adjustment factor based on the peak period and off-peak period obtained. If there is a time period that is a peak period, the time adjustment factor is T1, T2, T3, and T1<T2<T3. The specific time adjustment factor size is determined by the peak time period within the peak period. The time within the peak period is divided into 3 equal time periods, and the status data within the 3 time periods are compared. The maximum status The data is matched with T3, the minimum state data is matched with T1, and the intermediate state data is matched with T2. Similarly, if there is a time period of off-peak period, the time adjustment factor is T4, T5, T6, and T4<T5<T6. The time in the off-peak period is divided into 3 equal time periods, and the state data in the 3 time periods are compared. The maximum state data is matched with T4, the minimum state data is matched with T6, and the intermediate state data is matched with T5; then the minimum threshold of the state data of the corresponding server is multiplied by the time adjustment factor Ti to obtain the initial threshold of step 2 corresponding to Ti, i=1 or 2 or 3 or 4 or 5 or 6; Step 3: Calculate by combining the initial threshold of step 1 and multiple initial thresholds of step 2 respectively; multiply the initial threshold of step 1 by the time adjustment factor Ti in the second step to obtain the threshold of step 1 corresponding to Ti; multiply the initial threshold of step 2 corresponding to Ti by the digital energy factor YS in the first step to obtain the threshold of step 2 corresponding to Ti; Then, by setting the initial threshold interval as the normal interval, the normal interval is the middle area between the step one threshold and the step two threshold, and by setting the warning interval and critical interval; setting the normal interval: use the median and moving average in the historical data as the middle value of the normal working range of the server status, and then calculate the normal fluctuation range through the standard deviation (σ) or interquartile range (IQR) of the historical data, and set the upper and lower limits of the normal interval, usually set to the median plus or minus 2 times the standard deviation (ie ±2σ), or calculate the normal range through the interquartile range (Q1-1.5IQR to Q3+1.5IQR); setting the warning interval: expand the upper limit of the normal interval, determine the starting point of the warning interval through the 80th to 90th percentile in the historical data, and determine the upper limit of the warning interval by observing the operating data of the server when it is close to full load; such as CPU usage When it reaches 90%-95%, it can be regarded as a warning state; the critical interval (the critical interval is usually the range close to the resource limit, determined by the 99th percentile data) is set: by observing the status data when serious problems (such as system crashes, overloads) occurred in the past, the threshold is set; for example, when the CPU usage rate exceeds 95%, it enters the critical interval; and the collected status data is compared with the set threshold interval in real time; once the data approaches the warning interval or exceeds the critical interval, the management execution module triggers an alarm or automatic adjustment; (Automatic adjustment: When the system triggers an alarm multiple times or is in a critical state, the dynamic threshold interval will be automatically readjusted according to the actual operation data; for example, by detecting whether the system is frequently in the warning interval, the threshold is appropriately relaxed or tightened, and then the threshold setting process is continuously optimized through analysis of the system's historical alarms and adjustment records) The operation and maintenance management module analyzes the status data of the company's servers and performs scheduling management. The specific process is as follows: Through the company's server task priority scheduling analysis; first determine the real-time requirements of the task, and then calculate it through the real-time requirement evaluation formula: Output the real-time coefficient Sr, tr is the deadline of the task, td is the current time, tc' is the preset time correction factor, if tr is before td, then tc1 is the duration between the deadline of the task and the current time, if tr is after td, then tc2 is the duration between the deadline of the task and the current time, if tr=td, the real-time coefficient Sr takes a fixed value of 0.5; it should be noted that: the higher the real-time requirement, the lower the Sr value; then the task complexity is determined by the comprehensive complexity evaluation formula: Output complexity coefficient F, Ot is time complexity, Os is space complexity, R is required resource amount (such as CPU, memory, bandwidth), α1, α2 and α3 are preset weight coefficients; then judge the importance of the task, the importance of the task is determined according to the impact of the task on the overall performance of the system or business objectives, and each task is scored for key business (tasks related to core business processes have a higher priority, such as tasks involving security and emergency handling), system health score (tasks that have a significant impact on system stability or health have a higher priority, such as status monitoring, fault detection), user experience score (tasks that affect user experience, such as real-time feedback and rapid response operations, usually have a higher priority) and resource dependency score (tasks that consume a large amount of key resources may affect the execution of other tasks, so it is necessary to reasonably allocate priorities); then through the formula: Output the task importance coefficient P, where p1 is the business importance score, p2 is the system health impact score, p3 is the user experience score, and p4 is the resource consumption dependency score; β1, β2, β3, and β4 are all preset weight coefficients; finally, through comprehensive judgment and classification, the priority of each task is comprehensively determined according to the task importance, real-time and computational complexity coefficients; tasks can be classified according to high, medium and low priority, and the specific calculation formula is: Output the task priority value RWZ, b1, b2 and b3 are the preset weights of various indicators; for example: assume that the priorities of task A and task B are 2 and 1 respectively, and the weights are 0.5 and 1 respectively; if task A requires 20 units of resources and the server provides 10 units of resources, and substitute the data into the formula: , where Pj: the priority of the jth task, Tj: the processing time of task j, Rj: the resource requirement of the task, Cj: the computing power of the server, then the processing time of task A is δ unit time; compared with task B, its processing time is τ unit time, δ>τ>0; then the management execution module will prioritize task B over task A; Then, through the coordinated scheduling of edge computing and center tasks; when allocating tasks, edge servers are used to handle delay-sensitive tasks (tasks that require fast response and are time-sensitive, such as real-time video monitoring, automatic control instructions, etc.), and central servers are used to handle the delay of large-scale computing tasks. The delay of each task on the edge and central servers is calculated, and the path with the smallest delay is selected: Output the total time of task processing Tz, Tb, Tw and Tx are the processing time on the edge server, network transmission time and processing time on the central server respectively. It should be noted that Tb, Tw and Tx are all obtained through the detection and acquisition module; if the task requires ηms of processing time, the central server processing requires ωms, the network transmission delay is φms, ω>φ>η>0, the management execution module will give priority to assigning the task to the edge server to reduce the overall delay; Then analyze the adaptive resource prediction of the task; collect the current resource demand (Dc), the resource demand of the previous cycle (Dp) and the historical trend data (Tt); and obtain the corresponding preset weight parameters according to the historical data for matching, which are μ1, μ2 and μ3 respectively, and 1=μ1+μ2+μ3; then calculate the predicted resource demand and use the formula: The resource demand value DZ is output and predicted and scheduled by the management execution module. When the resource demand value DZ exceeds the preset demand threshold 1, resources are automatically increased; when the resource demand value DZ is lower than the preset demand threshold 2, unnecessary resources are released to avoid waste.

[0018] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. An operation and maintenance management system for intelligent detection equipment, comprising an operation and maintenance analysis module and an operation and maintenance management module, characterized in that: The operation and maintenance analysis module is used to analyze the status data, specifically as follows: the first step: by setting the status data threshold of the corresponding server, analyzing the digital energy factor, multiplying the status data threshold of the corresponding server by the digital energy factor to obtain the step one initial threshold; then, by statistically analyzing the time changes of the parameters corresponding to each status data and substituting them into the curve graph, the maximum value exceeding the corresponding parameter step one initial threshold within a fixed time interval, as well as the maximum slope and minimum slope exceeding the corresponding parameter step one initial threshold are obtained, the curve area exceeding the corresponding parameter step one initial threshold is shaded, the sum of all shaded areas within the fixed time interval is statistically calculated, and the status abnormal value is calculated; Step 2: Establish the minimum threshold of the status data of the corresponding server, and collect historical data for a long time to obtain data covering peak and off-peak periods, and set the corresponding time adjustment factors for peak and off-peak periods to obtain the initial threshold of step 2; Step 3: Multiply the initial threshold of step 1 by the time adjustment factor in step 2 to obtain the threshold of step 1; multiply the initial threshold of step 2 by the energy factor in step 1 to obtain the threshold of step 2; then set the initial threshold area, normal interval, warning interval and critical interval; The operation and maintenance management module is used to perform scheduling management after analyzing the status data of the company's server. The specific process is: analyzing the priority scheduling of the company's server tasks to obtain the task priority value; then coordinating the task edge computing and the center to obtain the total time for task processing; analyzing the adaptive resource prediction of the task to obtain the resource demand value.

2. The operation and maintenance management system for intelligent detection equipment according to claim 1, characterized in that: It also includes a detection and acquisition module and a management execution module; The detection and acquisition module is used to collect the status data of the company's servers, including CPU status, memory status, disk status, network status, system temperature and power status, process and service status, and database status. It is also used to obtain the number of company servers and the performance parameters of different servers, and identify and distinguish the location of each server according to the ID, and establish a 3D computer room model for display; The management execution module is used to execute and adjust the output results of the operation and maintenance analysis module and the operation and maintenance management module.

3. The operation and maintenance management system for intelligent detection equipment according to claim 1, characterized in that: The specific process of obtaining the initial threshold value of step 2 is as follows: The obtained peak period and off-peak period are set with corresponding time adjustment factors. If there is a time period of peak period, the time adjustment factors are T1, T2, T3, and T1<T2<T3. The size of the time adjustment factor is determined by the peak time period in the peak period. The time in the peak period is divided into three time periods. The state data in the three time periods are compared, the maximum state data is matched with T3, the minimum state data is matched with T1, and the middle state data is matched with T2. Similarly, if there is a time period of off-peak period, the time adjustment factors are T4, T5, T6, and T4<T5<T6. The time in the off-peak period is divided into three time periods. The state data in the three time periods are compared, the maximum state data is matched with T4, the minimum state data is matched with T6, and the middle state data is matched with T5. Then the minimum threshold of the state data of the corresponding server is multiplied by the time adjustment factor Ti to obtain the initial threshold of step two, i=1 or 2 or 3 or 4 or 5 or 6.

4. The operation and maintenance management system for intelligent detection equipment according to claim 1, characterized in that: The specific process of analyzing the company server task priority scheduling to obtain the task priority value is as follows: Determine the real-time requirements of the task and calculate it using the real-time requirements evaluation formula: Output the real-time coefficient Sr, tr is the deadline of the task, td is the current time, tc' is the preset time correction factor, if tr is before td, then tc1 is the time between the deadline of the task and the current time, if tr is after td, then tc2 is the time between the deadline of the task and the current time, if tr=td, the real-time coefficient Sr takes a fixed value of 0.5; The task complexity is determined by the comprehensive complexity evaluation formula: Output complexity coefficient F, Ot is time complexity, Os is space complexity, R is required resources, α1, α2 and α3 are preset weight coefficients; Determine the importance of tasks based on their impact on the overall system performance or business objectives, and give each task a critical business score, system health score, user experience score, and resource dependency score. By formula: Output the task importance coefficient P, where p1 is the business importance score, p2 is the system health impact score, p3 is the user experience score, and p4 is the resource consumption dependency score; β1, β2, β3, and β4 are all preset weight coefficients; Finally, a comprehensive classification is performed to determine the priority of each task based on the task's importance, real-time, and computational complexity. Tasks can be classified into high, medium, and low priority levels. The specific calculation formula is: Output the task priority value RWZ, b1, b2 and b3 are the preset weights of various indicators; And substitute the data into the formula: , Pj: priority of the jth task, Tj: processing time of task j, Rj: resource requirement of the task, Cj: computing power of the server, the processing time of task A is δ unit time; compared with task B, its processing time is τ unit time, δ>τ>0; and the management execution module will give priority to task B before task A.

5. The operation and maintenance management system for intelligent detection equipment according to claim 1, characterized in that: The specific process of obtaining the total time of task processing by coordinating task edge computing and center is as follows: When allocating tasks, edge servers are used to handle delay-sensitive tasks, and central servers are used to handle the delays of large-scale computing tasks. The delays of each task on the edge and central servers are calculated, and the path with the minimum delay is selected: The total time for output task processing is Tz, Tb, Tw and Tx, which are the processing time on the edge server, network transmission time and processing time on the central server respectively; Tb, Tw and Tx are all acquired through the detection and acquisition module; if the task requires ηms of processing time, the central server requires ωms for processing, the network transmission delay is φms, ω>φ>η>0, the management execution module will prioritize the task to the edge server to reduce the overall delay.

6. The operation and maintenance management system for intelligent detection equipment according to claim 5, characterized in that: The specific process of analyzing the adaptive resource prediction of the task to obtain the resource demand value is as follows: Collect the current resource demand Dc, the resource demand Dp of the previous cycle, and the historical trend data Tt; and obtain the corresponding preset weight parameters according to the historical data for matching, which are μ1, μ2, and μ3, respectively, and 1=μ1+μ2+μ3; then calculate the predicted resource demand, and use the formula: Output resource demand value DZ, and the management execution module performs prediction and scheduling.

Images