Adaptive operation and maintenance monitoring method and system for multiple live broadcasting rooms

Abstract

The invention discloses an adaptive operation and maintenance monitoring method and system for multiple live broadcast rooms, and relates to the technical field of live broadcast room operation and maintenance, the technical scheme is characterized by comprising the following steps: obtaining operation data of multiple live broadcast rooms, and each live broadcast room is connected with a live broadcast platform server through a network link; wherein the operation data comprises equipment operation data, network state data, interaction data and display data; analyzing the equipment operation data, the network state data, the interaction data and the display data to obtain an operation state deviation interval of the live broadcast room when the state of the live broadcast room is abnormal, judging an operation abnormal type of the live broadcast room according to the operation state deviation interval, and obtaining performance vacancy of the live broadcast room according to the operation abnormal type; wherein the performance vacancy is a deviation parameter of an expected live broadcast effect and an actual live broadcast effect corresponding to the live broadcast room; the method has the advantages that through continuous adaptive operation and maintenance monitoring, different live broadcast scenes and emergencies can be adapted continuously.

Description

Technical Field

[0001] The present invention relates to the technical field of live broadcast room operation and maintenance, and more specifically, it relates to a method and system for adaptive operation and maintenance monitoring of multiple live broadcast rooms. Background Art

[0002] With the booming development of the live broadcast industry, the number of live broadcast rooms is increasing day by day, and the live broadcast scenarios are becoming increasingly complex and diverse. The operation and maintenance management of multiple live broadcast rooms faces many challenges. Traditional live broadcast operation and maintenance management methods mostly rely on manual inspections and simple system monitoring indicators, and it is difficult to obtain the operation data of live broadcast rooms comprehensively and in real time. For example, by only observing conventional indicators such as the CPU and memory usage of the server, it is impossible to accurately determine whether the live broadcast screen freezes due to complex reasons such as device performance problems, network transmission failures, or untimely processing of interactive data. For some subtle performance changes and potential fault hazards, traditional monitoring means are often difficult to detect, resulting in the accumulation of problems and concentrated outbreaks at critical moments, affecting the normal progress of the live broadcast. Summary of the Invention

[0003] Aiming at the deficiencies of the existing technology, the purpose of the present invention is to provide a method and system for adaptive operation and maintenance monitoring of multiple live broadcast rooms.

[0004] To achieve the above purpose, the present invention provides the following technical solutions: An adaptive operation and maintenance monitoring method for multiple live broadcast rooms, the method includes the following steps: Obtain the operation data of multiple live broadcast rooms, and each live broadcast room is connected to the live broadcast platform server through a network link; wherein, the operation data includes device operation data, network status data, interactive data, and display data; When an abnormality occurs in the live broadcast room situation, analyze the device operation data, network status data, interactive data, and display data to obtain the operation status deviation range of the live broadcast room, judge the type of operation abnormality of the live broadcast room according to the operation status deviation range, and obtain the performance deficit of the live broadcast room according to the type of operation abnormality; wherein, the performance deficit is the deviation parameter between the corresponding expected live broadcast effect and the actual live broadcast effect of the live broadcast room; According to the expected live broadcast effect and the actual live broadcast effect of the live broadcast room in the deviation parameter, obtain the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time, and analyze the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time to obtain the first operation and maintenance coefficient of the live broadcast room; Classify various live broadcast tasks of the live broadcast room into guarantee tasks and non-urgent tasks, and analyze to obtain the second operation and maintenance coefficient of the live broadcast room according to the guarantee tasks and non-urgent tasks; Perform resource operation and maintenance processing on the live broadcast room according to the first operation and maintenance coefficient and the second operation and maintenance coefficient.

[0005] Preferably, it further includes: Set the target repair duration of the live broadcast room and obtain the time when the live broadcast room has an operation anomaly; Obtain the target repair time of the live broadcast room based on the time when the live broadcast room has an operation anomaly and the target repair duration.

[0006] Preferably, when the condition of the live broadcast room is abnormal, analyze the device operation data, network status data, interaction data, and display data to obtain the operation status deviation range of the live broadcast room. Specifically: Obtain the device operation status curve within the target repair time based on the device operation data; Obtain the network transmission quality curve within the target repair time based on the network status data; Obtain the audience interaction activity curve within the target repair time based on the interaction data; Obtain the content display effect curve within the target repair time based on the display data; Obtain the expected live broadcast effect range of the live broadcast room based on the device operation status curve, network transmission quality curve, and audience interaction activity curve, and obtain the actual live broadcast effect range of the live broadcast room based on the content display effect curve; And obtain the operation status deviation range of the live broadcast room based on the expected live broadcast effect range and the actual live broadcast effect range.

[0007] Preferably, judge the operation anomaly type of the live broadcast room based on the operation status deviation range, and obtain the performance deficit of the live broadcast room based on the operation anomaly type. Specifically: Set at least one standard deviation range, each standard deviation range corresponds to an operation anomaly type, and each operation anomaly type corresponds to a performance deficit; Compare the operation status deviation range of the live broadcast room with the standard deviation ranges, and mark the standard deviation range where the operation status deviation range of the live broadcast room is located as the target deviation range; Mark the operation anomaly type corresponding to the target deviation range as the operation anomaly type of the live broadcast room, and mark the performance deficit corresponding to the operation anomaly type as the performance deficit of the live broadcast room.

[0008] Preferably, obtain the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time according to the expected live broadcast effect and the actual live broadcast effect of the live broadcast room in the deviation parameter. Specifically: If the expected live broadcast effect corresponding to the live broadcast room in the deviation parameter is greater than or equal to the actual live broadcast effect, calculate the difference between the expected live broadcast effect and the actual live broadcast effect corresponding to the live broadcast room to obtain the optimization adjustment amount required to improve the live broadcast effect within the target repair time; And obtain the average workload of optimization and adjustment per unit time based on the optimization and adjustment amount and the target emergency repair duration, and perform optimization and adjustment on this live broadcast room according to the average workload and the optimization and adjustment amount; If the expected live broadcast effect of this live broadcast room in the deviation parameter is less than the actual live broadcast effect, then calculate the difference between the expected live broadcast effect and the actual live broadcast effect corresponding to this live broadcast room to obtain the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time.

[0009] Preferably, analyze the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time to obtain the first operation and maintenance coefficient of this live broadcast room, specifically: Mark the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time as the guarantee requirement; Obtain the unit resource guarantee cost corresponding to the unit live broadcast task in this live broadcast room; Obtain the total guarantee cost for resource guarantee of all live broadcast tasks according to the guarantee requirement and the unit resource guarantee cost of this live broadcast room; Obtain the first operation and maintenance coefficient of this live broadcast room according to the total guarantee cost.

[0010] Preferably, classify various live broadcast tasks of this live broadcast room into guarantee tasks and non-urgent tasks, specifically: Obtain the task types of each live broadcast task in this live broadcast room; Classify each live broadcast task of this live broadcast room according to the pre-set task classification standard to obtain the task category corresponding to each live broadcast task; among them, the task category includes the core task category, the important task category, and the ordinary task category; Set delayable tasks based on the core task category, the important task category, and the ordinary task category; Divide the live broadcast tasks into guarantee tasks and non-urgent tasks according to the delayable tasks.

[0011] Preferably, analyze and obtain the second operation and maintenance coefficient of this live broadcast room according to the guarantee tasks and non-urgent tasks, specifically: Obtain the total guarantee cost for resource guarantee of the guarantee tasks according to the guarantee tasks and the unit resource guarantee cost corresponding to the unit live broadcast task in this live broadcast room; Obtain the unit task delay compensation cost corresponding to the unit live broadcast task in this live broadcast room; Obtain the total compensation cost for non-urgent tasks according to the non-urgent tasks and the unit task delay compensation cost corresponding to the unit live broadcast task in this live broadcast room; Obtain the second operation and maintenance coefficient of this live broadcast room according to the total guarantee cost and the total compensation cost.

[0012] Preferably, resource operation and maintenance processing is performed on the live broadcast room according to the first operation and maintenance coefficient and the second operation and maintenance coefficient. Specifically: If the first operation and maintenance coefficient is greater than or equal to the second operation and maintenance coefficient, a first resource guarantee value corresponding to the live broadcast room is obtained based on the guarantee requirements and the target emergency repair duration of the live broadcast room, and resource operation and maintenance processing is performed on all live broadcast tasks of the live broadcast room according to the first resource guarantee value; If the first operation and maintenance coefficient is less than the second operation and maintenance coefficient, a second resource guarantee value corresponding to the live broadcast room is obtained based on the guarantee tasks and the target emergency repair duration of the live broadcast room, and resource operation and maintenance processing is performed on the guarantee tasks of the live broadcast room according to the second resource guarantee value.

[0013] An adaptive operation and maintenance monitoring system for multiple live broadcast rooms includes: An acquisition module that acquires the operation data of multiple live broadcast rooms, and each live broadcast room is connected to the live broadcast platform server through a network link; wherein, the operation data includes device operation data, network status data, interaction data, and display data; A judgment module that analyzes the device operation data, network status data, interaction data, and display data when the status of the live broadcast room is abnormal to obtain the operation status deviation range of the live broadcast room, judges the type of operation abnormality of the live broadcast room according to the operation status deviation range, and obtains the performance deficit of the live broadcast room according to the type of operation abnormality; wherein, the performance deficit is the deviation parameter between the expected live broadcast effect and the actual live broadcast effect corresponding to the live broadcast room; A first analysis module that obtains the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time according to the expected live broadcast effect and the actual live broadcast effect of the live broadcast room in the deviation parameter, and analyzes the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time to obtain the first operation and maintenance coefficient of the live broadcast room; A second analysis module that classifies various live broadcast tasks of the live broadcast room into guarantee tasks and non-emergency tasks, and analyzes the second operation and maintenance coefficient of the live broadcast room according to the guarantee tasks and non-emergency tasks; A processing module that performs resource operation and maintenance processing on the live broadcast room according to the first operation and maintenance coefficient and the second operation and maintenance coefficient.

[0014] Compared with the prior art, the present invention has the following beneficial effects: This application comprehensively acquires the device operation data, network status data, interaction data, and display data of multiple live broadcast rooms. By acquiring these multi-dimensional and all-round data, the accuracy and comprehensiveness of fault diagnosis are improved. The operation status deviation range obtained through analysis can accurately judge the type of operation abnormality, and then clarify the performance deficit. This judgment avoids misjudgment and inefficient troubleshooting caused by relying on experience or fuzzy judgment in the past, greatly saving operation and maintenance time and resources, and improving the reliability of the live broadcast system.

[0015] Determine the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time according to the performance deficit, and obtain the first operation and maintenance coefficient through analysis. At the same time, classify the live broadcast tasks into guarantee tasks and non-urgent tasks, analyze and obtain the second operation and maintenance coefficient according to the task characteristics, and perform resource operation and maintenance processing based on the first operation and maintenance coefficient and the second operation and maintenance coefficient to achieve intelligent optimization of resource allocation. In the case of limited resources, prioritize to meet the resource requirements of guarantee tasks to ensure that the core live broadcast effect is not affected; for non-urgent tasks, reasonably arrange according to the remaining resources, which improves the resource utilization efficiency, avoids excessive investment or unreasonable allocation of resources, ensures that the live broadcast can operate with the optimal resource allocation in different scenarios, reduces the operation cost while improving the live broadcast quality.

[0016] This application enables the live broadcast room to respond quickly in the face of various abnormal situations. Whether it is a sudden device failure, a temporary network fluctuation, or a sudden increase in the number of viewer interactions, etc., it can detect problems in time through data monitoring and analysis, and quickly take targeted measures based on the determined abnormal type and operation and maintenance coefficient. Through continuous adaptive operation and maintenance monitoring, the live broadcast system can continuously adapt to different live broadcast scenarios and emergencies, maintain a good operating state, better meet user needs, and provide strong technical support for the long-term stable development of the live broadcast business. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the steps of a method for adaptive operation and maintenance monitoring of multiple live broadcast rooms proposed by the present invention; Figure 2 It is a schematic diagram of the steps of dividing guarantee tasks and non-urgent tasks in a method for adaptive operation and maintenance monitoring of multiple live broadcast rooms proposed by the present invention; Figure 3 It is a schematic diagram of the modules of a system for adaptive operation and maintenance monitoring of multiple live broadcast rooms proposed by the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0018] Refer to Figures 1 to 3 as shown.

[0019] Embodiment 1 further describes a method for adaptive operation and maintenance monitoring of multiple live broadcast rooms proposed by the present invention.

[0020] A method for adaptive operation and maintenance monitoring of multiple live broadcast rooms, the method includes the following steps: Obtain the operation data of multiple live broadcast rooms, and each live broadcast room is connected to the live broadcast platform server through a network link; wherein, the operation data includes device operation data, network status data, interaction data and display data; When the live broadcast room is in an abnormal state, analyze the device operation data, network status data, interaction data, and display data to obtain the operation status deviation range of the live broadcast room. Determine the type of operation abnormality of the live broadcast room based on the operation status deviation range, and obtain the performance deficit of the live broadcast room based on the type of operation abnormality. Among them, the performance deficit is the deviation parameter between the expected live broadcast effect and the actual live broadcast effect corresponding to the live broadcast room. Obtain the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time according to the expected live broadcast effect and the actual live broadcast effect in the deviation parameter, and analyze the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time to obtain the first operation and maintenance coefficient of the live broadcast room. Classify various live broadcast tasks of the live broadcast room into guarantee tasks and non-urgent tasks, and analyze the second operation and maintenance coefficient of the live broadcast room based on the guarantee tasks and non-urgent tasks. Perform resource operation and maintenance processing on the live broadcast room according to the first operation and maintenance coefficient and the second operation and maintenance coefficient.

[0021] Each live broadcast room of this application is connected to the live broadcast platform server through a network link, and uses data collection tools or interfaces to continuously obtain operation data from multiple data sources such as live broadcast room devices, network nodes, and the background of the live broadcast platform in real time. For device operation data, install sensors on the device to collect hardware performance parameters of the device, such as the usage of CPU, memory, hard disk, etc., and physical indicators such as the temperature and fan speed of the device. These data reflect the real-time working state of the device. Network status data is obtained through network monitoring tools, including network bandwidth occupancy, data transmission delay, packet loss rate, etc., which are used to evaluate the transmission quality and stability of the network. Interaction data is extracted from the database of the live broadcast platform, covering behavior records such as messages, likes, votes, and shares of the audience during the live broadcast, reflecting the participation and interest points of the audience. Display data is obtained by parsing the live video stream, such as the resolution, frame rate, encoding format, color mode, etc. of the picture.

[0022] When an abnormal situation occurs in the live broadcast room, such as video freezing, sound interruption, interaction messages not being displayed, etc., the system triggers an abnormal analysis process. First, perform multi-dimensional correlation analysis on the stored device operation data, network status data, interaction data, and display data. For example, when it is found that the live video freezes, check whether the CPU and memory usage of the device are too high (device operation data), whether the network bandwidth is insufficient or the delay is too large (network status data), whether the server load is too high due to a sudden increase in the amount of audience interaction (interaction data), and whether the frame rate and resolution of the video stream are abnormal (display data).

[0023] Determine the operating state deviation range through comprehensive analysis. For example, through analysis, it is found that the network latency exceeds the normal threshold by 50 ms and the CPU usage rate of the device reaches more than 90%, which constitutes the operating state deviation range of the current live broadcast room. Based on this deviation range, accurately determine the type of operating anomaly according to the preset anomaly type judgment rules, such as network transmission anomaly, device performance bottleneck, etc. Further, calculate the performance deficit, that is, the deviation parameter between the two, by comparing the expected live broadcast effect (such as set smooth picture, real-time interaction response, etc.) and the actual live broadcast effect (the performance after being affected by the anomaly).

[0024] First, evaluate the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time. Taking video freezing as an example, if the performance deficit is caused by insufficient graphics processing power of the current device resulting in a frame rate lower than expected, then the amount of guarantee resources may include increasing the computing resources of the graphics processing unit (GPU), such as allocating more GPU cores or increasing the operating frequency of the GPU.

[0025] Analyze the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time to obtain the first operation and maintenance coefficient of this live broadcast room. This coefficient reflects the relative degree and comprehensive index of the operation and maintenance resources required to make up for the performance deficit and maintain the live broadcast effect, and is an important basis for subsequent resource allocation and operation and maintenance decision-making.

[0026] According to factors such as the nature, importance, and time urgency of the live broadcast tasks, classify various live broadcast tasks in the live broadcast room into guarantee tasks and non-urgent tasks. Guarantee tasks are usually tasks that play a decisive role in the normal progress of the live broadcast and the core effect, such as the key explanation links of the anchor, the display links of important products, and interactive links such as real-time lottery draws. Once problems occur in these tasks, it will directly affect the viewing experience of the audience and the achievement of the core goal of the live broadcast. Non-urgent tasks are relatively less important, can be appropriately delayed for processing, and have less impact on the key effects of the live broadcast, such as the collation and analysis of audience messages after the live broadcast, and the archiving of historical live broadcast data.

[0027] Analyze the resource requirement characteristics of guarantee tasks and non-urgent tasks respectively, including the demand intensity and time distribution of device resources, network resources, human resources, etc. For example, guarantee tasks may have higher requirements for the computing resources and network bandwidth of the device during the live broadcast and require real-time response, while non-urgent tasks have relatively lower resource requirements and can be carried out during idle periods of the live broadcast. Considering these factors comprehensively, calculate the second operation and maintenance coefficient through a specific evaluation model. This coefficient reflects the different characteristics of resource requirements and operation and maintenance management for different task types.

[0028] Take the first operation and maintenance coefficient and the second operation and maintenance coefficient as the key basis for resource operation and maintenance decision-making. First, determine the main resource input direction and scale required to solve the performance deficit according to the first operation and maintenance coefficient. If the first operation and maintenance coefficient is high, it indicates that the performance deficit is serious, and a large amount of resources need to be preferentially allocated for repair, such as increasing server computing resources, upgrading network devices, etc.

[0029] Based on the second operation and maintenance coefficient, rationally allocate the remaining resources on the basis of ensuring that the main resource input meets the performance repair requirements. For guarantee tasks, ensure that they have sufficient resource support during the live broadcast and give priority to ensuring their smooth progress; for non-urgent tasks, arrange them flexibly according to the remaining resources, such as allocating resources during the low peak period of the live broadcast to complete relevant tasks. In this way, the dynamic and optimized configuration of the live broadcast room resources is realized, ensuring that the live broadcast system can operate stably and efficiently under different circumstances and improving the overall live broadcast service quality.

[0030] It also includes: Set the target emergency repair duration of this live broadcast room and obtain the time when the live broadcast room has an operation anomaly; Based on the time when the live broadcast room has an operation anomaly and the target emergency repair duration, obtain the target repair time of this live broadcast room.

[0031] This application pre-sets a reasonable target emergency repair duration according to factors such as the business type, importance, and audience scale of the live broadcast room. For example, for large and popular entertainment live broadcasts, due to the large number of audiences and high real-time attention, a shorter target emergency repair duration is set, such as 10 minutes; while for some small-scale knowledge popularization live broadcasts with strong professionalism but relatively few audiences, the target emergency repair duration can be appropriately relaxed to 20 minutes. This duration is a reference standard for the upper limit of the time for the operation and maintenance team to carry out fault repair work, reflecting the differential consideration of the fault tolerance and repair urgency of different live broadcast rooms.

[0032] Through the monitoring tools and sensors deployed in the live broadcast room equipment, network nodes, and the live broadcast platform background, the running status of the live broadcast room is monitored in real time. If an operation anomaly is detected, such as equipment error reporting, network interruption, video freeze, etc., immediately record the exact time when the anomaly occurs. This time information is one of the important basic data for calculating the target repair time.

[0033] Based on the time when the live broadcast room experiences an abnormal operation obtained and the preset target emergency repair duration, the target repair time is calculated through a simple time calculation method. That is, the target repair time = the time when the abnormal operation occurs + the target emergency repair duration. For example, if a live broadcast room experiences a network lag abnormality at 10 am and its set target emergency repair duration is 15 minutes, then the target repair time for this live broadcast room is 10:15 am. The operation and maintenance team takes this as a time node and arranges a series of tasks such as fault troubleshooting, resource allocation, and repair implementation around this time to ensure the normal operation of the live broadcast room is restored within the specified time.

[0034] When the situation in the live broadcast room is abnormal, analyze the device operation data, network status data, interaction data, and display data to obtain the deviation range of the operation status of this live broadcast room. Specifically: Obtain the device operation status curve within the target repair time based on the device operation data; Obtain the network transmission quality curve within the target repair time based on the network status data; Obtain the viewer interaction activity curve within the target repair time based on the interaction data; Obtain the content display effect curve within the target repair time based on the display data; Obtain the expected live broadcast effect range of this live broadcast room based on the device operation status curve, network transmission quality curve, and viewer interaction activity curve, and obtain the actual live broadcast effect range of this live broadcast room based on the content display effect curve; And obtain the deviation range of the operation status of this live broadcast room based on the expected live broadcast effect range and the actual live broadcast effect range.

[0035] This application continuously collects device operation data during the operation of the live broadcast room, such as CPU usage rate, memory occupancy rate, hard disk read and write speed, device temperature, etc. Taking time as the abscissa and various device performance indicators as the ordinate, draw the device operation status curve within the target repair time. This curve dynamically reflects the change of device performance over time. For example, if the CPU usage rate gradually increases during a certain period, it indicates that the device may be facing performance pressure.

[0036] Real-time obtain network status data, including network bandwidth, latency, packet loss rate, etc., and draw the network transmission quality curve in chronological order. The abscissa is time and the ordinate is the numerical values of various network indicators. Through this curve, the transmission status of the network within the target repair time can be intuitively understood. For example, if the network latency suddenly increases, it may indicate that the network is congested or faulty.

[0037] Statistically analyze the interactive data (such as the number of audience messages, likes, and gift - giving times, etc.). Taking time as the axis and the amount of interactive data within a unit of time as the vertical coordinate, draw the audience interaction activity curve, which can reflect the change in the audience's participation during the live broadcast. If the interaction activity drops sharply during a certain period of the curve, it may mean that there are problems with the live - broadcast content or form.

[0038] Draw the content display effect curve based on the display data (such as picture resolution, frame rate, color saturation, etc.). The abscissa is time, and the ordinate is the value of the content display - related indicators. This curve reflects the presentation quality of the live - broadcast content within the target repair time. For example, a sudden drop in the frame rate will cause the picture to freeze and affect the audience's viewing experience.

[0039] Determine an expected live - broadcast effect interval by synthesizing the device operation status curve, network transmission quality curve, and audience interaction activity curve. This interval represents the range of various indicators that the live broadcast should achieve under ideal conditions. For example, stable device operation, network latency below a certain threshold, and active audience interaction together constitute the criteria for the expected live - broadcast effect interval.

[0040] The device operation status curve reflects the change of various device performance indicators over time within the target repair time, such as CPU usage rate, memory occupancy rate, hard - disk read - write speed, device temperature, etc. Under normal circumstances, the values of various device performance indicators should be within a reasonable range and will not have a negative impact on the live broadcast. For example, the CPU usage rate should generally be maintained at a relatively low level (such as below 80%, and the specific value can be adjusted according to the device performance and live - broadcast requirements), the memory occupancy is stable and there is a certain margin, the hard - disk read - write speed can meet the needs of live - broadcast data storage and reading, and the device temperature is within the normal working temperature range (such as server devices are generally at 20 - 25 degrees Celsius). By analyzing the curve trend and the indicator range, determine the fluctuation range of the performance indicators when the device is operating normally.

[0041] The network transmission quality curve shows the changes in indicators such as network bandwidth, latency, and packet - loss rate within the target repair time. High - quality network transmission should have stable and sufficient bandwidth (for example, high - definition live broadcast requires at least a certain Mbps of upstream bandwidth, and the specific value depends on the live - broadcast picture quality), low network latency (generally controlled within dozens of milliseconds), and a packet - loss rate approaching zero. It is necessary to find the value range of these indicators when the network state is good. For example, when the live - broadcast traffic is normal and there is no network fault, the network latency is basically stable at a certain low value, and the bandwidth remains at a level that meets the live - broadcast requirements. Based on these data, determine the ideal interval of network transmission quality.

[0042] The audience interaction activity curve reflects the changes in interaction data such as the number of audience messages, likes, and gift-giving times per unit time. During the live broadcast planning stage, a general interaction activity standard is usually set according to the live broadcast type, content, and expected goals. For example, for popular entertainment live broadcasts, a certain number of messages per minute and more likes and gift-giving are expected; for knowledge popularization live broadcasts, more attention may be paid to the audience's questions and in-depth discussions. When analyzing this curve, refer to the data of previous successful live broadcasts of the same type and the planning expectations of this live broadcast to determine a reasonable range of audience interaction activity, such as how many messages per minute should be reached, and the expected average values of the number of likes and gift-giving within a certain time period, etc.

[0043] Different types of live broadcasts have different dependencies on device operation, network transmission, and audience interaction. Therefore, weights need to be assigned to the indicators corresponding to each curve. For example, for game live broadcasts mainly featuring high-definition video display and real-time interaction, the device operation status (ensuring smooth video) and audience interaction activity (measuring audience participation) may have relatively high weights, and the network transmission quality is also important but the weight can be slightly lower; for e-commerce live broadcasts, the weights of the three may be relatively more balanced. The determination of weights needs to be combined with the specific business needs and characteristics of the live broadcast, and generally, it is determined by performing a correlation analysis on historical live broadcast data.

[0044] Integrate the ranges of device operation performance indicators, network transmission quality indicators, and audience interaction activity indicators obtained from the above analysis, and perform integration according to the determined weights. For each time point or time period, comprehensively consider the indicator ranges of the three aspects to determine a comprehensive indicator interval. For example, at a certain moment, the device operation indicator is within the normal range, the network transmission indicator is also normal, and the audience interaction activity reaches the expectation. Combine the indicator ranges of these three aspects to form the expected live broadcast effect indicator at this moment. Integrate the indicators at each moment during the entire target repair time to obtain the expected live broadcast effect interval. This interval covers multiple dimensions such as devices, networks, and audience interaction, representing the range of comprehensive effect indicators that the live broadcast room should reach under ideal circumstances, providing a reference standard for evaluating the actual live broadcast effect.

[0045] Determine the actual live broadcast effect interval based on the content display effect curve combined with the actual situation presented in the live broadcast. This interval reflects the actual ranges of various display indicators in the current state of the live broadcast. For example, the fluctuation ranges of indicators such as screen resolution and frame rate during the actual live broadcast process are the actual live broadcast effect intervals.

[0046] Compare the expected live broadcast effect interval and the actual live broadcast effect interval to find the difference range between the two to obtain the operation state deviation interval. This deviation interval reflects the degree of deviation between the actual effect and the expected effect of the live broadcast. By analyzing the specific indicator differences within the deviation interval, it is possible to deeply understand the problems existing in the live broadcast, providing a clear direction for subsequent fault repair and optimization work.

[0047] Judge the type of operation anomaly of the live broadcast room according to the operation status deviation interval, and obtain the performance deficit of the live broadcast room according to the type of operation anomaly, specifically: Set at least one standard deviation interval, each standard deviation interval corresponds to a type of operation anomaly, and each type of operation anomaly corresponds to a performance deficit; Compare the operation status deviation interval of the live broadcast room with the standard deviation interval, and mark the standard deviation interval where the operation status deviation interval of the live broadcast room is located as the target deviation interval; Mark the type of operation anomaly corresponding to the target deviation interval as the type of operation anomaly of the live broadcast room, and mark the performance deficit corresponding to the type of operation anomaly as the performance deficit of the live broadcast room.

[0048] This application pre-sets at least one standard deviation interval. Each standard deviation interval corresponds to a specific type of operation anomaly, and each type of operation anomaly is associated with a corresponding performance deficit. For example, set the standard deviation interval A as the network latency is higher than a certain threshold and the bandwidth is lower than a specific value. The corresponding type of operation anomaly is "poor network transmission". The performance deficits under this anomaly type include the quantization indexes of situations such as live broadcast picture freezing and interactive message delay; the standard deviation interval B is that the device CPU usage rate exceeds a certain proportion and the memory occupancy is too high, corresponding to the "device performance bottleneck" anomaly type, and the performance deficit is manifested as the quantization description of picture frame drops and slow program responses. These standard deviation intervals and corresponding relationships are obtained based on a large amount of live broadcast experience data, industry standards, and the analysis and summary of common faults.

[0049] After obtaining the operation status deviation interval of the live broadcast room, compare it with the pre-set standard deviation intervals one by one. During the comparison process, check whether the indicators of the operation status deviation interval (such as device performance indicators, network indicators, interactive indicators, etc.) meet the range requirements of a certain standard deviation interval. For example, if the network latency of the live broadcast room is continuously higher than the standard value, the bandwidth is also lower than the normal level, and other indicators match, then the operation status deviation interval conforms to the above-mentioned standard deviation interval A, and the conforming standard deviation interval is marked as the target deviation interval.

[0050] After determining the target deviation interval, mark the type of operation anomaly corresponding to it as the current type of operation anomaly of the live broadcast room. For example, if the target deviation interval is A, then the type of operation anomaly is "poor network transmission". At the same time, mark the performance deficit corresponding to this type of operation anomaly as the performance deficit of the live broadcast room, that is, clarify the specific quantified performance index deviations such as the degree of picture freezing and the delay time of interactive messages caused by poor network transmission in the current live broadcast room.

[0051] Based on the expected live broadcast effect and the actual live broadcast effect of this live broadcast room in the deviation parameters, obtain the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time, specifically as follows: If the expected live broadcast effect corresponding to this live broadcast room in the deviation parameters is greater than or equal to the actual live broadcast effect, calculate the difference between the expected live broadcast effect and the actual live broadcast effect of this live broadcast room to obtain the amount of optimization and adjustment required to improve the live broadcast effect within the target repair time; And obtain the average workload of optimization and adjustment per unit time based on the amount of optimization and adjustment and the target emergency repair duration, and perform optimization and adjustment on this live broadcast room based on the average workload and the amount of optimization and adjustment; If the expected live broadcast effect of this live broadcast room in the deviation parameters is less than the actual live broadcast effect, then calculate the difference between the expected live broadcast effect and the actual live broadcast effect of this live broadcast room to obtain the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time.

[0052] This application compares the expected live broadcast effect and the actual live broadcast effect of the live broadcast room in the deviation parameters. The expected live broadcast effect is the ideal live broadcast state set in the live broadcast planning stage, covering multiple indicators such as picture quality, interaction frequency, and content display effect; the actual live broadcast effect is the current actual performance of the live broadcast obtained by real-time monitoring of device operation data, network status data, interaction data, and display data, etc., and judge the size relationship between the two to determine whether the live broadcast effect fails to meet expectations or exceeds expectations.

[0053] If the expected live broadcast effect is greater than or equal to the actual live broadcast effect, it means that there is room for improvement in the live broadcast effect. Calculate the difference between the expected live broadcast effect and the actual live broadcast effect to obtain the amount of optimization and adjustment required to improve the live broadcast effect within the target repair time. For example, if the expected live broadcast picture resolution is 1080P and the actual is only 720P, the calculated resolution improvement amount is part of the amount of optimization and adjustment.

[0054] Based on the obtained amount of optimization and adjustment and the set target emergency repair duration, calculate the average workload of each optimization and adjustment. Assume that the target emergency repair duration is 30 minutes, and the amount of optimization and adjustment is the sum of multiple tasks such as improving the picture resolution and improving network latency. Distribute these task amounts evenly over 30 minutes to obtain the workload that needs to be completed per unit time, that is, the average workload.

[0055] According to the average workload and the amount of optimization and adjustment, arrange maintenance personnel and resources to optimize and adjust the live broadcast room, and gradually complete various optimization tasks within the target repair time according to the calculated work rhythm, such as upgrading the graphics processing device to improve the resolution, optimizing the network configuration to reduce latency, etc., so as to improve the live broadcast effect.

[0056] When the expected live broadcast effect is less than the actual live broadcast effect, it indicates that the current live broadcast performance is good. To maintain this high-quality live broadcast state, the difference between the expected live broadcast effect and the actual live broadcast effect is calculated to obtain the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time. For example, in an actual live broadcast, the amount of audience interaction far exceeds the expectation. To ensure the smoothness of the interaction, it may be necessary to increase server resources, optimize the interaction program, etc. After calculating the specific amount of resources required, the corresponding resources are invested to maintain a good live broadcast effect.

[0057] Analyze the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time to obtain the first operation and maintenance coefficient of this live broadcast room, specifically: Mark the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time as the guarantee requirement; Obtain the unit resource guarantee cost corresponding to the unit live broadcast task in this live broadcast room; Based on the guarantee requirement of this live broadcast room and the unit resource guarantee cost, obtain the total guarantee cost for resource guarantee of all live broadcast tasks; Obtain the first operation and maintenance coefficient of this live broadcast room based on the total guarantee cost.

[0058] This application marks the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time as the guarantee requirement. For example, if there is a problem of video freezing in the live broadcast room due to insufficient graphics processing ability and insufficient network bandwidth, then the amount of graphics processing device resources and network bandwidth resources required to solve these two problems constitutes the guarantee requirement of this live broadcast room.

[0059] Pre-calculate the costs of various resources in the live broadcast room in advance to determine the unit resource guarantee cost corresponding to the unit live broadcast task. The unit live broadcast task can be a basic function module or a specific business operation in the live broadcast room. For example, the cost corresponding to the device resources, network resources, etc. consumed for every 10 minutes of product display live broadcast task is calculated, and then the unit resource guarantee cost of the unit live broadcast task is obtained. This cost data covers multiple aspects such as equipment purchase and maintenance costs, network lease costs, and labor costs, and is an important basis for subsequent calculations.

[0060] Perform a multiplication operation based on the determined guarantee requirement and the obtained unit resource guarantee cost to obtain the total guarantee cost for resource guarantee of all live broadcast tasks. Through this calculation method, the guarantee requirement is converted into a specific cost value to intuitively reflect the resource cost investment required to ensure the normal operation of the live broadcast room.

[0061] Derive the first operation and maintenance coefficient of the live broadcast room based on the calculated total guarantee cost. This coefficient can be a function value of the total guarantee cost. For example, the first operation and maintenance coefficient = total guarantee cost × a certain proportional factor (this proportional factor is determined based on a large amount of historical data). The first operation and maintenance coefficient comprehensively reflects the cost investment of the live broadcast room in terms of resource guarantee, providing a quantitative basis for subsequent resource allocation and operation and maintenance decisions.

[0062] Classify various live broadcast tasks of the live broadcast room into guarantee tasks and non-urgent tasks. Specifically: Obtain the task types of each live broadcast task in the live broadcast room; Classify each live broadcast task of the live broadcast room according to the pre-set task classification criteria to obtain the task categories corresponding to each live broadcast task; among them, the task categories include core task categories, important task categories, and ordinary task categories; Set delayable tasks based on core task categories, important task categories, and ordinary task categories; Divide the live broadcast tasks into guarantee tasks and non-urgent tasks according to the delayable tasks.

[0063] This application obtains the task types of each live broadcast task in the live broadcast room through the recording of live broadcast-related business processes. These task types cover various specific activities during the live broadcast, such as the host's product introduction, interaction games with the audience, and background data statistics. Each task has its specific function and role.

[0064] Classify the obtained live broadcast tasks according to the pre-set task classification criteria. The task classification criteria are usually formulated based on factors such as the importance of the task to the live broadcast effect, time urgency, and impact on the audience experience.

[0065] The core task category is crucial for achieving the core goal of the live broadcast, directly affecting the main experience of the audience and the key value output of the live broadcast. For example, in e-commerce live broadcasts, the host's detailed explanation of core products and the display of key product features; in education live broadcasts, the teacher's teaching of key knowledge, etc. Once problems occur in these tasks, it will seriously affect the realization of the live broadcast purpose and audience satisfaction.

[0066] Although the important task category does not directly determine the success or failure of the live broadcast like the core task, it has a greater impact on the overall effect of the live broadcast, audience retention, and participation. For example, the introduction of time-limited preferential activities in e-commerce live broadcasts, the Q&A session in education live broadcasts, etc., can enhance the audience's sense of participation and interest.

[0067] The ordinary task category is relatively less important and has a smaller impact on the core effect of the live broadcast, such as a simple summary after the live broadcast and the display of some secondary information.

[0068] Determine the deferrable tasks based on the classified core task categories, important task categories, and ordinary task categories. Generally speaking, most tasks in the ordinary task category can be set as deferrable tasks because their deferred execution will not seriously affect the key objectives and core experience of the live broadcast. For some tasks in the important task category, if a certain degree of delay is allowed in the schedule, they can also be included in the scope of deferrable tasks. For example, during a live broadcast, if there is a technical problem in the current live broadcast, a planned interactive mini-game can be temporarily postponed.

[0069] Tasks other than the deferrable tasks, that is, the non-deferrable tasks in the core task category and some important task categories, are classified as guarantee tasks. These tasks need to be given priority in resource and operation and maintenance support during the live broadcast to ensure their smooth execution, while the deferrable tasks are classified as non-urgent tasks, which have certain flexibility in resource allocation and operation and maintenance processing and can be arranged for execution according to the actual situation of the live broadcast and the resource status.

[0070] And obtain the second operation and maintenance coefficient of this live broadcast room based on the analysis of the guarantee tasks and non-urgent tasks. Specifically: Obtain the total guarantee cost for resource guarantee of the guarantee tasks based on the guarantee tasks and the unit resource guarantee cost corresponding to each unit live broadcast task in this live broadcast room; Obtain the unit task delay compensation cost corresponding to each unit live broadcast task in this live broadcast room; Obtain the total compensation cost for non-urgent tasks based on the non-urgent tasks and the unit task delay compensation cost corresponding to each unit live broadcast task in this live broadcast room; Obtain the second operation and maintenance coefficient of this live broadcast room based on the total guarantee cost and the total compensation cost.

[0071] After determining the guarantee tasks in this application, combined with the unit resource guarantee cost corresponding to each unit live broadcast task in the live broadcast room (that is, the cost of resources required to guarantee each unit live broadcast task, covering resource costs such as equipment, network, and manpower), the total guarantee cost for resource guarantee of the guarantee tasks is obtained through multiplication.

[0072] Determine the unit task delay compensation cost corresponding to each unit live broadcast task in this live broadcast room through historical data statistics. This cost reflects the compensation cost for possible losses caused by the deferred execution of unit live broadcast tasks, such as the decline in audience satisfaction and the loss of business opportunities.

[0073] For example, during a live broadcast, although some tasks (non-urgent tasks) can be delayed, the delay may bring a series of negative impacts. For example, delaying the lottery session during the live broadcast may lead to a decrease in the patience of the audience during the waiting process, thereby reducing the satisfaction and enthusiasm of the audience for the live broadcast; another example is delaying the processing of the task of replying to the audience's messages, which may make the audience feel ignored and affect the audience's favorability towards the anchor and the live broadcast platform. In addition, if some non-urgent tasks are delayed, the best opportunity may be missed, resulting in business losses. For example, delaying the release of live broadcast-related promotional content may miss the traffic peak, reduce the exposure volume, and the potential audience drainage. These declines in the audience experience and business opportunity losses caused by task delays all require a certain cost to make up for or reduce the impact, which generates the unit task delay compensation cost.

[0074] The total compensation cost for non-urgent tasks is obtained through multiplication based on the determined number of non-urgent tasks and the unit task delay compensation cost.

[0075] The second operation and maintenance coefficient of the live broadcast room is calculated by weighted summation of the comprehensive total guarantee cost and the total compensation cost. This coefficient reflects the comprehensive situation of the live broadcast room in terms of the resource investment for guarantee tasks and the risk of non-urgent task delays, providing an important basis for the operation and maintenance decision-making of live broadcast resources.

[0076] Resource operation and maintenance processing is performed on the live broadcast room according to the first operation and maintenance coefficient and the second operation and maintenance coefficient. Specifically: If the first operation and maintenance coefficient is greater than or equal to the second operation and maintenance coefficient, the corresponding first resource guarantee value of the live broadcast room is obtained based on the guarantee requirements and the target emergency repair duration of the live broadcast room, and resource operation and maintenance processing is performed on all live broadcast tasks of the live broadcast room according to the first resource guarantee value; If the first operation and maintenance coefficient is less than the second operation and maintenance coefficient, the corresponding second resource guarantee value of the live broadcast room is obtained based on the guarantee tasks and the target emergency repair duration of the live broadcast room, and resource operation and maintenance processing is performed on the guarantee tasks of the live broadcast room according to the second resource guarantee value.

[0077] This application first compares the magnitudes of the first operation and maintenance coefficient and the second operation and maintenance coefficient. The first operation and maintenance coefficient reflects the degree of resource guarantee cost investment required to maintain the actual live broadcast effect, which is related to the overall performance deficit and guarantee requirements of the live broadcast room; the second operation and maintenance coefficient reflects the comprehensive situation of the resource cost of guarantee tasks and the non-urgent task delay compensation cost, focusing on the cost consideration after task classification. By comparing these two coefficients, the key direction of resource allocation is judged.

[0078] When the first operation and maintenance coefficient is greater than or equal to the second operation and maintenance coefficient, according to the guarantee requirements of the live broadcast room (that is, the amount of guarantee resources required to maintain the actual live broadcast effect) and the target emergency repair duration, the first resource guarantee value is obtained through a calculation method (such as dividing the guarantee requirements by the target emergency repair duration to obtain the resource guarantee requirements per unit time). This value represents the amount of resource guarantee that should be provided to meet the overall live broadcast requirements within the target emergency repair duration.

[0079] Resource operation and maintenance processing is performed on all live broadcast tasks in the live broadcast room according to the first resource guarantee value. This means that in terms of resource allocation, regardless of task categories, the resource requirements of each task are comprehensively guaranteed to ensure the stable operation of the overall live broadcast. For example, when there are multiple problems such as equipment performance and network transmission in the live broadcast and the first operation and maintenance coefficient is relatively high, equipment upgrade resources, network optimization resources, etc. are allocated to ensure the smooth progress of all live broadcast tasks.

[0080] When the first operation and maintenance coefficient is less than the second operation and maintenance coefficient, the second resource guarantee value is calculated based on the guarantee tasks and the target emergency repair duration of the live broadcast room. First, determine the amount of resources required for the guarantee tasks, and then combine the target emergency repair duration to obtain the second resource guarantee value. This value represents the amount of resources that should be provided to ensure the smooth progress of key tasks within the target emergency repair duration.

[0081] Resource operation and maintenance processing is performed on the guarantee tasks in the live broadcast room according to the second resource guarantee value. At this time, resources are focused on tasks that play a key role in the core effect of the live broadcast, and the resource requirements of non-urgent tasks are not considered or processed later. For example, in a live broadcast, if the interaction session is a guarantee task, when the first operation and maintenance coefficient is less than the second operation and maintenance coefficient, the resources for the interaction session are preferentially guaranteed, such as ensuring the processing ability of the server for interaction messages and the stability of the network for interaction data transmission. For non-urgent tasks such as the statistical task after the live broadcast, they can be delayed when resources are limited. Embodiment 2

[0082] Based on Embodiment 1, the following technical features are added: An adaptive operation and maintenance monitoring system for multiple live broadcast rooms, including: An acquisition module that acquires the operation data of multiple live broadcast rooms, and each live broadcast room is connected to the live broadcast platform server through a network link; among them, the operation data includes equipment operation data, network status data, interaction data, and display data; A judgment module that analyzes the equipment operation data, network status data, interaction data, and display data when the situation in the live broadcast room is abnormal to obtain the operation status deviation range of the live broadcast room, judges the type of operation abnormality of the live broadcast room based on the operation status deviation range, and obtains the performance deficit of the live broadcast room according to the type of operation abnormality; among them, the performance deficit is the deviation parameter between the corresponding expected live broadcast effect and the actual live broadcast effect of the live broadcast room; The first analysis module obtains the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time based on the expected live broadcast effect and the actual live broadcast effect of the live broadcast room in the deviation parameter, and analyzes the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time to obtain the first operation and maintenance coefficient of the live broadcast room; The second analysis module classifies various live broadcast tasks of the live broadcast room into guarantee tasks and non-urgent tasks, and analyzes the second operation and maintenance coefficient of the live broadcast room based on the guarantee tasks and non-urgent tasks; The processing module performs resource operation and maintenance processing on the live broadcast room according to the first operation and maintenance coefficient and the second operation and maintenance coefficient.

[0083] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative work.

[0084] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0085] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An adaptive operation and maintenance monitoring method for multiple live rooms, characterized in that The method includes the following steps: Obtain the operation data of multiple live rooms; wherein, the operation data includes device operation data, network status data, interaction data, and display data; When an abnormality occurs in the live room situation, analyze the device operation data, network status data, interaction data, and display data to obtain the operation status deviation range of the live room, determine the operation abnormality type of the live room according to the operation status deviation range, and obtain the performance deficit of the live room according to the operation abnormality type; wherein, the performance deficit is the deviation parameter between the corresponding expected live broadcast effect and the actual live broadcast effect of the live room; Obtain the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time according to the expected live broadcast effect and the actual live broadcast effect in the deviation parameter, and analyze the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time to obtain the first operation and maintenance coefficient of the live room; Classify various live broadcast tasks of the live room into guarantee tasks and non-urgent tasks, and analyze to obtain the second operation and maintenance coefficient of the live room according to the guarantee tasks and non-urgent tasks; Perform resource operation and maintenance processing on the live room according to the first operation and maintenance coefficient and the second operation and maintenance coefficient.

2. The adaptive operation and maintenance monitoring method for multiple live broadcast rooms according to claim 1, characterized in that It further includes: Set the target emergency repair duration of the live room, and obtain the time when the operation abnormality occurs in the live room; Obtain the target repair time of the live room according to the time when the operation abnormality occurs in the live room and the target emergency repair duration.

3. The adaptive operation and maintenance monitoring method for multiple live rooms according to claim 2, wherein, When an abnormality occurs in the live room situation, analyze the device operation data, network status data, interaction data, and display data to obtain the operation status deviation range of the live room, specifically: Obtain the device operation status curve within the target repair time according to the device operation data; Obtain the network transmission quality curve within the target repair time according to the network status data; Obtain the audience interaction activity curve within the target repair time according to the interaction data; Obtain the content display effect curve within the target repair time according to the display data; Obtain the expected live broadcast effect range of the live room according to the device operation status curve, network transmission quality curve, and audience interaction activity curve, and obtain the actual live broadcast effect range of the live room according to the content display effect curve; And obtain the operation status deviation range of the live room according to the expected live broadcast effect range and the actual live broadcast effect range.

4. A method for adaptively operating and maintaining monitoring of multiple live broadcast rooms according to claim 3, characterized in that Judge the operation abnormality type of the live room according to the operation status deviation range, and obtain the performance deficit of the live room according to the operation abnormality type, specifically: Set at least one standard deviation range, each standard deviation range corresponds to an operation abnormality type, and each operation abnormality type corresponds to a performance deficit; Compare the operation status deviation range of the live room with the standard deviation range, and mark the standard deviation range where the operation status deviation range of the live room is located as the target deviation range; Mark the operation abnormality type corresponding to the target deviation range as the operation abnormality type of the live room, and mark the performance deficit corresponding to the operation abnormality type as the performance deficit of the live room.

5. The adaptive operation and maintenance monitoring method for multiple live rooms according to claim 4, characterized in that, Obtain the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time according to the expected live broadcast effect and the actual live broadcast effect in the deviation parameter, specifically: If the expected live broadcast effect of the live broadcast room in the deviation parameter is less than the actual live broadcast effect, the difference between the expected live broadcast effect and the actual live broadcast effect corresponding to the live broadcast room is calculated to obtain the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time.

6. The adaptive operation and maintenance monitoring method for multiple live rooms according to claim 5, wherein, Analyze the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time to obtain the first operation and maintenance coefficient of the live broadcast room, specifically: Mark the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time as the guarantee requirement; Obtain the unit resource guarantee cost corresponding to the unit live broadcast task in the live broadcast room; Obtain the total guarantee cost for resource guarantee of all live broadcast tasks based on the guarantee requirement and the unit resource guarantee cost of the live broadcast room; Obtain the first operation and maintenance coefficient of the live broadcast room based on the total guarantee cost.

7. A method for adaptively operating and maintaining monitoring of multiple live broadcast rooms according to claim 6, characterized in that, Classify various live broadcast tasks of the live broadcast room into guarantee tasks and non-urgent tasks, specifically: Obtain the task types of each live broadcast task in the live broadcast room; Classify each live broadcast task of the live broadcast room according to the pre-set task classification standard to obtain the task category corresponding to each live broadcast task; among them, the task category includes the core task category, the important task category and the ordinary task category; Set delayable tasks based on the core task category, the important task category and the ordinary task category; Divide the live broadcast tasks into guarantee tasks and non-urgent tasks according to the delayable tasks.

8. A method for adaptively monitoring the operation and maintenance of multiple live broadcast rooms according to claim 7, characterized in that, And analyze the second operation and maintenance coefficient of the live broadcast room based on the guarantee tasks and non-urgent tasks, specifically: Obtain the total guarantee cost for resource guarantee of the guarantee tasks based on the guarantee tasks and the unit resource guarantee cost corresponding to the unit live broadcast task in the live broadcast room; Obtain the unit task delay compensation cost corresponding to the unit live broadcast task in the live broadcast room; Obtain the total compensation cost for non-urgent tasks based on the non-urgent tasks and the unit task delay compensation cost corresponding to the unit live broadcast task in the live broadcast room; Obtain the second operation and maintenance coefficient of the live broadcast room based on the total guarantee cost and the total compensation cost.

9. The adaptive operation and maintenance monitoring method for multiple live broadcast rooms according to claim 8, characterized in that Perform resource operation and maintenance processing on the live broadcast room according to the first operation and maintenance coefficient and the second operation and maintenance coefficient, specifically: If the first operation and maintenance coefficient is greater than or equal to the second operation and maintenance coefficient, obtain the first resource guarantee value corresponding to the live broadcast room based on the guarantee requirement and the target emergency repair duration of the live broadcast room, and perform resource operation and maintenance processing on all live broadcast tasks of the live broadcast room according to the first resource guarantee value; If the first operation and maintenance coefficient is less than the second operation and maintenance coefficient, obtain the second resource guarantee value corresponding to the live broadcast room based on the guarantee tasks and the target emergency repair duration of the live broadcast room, and perform resource operation and maintenance processing on the guarantee tasks of the live broadcast room according to the second resource guarantee value.

10. A multi-live broadcast room adaptive operation and maintenance monitoring system is applied to the multi-live broadcast room adaptive operation and maintenance monitoring method described in any one of claims 1-9, and is characterized in that, Including: An acquisition module, which acquires the operation data of multiple live broadcast rooms, and each live broadcast room is connected to the live broadcast platform server through a network link; among them, the operation data includes device operation data, network status data, interaction data and display data; A judgment module analyzes device operation data, network status data, interaction data, and display data when the live broadcast room status is abnormal to obtain the operation status deviation range of the live broadcast room, determines the type of operation abnormality of the live broadcast room based on the operation status deviation range, and obtains the performance deficit of the live broadcast room according to the type of operation abnormality; wherein, the performance deficit is the deviation parameter between the corresponding expected live broadcast effect and the actual live broadcast effect of the live broadcast room. A first analysis module obtains the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time based on the expected live broadcast effect and the actual live broadcast effect of the live broadcast room in the deviation parameter, and analyzes the amount of guarantee resources required to maintain the actual live broadcast effect within the target repair time to obtain the first operation and maintenance coefficient of the live broadcast room. A second analysis module classifies various live broadcast tasks of the live broadcast room into guarantee tasks and non-urgent tasks, and analyzes the second operation and maintenance coefficient of the live broadcast room based on the guarantee tasks and non-urgent tasks. A processing module performs resource operation and maintenance processing on the live broadcast room according to the first operation and maintenance coefficient and the second operation and maintenance coefficient.

Images