Operation and maintenance method, device, equipment, medium and program product

Abstract

The invention provides an operation and maintenance method and device, equipment, a medium and a program product, relates to the technical field of communication, and is used for improving the resource utilization rate and operation and maintenance efficiency of operation and maintenance, and the specific technical scheme is as follows: obtaining operation and maintenance data and first state information of terminal equipment, the first state information comprising a terminal computing power load and a terminal network bandwidth; under the condition that it is detected that second state information of the cloud equipment meets a preset condition, an operation and maintenance allocation decision value is obtained through calculation based on the terminal computing power load, the terminal network bandwidth and feature information corresponding to the operation and maintenance data, and the second state information comprises the cloud computing power load and the cloud storage remaining capacity; the feature information at least comprises task complexity, data privacy and real-time requirements; based on the privacy level and the operation and maintenance allocation decision value corresponding to the operation and maintenance data, determining an operation and maintenance device for processing the operation and maintenance data, the operation and maintenance device being a terminal device or a cloud device; the method is applied to an operation and maintenance early warning scene.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to an operation and maintenance method, apparatus, equipment, medium and program product. Background Technology

[0002] Currently, with the deepening of enterprise digital transformation, the scale and complexity of information technology infrastructure are increasing dramatically, posing unprecedented challenges to the real-time, intelligent, and secure aspects of operation and maintenance management. Against this backdrop, building an efficient and reliable operation and maintenance system has become an important development trend in this field.

[0003] To address the aforementioned needs, existing technologies propose a cloud-dominated semi-collaborative operation and maintenance (O&M) solution. This solution typically deploys lightweight data acquisition tools on the terminal device side to perform preliminary preprocessing of the raw O&M data according to pre-designed processing logic. Then, the processed data is uploaded to a cloud platform, where cloud devices centrally store, analyze, and make decisions, ultimately issuing instructions to the terminal devices for execution.

[0004] However, since the terminal devices send pre-processed maintenance data to the cloud devices for further processing, the maintenance analysis is essentially still performed through the cloud devices. Therefore, the computing resources of the terminal devices and the cloud devices cannot be coordinated and fully utilized. Consequently, the overall system resource utilization and maintenance efficiency are low. Summary of the Invention

[0005] This application provides an operation and maintenance method, apparatus, equipment, medium, and program product to improve the resource utilization and efficiency of operation and maintenance.

[0006] In a first aspect, embodiments of this application provide an operation and maintenance method, which includes: acquiring operation and maintenance data and first state information of a terminal device, the first state information including: terminal computing power load and terminal network bandwidth; when detecting that the second state information of a cloud device meets preset conditions, calculating an operation and maintenance allocation decision value based on the terminal computing power load, terminal network bandwidth and characteristic information corresponding to the operation and maintenance data, the second state information including: cloud computing power load and remaining cloud storage capacity, the characteristic information including at least: task complexity, data privacy, and real-time requirements; and determining the operation and maintenance device that processes the operation and maintenance data based on the privacy level corresponding to the operation and maintenance data and the operation and maintenance allocation decision value, the operation and maintenance device being either a terminal device or a cloud device.

[0007] The technical solution provided in this application brings at least the following beneficial effects: When the second state information of the cloud device is detected to meet the preset conditions, the operation and maintenance allocation decision value is calculated by using the first state information of the terminal device and the feature information corresponding to the operation and maintenance data. Then, based on the privacy level corresponding to the operation and maintenance data and the operation and maintenance allocation decision value, the operation and maintenance equipment for processing the operation and maintenance data is determined. That is, when the computing power resources of the cloud device are detected to meet the preset conditions, the terminal device can make autonomous decisions and the cloud device can be optimized globally through the operation and maintenance allocation decision algorithm and privacy level. This enables the computing power resources of the cloud device and the terminal device to be coordinated and fully utilized, thereby improving the resource utilization rate and operation and maintenance efficiency.

[0008] One possible implementation involves determining the maintenance equipment that processes the maintenance data based on the privacy level and maintenance allocation decision value corresponding to the maintenance data. This includes: identifying the terminal device as the maintenance equipment when the privacy level is the first privacy level; or identifying the terminal device as the maintenance equipment when the privacy level is the second or third privacy level and the maintenance allocation decision value is greater than or equal to a preset decision threshold; or identifying the cloud device as the maintenance equipment when the privacy level is the second or third privacy level and the maintenance allocation decision value is less than a preset decision threshold. The first privacy level is greater than the second privacy level, and the second privacy level is greater than the third privacy level.

[0009] Another possible implementation method, the above method further includes: calculating a privacy value based on the amount of privacy data contained in the operation and maintenance data; determining the privacy level corresponding to the operation and maintenance data as the first privacy level when the privacy value is greater than the maximum value of the first privacy range; determining the privacy level corresponding to the operation and maintenance data as the second privacy level when the privacy value is within the first privacy range; and determining the privacy level corresponding to the operation and maintenance data as the third privacy level when the privacy value is less than the minimum value of the first privacy range.

[0010] Another possible implementation method, after determining the operation and maintenance equipment for processing operation and maintenance data based on the privacy level and operation and maintenance allocation decision value corresponding to the operation and maintenance data, the method further includes: when the operation and maintenance equipment is a terminal device, inputting the operation and maintenance data into the operation and maintenance inference model in the terminal device, performing fault identification on the operation and maintenance data, and obtaining the operation and maintenance results; wherein, the operation and maintenance inference model is trained based on cross-entropy loss and distillation loss.

[0011] Another possible implementation, after determining the maintenance equipment for processing maintenance data based on the privacy level and maintenance allocation decision value corresponding to the maintenance data, further includes: when it is determined that the maintenance equipment is a cloud device and an abnormal state is detected in the cloud device, controlling the terminal device to perform fault identification based on the maintenance data and the first local emergency policy stored in the terminal device to obtain the maintenance result, wherein the first local emergency policy is sent to the terminal device when the cloud device is in a normal state; when it is detected that the cloud device has returned to a normal state, the maintenance result is sent to the cloud device, and the second local emergency policy sent by the cloud device is forwarded to the terminal device.

[0012] Another possible implementation method, after determining the operation and maintenance equipment for processing operation and maintenance data based on the privacy level and operation and maintenance allocation decision value corresponding to the operation and maintenance data, the method further includes: when it is determined that the operation and maintenance equipment is a terminal device and the terminal device is detected to be in an abnormal state, controlling the cloud device to perform fault identification based on the operation and maintenance data to obtain the operation and maintenance result; when the terminal device is detected to have returned to normal state, sending the operation and maintenance result to the terminal device.

[0013] Another possible implementation involves calculating the operation and maintenance allocation decision value based on the characteristic information corresponding to the terminal computing power load, terminal network bandwidth, and operation and maintenance data. This includes: weighting and summing the terminal computing power load, terminal network bandwidth, task complexity, data privacy, and real-time requirements based on the weights corresponding to the terminal computing power load, terminal network bandwidth, task complexity, data privacy, and real-time requirements to obtain the operation and maintenance allocation decision value; wherein the sum of the weights corresponding to the terminal computing power load, terminal network bandwidth, task complexity, data privacy, and real-time requirements is 1.

[0014] Secondly, embodiments of this application provide an operation and maintenance device, including: an acquisition module, a processing module, and a determination module. The acquisition module is used to acquire operation and maintenance data and first state information of a terminal device, the first state information including: terminal computing power load and terminal network bandwidth. The processing module is used to calculate an operation and maintenance allocation decision value based on the terminal computing power load, terminal network bandwidth, and characteristic information corresponding to the operation and maintenance data, when it detects that the second state information of the cloud device meets preset conditions. The second state information includes: cloud computing power load and remaining cloud storage capacity, and the characteristic information includes at least task complexity, data privacy, and real-time requirements. The determination module is used to determine the operation and maintenance device that processes the operation and maintenance data based on the privacy level corresponding to the operation and maintenance data and the operation and maintenance allocation decision value, wherein the operation and maintenance device is a terminal device or a cloud device.

[0015] One possible implementation is that the aforementioned determining module is specifically used to determine the terminal device as an operation and maintenance device when the privacy level is the first privacy level; or, when the privacy level is the second or third privacy level and the operation and maintenance allocation decision value is greater than or equal to a preset decision threshold, the terminal device is determined as an operation and maintenance device; or, when the privacy level is the second or third privacy level and the operation and maintenance allocation decision value is less than a preset decision threshold, the cloud device is determined as an operation and maintenance device; wherein the first privacy level is greater than the second privacy level, and the second privacy level is greater than the third privacy level.

[0016] In another possible implementation, the aforementioned processing module is further configured to calculate a privacy value based on the amount of privacy data contained in the operation and maintenance data; if the privacy value is greater than the maximum value of the first privacy range, the privacy level corresponding to the operation and maintenance data is determined as the first privacy level; if the privacy value is within the first privacy range, the privacy level corresponding to the operation and maintenance data is determined as the second privacy level; and if the privacy value is less than the minimum value of the first privacy range, the privacy level corresponding to the operation and maintenance data is determined as the third privacy level.

[0017] In another possible implementation, the aforementioned processing module is further used to determine the operation and maintenance equipment for processing the operation and maintenance data based on the privacy level and operation and maintenance allocation decision value corresponding to the operation and maintenance data. If the operation and maintenance equipment is a terminal device, the operation and maintenance data is input into the operation and maintenance inference model in the terminal device to perform fault identification on the operation and maintenance data and obtain the operation and maintenance results. The operation and maintenance inference model is trained based on cross-entropy loss and distillation loss.

[0018] In another possible implementation, the aforementioned processing module is further configured to, after determining the maintenance device processing the maintenance data based on the privacy level and maintenance allocation decision value corresponding to the maintenance data, and upon determining that the maintenance device is a cloud device and detecting that the cloud device is in an abnormal state, control the terminal device to perform fault identification based on the maintenance data and the first local emergency policy stored on the terminal device to obtain the maintenance result. The first local emergency policy is to be sent to the terminal device when the cloud device is in a normal state. Upon detecting that the cloud device has returned to a normal state, the maintenance result is sent to the cloud device, and the second local emergency policy sent by the cloud device is forwarded to the terminal device.

[0019] In another possible implementation, the aforementioned processing module is further configured to determine the maintenance device for processing the maintenance data based on the privacy level and maintenance allocation decision value corresponding to the maintenance data; and, if the maintenance device is determined to be a terminal device and the terminal device is detected to be in an abnormal state, control the cloud device to perform fault identification based on the maintenance data and obtain the maintenance result; and if the terminal device is detected to have returned to normal, send the maintenance result to the terminal device.

[0020] Another possible implementation is that the aforementioned processing module is specifically used to perform a weighted summation of the terminal computing power load, terminal network bandwidth, task complexity, data privacy, and real-time requirements based on the weights corresponding to the terminal computing power load, terminal network bandwidth, task complexity, data privacy, and real-time requirements, to obtain the operation and maintenance allocation decision value; wherein the sum of the weights corresponding to the terminal computing power load, terminal network bandwidth, task complexity, data privacy, and real-time requirements is 1.

[0021] Thirdly, this application provides an electronic device comprising: a processor and a memory; the memory stores a program or instructions executable on the processor, wherein the program or instructions, when executed by the processor, implement the method of the first aspect described above.

[0022] Fourthly, this application provides a readable storage medium on which a program or instructions are stored, which, when executed by a computer, implement the method of the first aspect described above.

[0023] Fifthly, this application provides a computer program product stored in a storage medium, which, when executed by a computer, implements the method described in the first aspect.

[0024] In a sixth aspect, embodiments of this application provide a chip including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the method described in the first aspect.

[0025] The beneficial effects of the second to sixth aspects mentioned above are described in the corresponding description of the first aspect and will not be repeated here. Attached Figure Description

[0026] Figure 1 A schematic diagram of the network architecture for an operation and maintenance method application provided in this application embodiment;

[0027] Figure 2 A flowchart illustrating an operation and maintenance method provided in an embodiment of this application;

[0028] Figure 3 A flowchart illustrating another operation and maintenance method provided in this application embodiment;

[0029] Figure 4 A flowchart illustrating another operation and maintenance method provided in this application embodiment;

[0030] Figure 5 A flowchart illustrating another operation and maintenance method provided in this application embodiment;

[0031] Figure 6 A flowchart illustrating another operation and maintenance method provided in this application embodiment;

[0032] Figure 7 A flowchart illustrating another operation and maintenance method provided in this application embodiment;

[0033] Figure 8 A flowchart illustrating another operation and maintenance method provided in this application embodiment;

[0034] Figure 9 A flowchart illustrating another operation and maintenance method provided in this application embodiment;

[0035] Figure 10 This is a schematic diagram of the structure of an operation and maintenance system provided in an embodiment of this application;

[0036] Figure 11 This is a schematic diagram of the structure of an operation and maintenance device provided in an embodiment of this application;

[0037] Figure 12 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0038] The operation and maintenance methods, devices, equipment, media, and program products provided in this application will be described in detail below with reference to the accompanying drawings.

[0039] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0040] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0041] The terms "at least one," "at least one of," etc., used in the specification and claims of this application refer to any one, any two, or a combination of two or more of the included items. For example, at least one of a, b, and c can mean: "a," "b," "c," "a and b," "a and c," "b and c," and "a, b, and c," where a, b, and c can be single or multiple. Similarly, "at least two" refers to two or more items, and its meaning is similar to that of "at least one."

[0042] In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0043] The present application provides an operation and maintenance method, apparatus, equipment, medium, and program product that can be applied to system resource operation and maintenance scenarios.

[0044] Currently, with the acceleration of digital transformation, the complexity of enterprise information technology architecture is growing exponentially, and the scale of maintenance objects such as servers, network devices, and databases has exceeded one million, posing a severe challenge to traditional operation and maintenance models. Current mainstream operation and maintenance solutions can be mainly divided into three categories.

[0045] 1. Pure Cloud-Based Centralized O&M Solution: This solution uploads all O&M data to a cloud platform, relying on cloud computing power for data storage, analysis, and fault diagnosis. Its advantages include the ability to handle large-scale data and support cross-regional O&M. However, it suffers from three major drawbacks: First, high data transmission latency. O&M data (especially real-time metrics such as Central Processing Unit (CPU) utilization and network bandwidth) must be uploaded to the cloud via the network, with a single-trip latency of 100-500ms, resulting in delayed fault response and failing to meet the millisecond-level O&M requirements of industrial control, etc. Second, high bandwidth costs. The daily traffic costs generated by TB-level O&M data transmission account for 15%-20% of the enterprise's IT costs. Third, data privacy risks. Uploading core business logs (such as user payment records and system permission operations) to a third-party cloud platform poses a risk of leakage and tampering, failing to meet the requirements of "local data storage."

[0046] 2. Pure edge-side distributed operation and maintenance solution: Data processing and analysis functions are deployed on the enterprise's local server. Its advantages are that data does not need to be transmitted externally and the response speed is fast. However, it has a computing power bottleneck and cannot support deep learning analysis of large-scale data (such as fault root cause location and capacity prediction). In addition, data between different edge devices is isolated and lacks unified collaborative scheduling capabilities. When the computing power of a certain edge device is saturated, problems such as operation and maintenance task backlog, missed reports and false reports are likely to occur.

[0047] 3. Cloud-led Semi-Collaborative O&M Solution: Some vendors attempt to deploy lightweight data collection tools on the device side, uploading only pre-processed key data to the cloud for analysis. However, this solution is essentially a passive "cloud-based decision-making + device-based execution" model, with two major limitations: First, the decision-making logic is rigid, and the device side can only perform simple operations such as data filtering and format conversion, unable to dynamically adjust the data upload strategy based on local device load and network status. Second, there is a lack of collaboration capabilities; there is no two-way intelligent scheduling mechanism between the device side and the cloud. For example, when cloud computing power is saturated, some non-core analysis tasks (such as historical log archiving) cannot be offloaded to the device side for execution, resulting in an imbalance in resource utilization. In other words, there is a lack of dynamic collaboration mechanism, and the device side's computing power is not fully utilized (utilization rate is only 10%-15% when idle), while the cloud is in a high-load state for a long time (peak utilization rate exceeds 80%), resulting in an overall O&M resource utilization rate of less than 40%. Moreover, it cannot dynamically adjust the processing strategy according to the O&M scenario. For example, when the network is interrupted, the pure cloud solution completely fails, and the semi-collaborative solution can only store a small amount of local data, unable to complete fault diagnosis and self-healing operations.

[0048] To address the aforementioned technical issues, embodiments of this application provide an operation and maintenance method, apparatus, device, medium, and program product. By using the first state information of the terminal device and the second state information of the cloud device, an operation and maintenance allocation decision value is calculated. Based on the privacy level corresponding to the operation and maintenance data and the operation and maintenance allocation decision value, the operation and maintenance device processing the data is determined. In other words, through the operation and maintenance allocation decision algorithm and privacy level, the functions of autonomous decision-making by the terminal device and global optimization by the cloud device are realized, enabling the computing resources of the cloud device and the terminal device to be coordinated and fully utilized, thereby improving the resource utilization rate and efficiency of operation and maintenance.

[0049] The main inventive objectives of this application will be explained in detail below.

[0050] A two-way collaborative architecture of "edge-side autonomous decision-making + cloud-based global optimization" is constructed, and the utilization rate of edge-cloud resources is increased to over 70% through dynamic task scheduling algorithms.

[0051] The design incorporates a collaborative mechanism between a lightweight edge-side inference engine and a large-scale cloud-based analytics engine, reducing end-to-end latency for operations and maintenance tasks to less than 50ms, thus meeting real-time operations and maintenance requirements.

[0052] Establish a data tiered processing and privacy protection mechanism, where core sensitive data is processed only on the device side and non-sensitive data is uploaded to the cloud, in compliance with data security and compliance requirements.

[0053] It enables adaptive fault handling capabilities, allowing the edge device to independently complete basic operation and maintenance tasks (such as local device restart and temporary alarm blocking) in abnormal scenarios such as network interruption and cloud failure, ensuring the continuity of operation and maintenance.

[0054] The following description, in conjunction with the accompanying drawings, details the operation and maintenance methods, apparatus, equipment, media, and program products provided in the embodiments of this application.

[0055] Figure 1 This illustration shows a network architecture for an operation and maintenance method application provided in an embodiment of this application. For example... Figure 1 As shown, the network architecture includes: an operation and maintenance device 101, a terminal device 102, and a cloud device 103. The operation and maintenance device 101, the terminal device 102, and the cloud device 103 are interconnected.

[0056] In some embodiments, the maintenance device 101 may be a server, a computer, or a processor or processing unit within a server or computer. The server may be a single server or a server cluster consisting of multiple servers. It should be noted that the embodiments of this application do not limit the specific device form of the maintenance device 101. Figure 1 The example shown is a single server using the maintenance device 101.

[0057] In some embodiments, the terminal device may be a mobile phone, tablet computer, laptop computer, handheld computer, in-vehicle electronic device, mobile internet device (MID), augmented reality (AR) / virtual reality (VR) device, robot, wearable device, personal computer (PC), ultra-mobile personal computer (UMPC), netbook, or personal digital assistant (PDA), etc., and the embodiments of this application do not specifically limit it. Figure 1 The example shown is a mobile phone, with terminal device 102 as an example.

[0058] In some embodiments, cloud devices are cloud-based devices built using big data and cloud servers, including functions such as network web configuration, device management, device monitoring, fault early warning, device data analysis and applications. These capabilities can meet enterprises' needs for building agility and innovation, and improve enterprise management capabilities. A typical example of a cloud device is a cloud server. Figure 1 The example shown is a single server, cloud device 103.

[0059] In some embodiments, terminal device 102 sends first status information to maintenance device 101, and cloud device 102 sends second status information to maintenance device 101. Maintenance device 101 receives the first and second status information and obtains maintenance data. Then, when it detects that the second status information of the cloud device meets preset conditions, it calculates maintenance allocation decision value based on terminal computing load, terminal network bandwidth, and feature information corresponding to maintenance data. The second status information includes cloud computing load and cloud storage remaining capacity. The feature information includes at least task complexity information, data privacy information, and real-time requirement information. Based on the privacy level corresponding to maintenance data and maintenance allocation decision value, it determines the maintenance device that processes the maintenance data. The maintenance device is either a terminal device or a cloud device, and the corresponding maintenance device is notified.

[0060] It should be noted that the network architecture described in the embodiments of this application is for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and does not constitute a limitation on the technical solutions provided in the embodiments of this application. As network architectures evolve, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0061] See Figure 2 This is a flowchart illustrating an operation and maintenance method provided in an embodiment of this application. Figure 2 As shown, the operation and maintenance method provided in this application embodiment can be implemented by the above-mentioned operation and maintenance device, specifically including the following steps 201 to 203.

[0062] Step 201: The operation and maintenance device acquires operation and maintenance data and the first status information of the terminal equipment.

[0063] In some embodiments, the first state information mentioned above includes: terminal computing load and terminal network bandwidth.

[0064] In some embodiments, the aforementioned terminal network bandwidth includes at least one of the following: the network bandwidth currently available to the terminal device, the sum of the maximum bandwidth of the wireless communication network hardware and the Wireless Fidelity (WIFI) chip hardware on the terminal device, the uplink bandwidth of the electronic device, the downlink bandwidth of the electronic device, the bandwidth involved in load balancing, and the bandwidth of the network transmission module.

[0065] For example, the network bandwidth currently available to the aforementioned terminal device may include at least one of the following: available bandwidth based on mobile communication networks, and available bandwidth based on Wi-Fi.

[0066] For example, the available bandwidth based on the mobile communication network mentioned above may include at least one of the following: available bandwidth based on 5th-generation mobile communication technology (5G) or available bandwidth based on 4th-generation mobile communication technology (4G).

[0067] For example, the available bandwidth based on WIFI mentioned above may include at least one of the following: the difference between the bandwidth of the wireless Fidelity (WIFI) hardware device built into the terminal device and the bandwidth of the wireless local area network (WLAN) hardware device already in use; the difference between the capability of the WIFI router hardware device connected to the terminal device and the bandwidth of the WLAN already in use; and the difference between the bandwidth limited by the uplink port of the WIFI router connected to the terminal device and the bandwidth of the WLAN already in use.

[0068] In some embodiments, the operation and maintenance device can obtain the CPU utilization and memory usage of the terminal device, and then calculate the terminal computing load based on the weight corresponding to the CPU utilization and the weight corresponding to the memory usage of the terminal device.

[0069] In some embodiments, the weight corresponding to the CPU utilization can be user-defined or preset by the operation and maintenance device. The specific weight can be determined according to actual usage requirements, and this application embodiment does not impose any limitations.

[0070] In some embodiments, the weight corresponding to the memory occupancy rate can be user-defined or preset by the operation and maintenance device. The specific weight can be determined according to actual usage requirements, and this application embodiment does not impose any limitations.

[0071] For example, the operation and maintenance device can multiply the CPU utilization rate of the terminal device by the weight corresponding to the CPU utilization rate of the terminal device to obtain a first value, then multiply the memory occupancy rate of the terminal device by the weight corresponding to the memory occupancy rate of the terminal device to obtain a second value, and finally add the first value and the second value to obtain the aforementioned terminal computing load. Specifically, this can be achieved through the following formula (1).

[0072] (1)

[0073] in, The terminal computing load is represented by C1, the CPU utilization rate of the terminal device is represented by M1, the memory usage rate of the terminal device is represented by 0.6, the weight corresponding to the CPU utilization rate is represented by 0.4, and the weight corresponding to the memory usage rate is represented by 0.4.

[0074] It should be noted that the computing load of the aforementioned terminals is between (0, 1).

[0075] In some embodiments, the aforementioned operation and maintenance data may include at least one of the following: metric data, log data, and alarm data, etc. The specific data can be determined according to actual usage requirements, and this application embodiment does not impose any limitations.

[0076] For example, the above-mentioned metrics may include at least one of the following: CPU utilization, memory utilization, hard disk utilization, etc.

[0077] For example, the log data mentioned above may include at least one of the following: database error log, system error log, application error log, etc.

[0078] For example, the alarm data mentioned above may include at least one of the following: device offline alarm, device CPU resource alarm, device memory alarm, etc.

[0079] In some embodiments, the aforementioned maintenance data may be sent from a terminal device to the maintenance device; or sent from a cloud device to the maintenance device; or automatically acquired by the maintenance device at preset intervals. The specific details can be determined based on actual usage requirements, and this application embodiment does not impose any limitations.

[0080] In some embodiments, the aforementioned preset duration can be user-defined; or, it can be preset by the maintenance device. The specific duration can be determined according to actual usage requirements, and this application embodiment does not impose any limitations.

[0081] For example, the preset duration can be 1 hour, 2 hours or 3 hours.

[0082] In some embodiments, the above-mentioned maintenance device may be deployed in terminal devices or cloud devices, or it may be independent of terminal devices or cloud devices.

[0083] In some embodiments, the above-mentioned maintenance device can connect to the terminal device wirelessly or via a wired connection to obtain the first status information.

[0084] For example, the wireless connection described above can be a Wireless Fidelity (WIFI) connection.

[0085] It should be noted that the maintenance device can send a status acquisition request to the terminal device at preset intervals, and the terminal device can then send first status information to the maintenance device based on the status acquisition request. In other words, the first status information is based on the real-time status changes of the terminal device.

[0086] Step 202: When the maintenance device detects that the second state information of the cloud device meets the preset conditions, it calculates the maintenance allocation decision value based on the characteristic information corresponding to the terminal computing power load, terminal network bandwidth and maintenance data.

[0087] In some embodiments, the second state information includes: cloud computing load and remaining cloud storage capacity, and the feature information includes at least: task complexity, data privacy, and real-time requirements.

[0088] In some embodiments, the operation and maintenance device can obtain the CPU utilization and memory usage of the cloud device, and then calculate the cloud computing load based on the weight corresponding to the CPU utilization and the weight corresponding to the memory usage of the cloud device.

[0089] For example, the operation and maintenance device can multiply the CPU utilization rate of the cloud device with the weight corresponding to the CPU utilization rate of the cloud device to obtain a third value, then multiply the memory occupancy rate of the cloud device with the weight corresponding to the memory occupancy rate of the cloud device to obtain a fourth value, and finally add the third value and the fourth value to obtain the above-mentioned cloud computing load. Specifically, this can be achieved through the following formula (2).

[0090] (2)

[0091] in, C2 represents the cloud computing load, M2 represents the cloud device's CPU utilization, 0.6 represents the weight of CPU utilization, and 0.4 represents the weight of memory utilization.

[0092] It should be noted that the computing power load mentioned above is between (0, 1).

[0093] In some embodiments, the aforementioned maintenance device can connect to cloud devices wirelessly to obtain second status information.

[0094] It should be noted that the maintenance device can send status acquisition requests to the cloud device at preset intervals, and the cloud device can then send second status information to the maintenance device based on the status acquisition requests. In other words, the second status information is based on the real-time status changes of the computing device.

[0095] In some embodiments, the second state information of the cloud device may satisfy the preset conditions as follows: the cloud computing load is greater than or equal to a preset remaining load threshold, and the cloud storage remaining capacity is greater than or equal to a preset remaining capacity threshold.

[0096] In some embodiments, both the preset load threshold and the preset capacity threshold can be user-defined or preset by the operation and maintenance device. The specific values ​​can be determined according to actual usage requirements, and this application embodiment does not impose any limitations.

[0097] In some embodiments, the operation and maintenance equipment can determine the task complexity based on the computing power required for the operation and maintenance data.

[0098] For example, the complexity of the task can be determined by the number of CPUs, CPU utilization, or memory size used by the maintenance equipment to execute the maintenance task corresponding to the maintenance data. That is, the greater the number of CPUs, CPU utilization, or memory used by the maintenance task, the higher the task complexity.

[0099] For example, if the operation and maintenance task corresponding to the operation and maintenance data is: single device CPU overload identification, the task complexity is 0.2; if the operation and maintenance task corresponding to the operation and maintenance data is: cross-regional multi-device fault root cause location, the task complexity is 0.9.

[0100] It is understandable that the above task complexity refers to assessing the complexity of an operation and maintenance task based on its difficulty and the resources required.

[0101] In some embodiments, the maintenance equipment may determine the level of data privacy based on the amount of privacy data included in the maintenance data.

[0102] For example, the more privacy data the operation and maintenance data contains, the greater the data privacy level.

[0103] For example, if the operation and maintenance data includes database password log analysis, the data privacy level is 0.9; if the operation and maintenance data includes public network device traffic statistics, the data privacy level is 0.1.

[0104] It is understood that the aforementioned data privacy level refers to the degree of protection of operation and maintenance data during collection, storage, processing and use. Data privacy can prevent unauthorized access, disclosure or misuse, and its purpose is to ensure the confidentiality, integrity and availability of operation and maintenance data.

[0105] In some embodiments, the operation and maintenance device can determine the real-time requirements based on the urgency of the tasks corresponding to the operation and maintenance data.

[0106] For example, the shorter the urgency of the task corresponding to the operation and maintenance data, the higher the value of the real-time requirement.

[0107] For example, if the task corresponding to the operation and maintenance data is device offline alarm response, the real-time requirement is 0.9; if the task corresponding to the operation and maintenance data is historical log archiving and analysis, the real-time requirement is 0.2.

[0108] In this way, the operation and maintenance device can dynamically calculate the operation and maintenance allocation decision value according to the task complexity, data privacy and real-time requirements corresponding to the operation and maintenance data, which improves the accuracy of the operation and maintenance device in calculating the operation and maintenance allocation decision value, and can ensure the optimal utilization of resources through task complexity, data privacy and real-time requirements.

[0109] It is understandable that if the maintenance device detects that the second state information of the cloud device does not meet the preset conditions, the maintenance device will directly treat the terminal device as the maintenance device and process the maintenance data.

[0110] It should be noted that the specific implementation process of step 202 above can be implemented according to the following embodiments, and will not be repeated here to avoid repetition.

[0111] In some embodiments, combined with Figure 2 ,like Figure 3 As shown, step 202 above can be specifically implemented through step 202a below.

[0112] Step 202a: When the second state information of the cloud device is detected to meet the preset conditions, the operation and maintenance device performs a weighted summation of the terminal computing power load, terminal network bandwidth, task complexity, data privacy, and real-time requirements based on the weights corresponding to the terminal computing power load, terminal network bandwidth, task complexity, data privacy, and real-time requirements to obtain the operation and maintenance allocation decision value.

[0113] In some embodiments, the sum of the weights corresponding to the terminal computing power load, the terminal network bandwidth, the task complexity, the data privacy, and the real-time requirements is 1.

[0114] In some embodiments, the weights corresponding to the terminal computing power load, the terminal network bandwidth, the task complexity information, the data privacy information, and the real-time requirement information can be preset by the operation and maintenance device; or, user-defined. The specific weights can be determined according to actual usage requirements, and this application embodiment does not impose any limitations.

[0115] For example, the operation and maintenance equipment can multiply the weight corresponding to the terminal computing power load with the terminal computing power load to obtain the fifth value; multiply the weight corresponding to the terminal network bandwidth with the terminal network bandwidth to obtain the sixth value; multiply the weight corresponding to the task complexity with the task complexity to obtain the seventh value; multiply the weight corresponding to the data privacy with the data privacy to obtain the eighth value; multiply the weight corresponding to the real-time requirement with the real-time requirement to obtain the ninth value; and finally add the fifth, sixth, seventh, and eighth values ​​to obtain the operation and maintenance allocation decision value. Specifically, this can be achieved through the following formula (3).

[0116] (3)

[0117] in, Assign decision values ​​to operations and maintenance personnel, where W1 represents the weight corresponding to the task complexity. W1 represents the task complexity, and W2 represents the weight corresponding to data privacy. For data privacy, W3 represents the weight corresponding to real-time requirements. To meet real-time requirements, W4 represents the weight corresponding to the terminal's computing power load. W5 represents the terminal's computing power load, while W5 represents the weight corresponding to the terminal's network bandwidth. For bandwidth threshold, This refers to the terminal network bandwidth.

[0118] For example, W1=0.3, W2=0.25, W2=0.2, W4=0.15, W5=0.1.

[0119] It should be noted that, This is to ensure that no additional points are awarded once the bandwidth exceeds the threshold.

[0120] In this embodiment, the operation and maintenance device can determine whether the operation and maintenance task is executed on the terminal or in the cloud based on three dimensions: operation and maintenance task characteristics, terminal status, and cloud status, to ensure optimal resource utilization and improve overall resource utilization and operation and maintenance efficiency.

[0121] Step 203: The operation and maintenance device determines the operation and maintenance equipment to process the operation and maintenance data based on the privacy level and operation and maintenance allocation decision value corresponding to the operation and maintenance data.

[0122] In some embodiments, the aforementioned maintenance equipment is a terminal device or a cloud device.

[0123] In some embodiments, the privacy level may include three privacy levels: a first privacy level, a second privacy level, and a third privacy level.

[0124] For example, the operation and maintenance device can collect the state vectors of all terminal devices in real time. When the computing load of a terminal device is low (e.g., below 30%) and its network bandwidth is sufficient, the operation and maintenance device will, based on the operation and maintenance allocation decision value, assign newly arrived operation and maintenance tasks (e.g., tasks with low complexity) suitable for the terminal device to perform. Lower sensitivity Higher-load tasks are prioritized for allocation to cloud devices. Conversely, when a terminal device is overloaded (e.g., exceeding 80%) and the task's real-time requirements are not high, the maintenance unit will allocate the maintenance task to the cloud device for processing. Through this fine-grained scheduling based on real-time status, the computing power of previously idle terminal devices is fully utilized, the pressure on cloud devices is alleviated, thereby increasing the average resource utilization rate of the entire system from less than 40% in the existing solution to over 72% in the simulated environment.

[0125] It's important to note that the latency reduction stems from two aspects. First, rapid local processing: For maintenance tasks with high real-time requirements, the decision algorithm tends to allocate them to terminal devices. Thanks to the lightweight model of the terminal devices (inference takes only 5ms), these tasks do not require network transmission or cloud queuing, and can complete millisecond-level responses on the device side. For example, device offline alarms can be locally identified and triggered within 30ms. Second, optimized data upload: For complex tasks that must be processed on cloud devices, since the terminal devices have already cleaned and extracted the maintenance data, and the secure communication gateway only uploads necessary non-sensitive or de-identified data, the average amount of data uploaded per session is reduced by 60%. This directly reduces network transmission latency. Combined with the efficient processing of the cloud-based complex task analysis module, the total end-to-end time for a typical cross-device fault root cause analysis task, from data collection on the device side to the return of diagnostic results from the cloud, is measured to be an average of 45ms, meeting the design requirement of less than 50ms.

[0126] It should be noted that the specific implementation process of step 203 above can be found in the following embodiments, and will not be repeated here to avoid repetition.

[0127] In some embodiments, combined with Figure 2 ,like Figure 4 As shown, step 203 can be implemented through step 203a, step 203b, or step 203c as described below.

[0128] Step 203a: When the privacy level is the first privacy level, the maintenance device identifies the terminal device as the maintenance device.

[0129] In some embodiments, the aforementioned first privacy level can be understood as the privacy data included in the operation and maintenance data being greater than or equal to the first privacy quantity threshold.

[0130] In some embodiments, the aforementioned privacy threshold can be user-defined or preset by the maintenance device.

[0131] It is understandable that when the privacy level of the operation and maintenance data is the highest level, the operation and maintenance device can determine that the data included in the operation and maintenance data is basically private data. In order to ensure the security of the private data, operation and maintenance processing can only be performed on the terminal device.

[0132] Step 203b: When the privacy level is the second or third privacy level and the operation and maintenance allocation decision value is greater than or equal to the preset decision threshold, the operation and maintenance device will identify the terminal device as the operation and maintenance device.

[0133] In some embodiments, the second privacy level can be understood as the privacy data included in the operation and maintenance data being within a privacy quantity range, the maximum value of which is the first privacy quantity threshold mentioned above.

[0134] In some embodiments, the aforementioned third privacy level can be understood as the privacy data included in the operation and maintenance data being at a second privacy quantity threshold, which is the minimum value of the privacy quantity range.

[0135] In some embodiments, the aforementioned preset decision threshold can be preset by the operation and maintenance device or user-defined. The specific threshold can be determined according to actual usage requirements, and this application embodiment does not impose any limitations.

[0136] Step 203c: When the privacy level is the second or third privacy level and the operation and maintenance allocation decision value is less than the preset decision threshold, the operation and maintenance device will identify the cloud device as the operation and maintenance device.

[0137] In some embodiments, the first privacy level is greater than the second privacy level, and the second privacy level is greater than the third privacy level.

[0138] In the operation and maintenance method provided in this application embodiment, operation and maintenance data and first state information of terminal devices are obtained. The first state information includes: terminal computing power load and terminal network bandwidth. When the second state information of the cloud device is detected to meet preset conditions, an operation and maintenance allocation decision value is calculated based on the characteristic information corresponding to the terminal computing power load, terminal network bandwidth and operation and maintenance data. The second state information includes: cloud computing power load and cloud storage remaining capacity. The characteristic information includes at least: task complexity, data privacy, and real-time requirements. Based on the privacy level corresponding to the operation and maintenance data and the operation and maintenance allocation decision value, the operation and maintenance device for processing the operation and maintenance data is determined. The operation and maintenance device is either a terminal device or a cloud device. In this solution, when the second state information of the cloud device meets the preset conditions, the operation and maintenance allocation decision value is calculated by using the first state information of the terminal device and the feature information corresponding to the operation and maintenance data. Based on the privacy level of the operation and maintenance data and the operation and maintenance allocation decision value, the operation and maintenance equipment for processing the operation and maintenance data is determined. That is, when the computing power resources of the cloud device meet the preset conditions, the operation and maintenance allocation decision algorithm and privacy level enable the terminal device to make autonomous decisions and the cloud device to optimize globally. This allows the computing power resources of the cloud device and the terminal device to be coordinated and fully utilized, thereby improving the resource utilization rate and operation and maintenance efficiency.

[0139] In some embodiments, combined with Figure 3 ,like Figure 5 As shown, the operation and maintenance method provided in this application embodiment also includes the following steps 301 to 304.

[0140] Step 301: The operation and maintenance device calculates the privacy value based on the amount of privacy data contained in the operation and maintenance data.

[0141] It should be noted that the specific implementation process of step 301 above can be found in the description in the relevant technology. To avoid repetition, it will not be repeated here.

[0142] Step 302: If the privacy value is greater than the maximum value of the first privacy range, the operation and maintenance device determines the privacy level corresponding to the operation and maintenance data as the first privacy level.

[0143] For example, the operation and maintenance data corresponding to the first privacy level mentioned above may include at least one of the following: system account password and database query logs, etc.

[0144] It should be noted that the operation and maintenance data corresponding to the first privacy level mentioned above is processed only on the device side and is not uploaded to the cloud. The processing flow is "collection → preprocessing → device-side inference → local storage", with a storage period of 30 days. After the storage period expires, the data will be automatically de-identified and deleted.

[0145] Step 303: If the privacy value is within the first privacy range, the operation and maintenance device determines the privacy level corresponding to the operation and maintenance data as the second privacy level.

[0146] For example, the operation and maintenance data corresponding to the second privacy level mentioned above may include at least one of the following: CPU utilization, memory utilization, device configuration information, etc.

[0147] It should be noted that the operation and maintenance data corresponding to the second privacy level mentioned above is uploaded to the cloud after being anonymized (e.g., the device IP is replaced with the device ID). The processing flow is "collection → preprocessing → anonymization → cloud analysis → dual storage on both the end and the cloud".

[0148] Step 304: If the privacy value is less than the minimum value of the first privacy range, the operation and maintenance device determines the privacy level corresponding to the operation and maintenance data as the third privacy level.

[0149] For example, the operation and maintenance data corresponding to the third privacy level mentioned above may include at least one of the following: publicly available network traffic statistics and device model information, etc.

[0150] It should be noted that the operation and maintenance data corresponding to the third privacy level mentioned above can be directly uploaded to cloud devices, and the processing flow is "collection → cloud analysis → cloud storage".

[0151] For example, the operation and maintenance device can determine the privacy level corresponding to the operation and maintenance data using the following formula (4), which is specifically:

[0152] (4)

[0153] Where X represents operation and maintenance data, For the k-th sensitive feature indicator (if X contains sensitive feature k, then...) (X) = 1, otherwise 0). Weights for sensitive features (such as "password" features) =0.5, "IP address" characteristic =0.2), M is the total number of sensitive features.

[0154] For example, M can be 10.

[0155] It should be noted that the execution order of steps 301 to 304 described above is not limited in this embodiment. For example, the maintenance device may execute steps 301 to 304 before step 201; or, the maintenance device may execute steps 301 to 304 after step 202. The specific execution order can be determined according to actual usage requirements, and this embodiment does not impose any restrictions. Figure 6 This example illustrates the process of executing steps 301 to 304 first, followed by step 201.

[0156] In this embodiment, the operation and maintenance device can establish a data hierarchical processing and privacy protection mechanism, where core sensitive data is processed only on the device side and non-sensitive data is uploaded to the cloud, thereby improving the security of operation and maintenance data.

[0157] In some embodiments, combined with Figure 2 ,like Figure 6 As shown, after step 203 above, the operation and maintenance method provided in this application embodiment further includes the following step 401.

[0158] Step 401: When the maintenance equipment is a terminal device, the maintenance device inputs the maintenance data into the maintenance reasoning model in the terminal device, performs fault identification on the maintenance data, and obtains the maintenance results.

[0159] In some embodiments, the above-described operation and maintenance inference model is trained based on cross-entropy loss and distillation loss.

[0160] In some embodiments, the above-described operation and maintenance reasoning model can be a lightweight student model.

[0161] For example, the student model described above can be any of the following: a neural network model, an artificial intelligence (AI) model, etc. The specific model can be determined according to actual usage requirements, and this application embodiment does not impose any limitations.

[0162] The training process of the operation and maintenance inference model in the terminal device will be explained in detail below.

[0163] Step 1: The operation and maintenance device builds a training dataset.

[0164] For example, the maintenance and operation device can collect maintenance and operation fault datasets. ,in, The feature vector of operation and maintenance data (such as CPU utilization, memory usage, disk input and output) has a dimension of d (d=20 in this embodiment of the application). The fault labels are K categories (e.g., 0 = normal, 1 = CPU overload, 2 = disk failure) (K = 10 in this embodiment).

[0165] Step 2: Define the teacher model and student model for the operation and maintenance device.

[0166] For example, teacher model Using a cloud-based ResNet-50 model, the input is The output is the failure probability distribution. ,in The model outputs a feature vector before the layer, where τ is a temperature parameter (τ=5 in this embodiment); Student model The MobileNet-V2 model is used on the device side, with the input being... The output is the failure probability distribution. ,in The feature vectors of the student model.

[0167] For example, the operation and maintenance device constructs a teacher model and a student model respectively. The teacher model adopts a Transformer architecture with a large number of parameters, a vocabulary size of 50,000, a hidden layer dimension of 1024, and 24 network layers, including an embedding layer, a Transformer encoding and decoding layer, and a fully connected output layer, to extract deep fault features from the operation and maintenance data. The student model adopts a lightweight Transformer architecture, with the same vocabulary size as the teacher model, a hidden layer dimension of 512, and 12 network layers. The network structure is consistent with the teacher model, but the number of parameters is halved, adapting to the computing power of the edge side.

[0168] Step 3: Design the loss function for the operation and maintenance equipment.

[0169] For example, the training objective of the student model is to minimize the combined loss function to balance the distillation loss and the cross-entropy loss. The loss function can be shown in Equation (5).

[0170] (5)

[0171] in, For cross-entropy loss, this The cross-entropy loss is used to measure the difference between the student model and the true label; α is the weighting coefficient (α=0.7 in the embodiment of this application). For distillation loss, this The distillation loss is used to make the student model's output distribution approximate the teacher model, while preserving the teacher model's fault identification knowledge.

[0172] Step 4: The operation and maintenance equipment performs model training and optimization.

[0173] For example, the operation and maintenance device determines a weight coefficient α. This weight coefficient α is not simply fixed at 0.7, but rather a higher α value (e.g., 0.9) is set in the early stages of training (the first 20 training batches) to allow the student model to focus on imitating the reasoning process of the teacher model and quickly learn the similarity relationships between faults. As training progresses, α is gradually reduced to 0.5 to strengthen the constraint of the real labels on the student model and ensure classification accuracy. This dynamic adjustment strategy better balances knowledge transfer and classification accuracy.

[0174] Then, the operation and maintenance system performs learning rate scheduling on the student model, that is, it replaces the traditional Adam optimizer with the AdamW optimizer and introduces weight decay (set to 1e-4) to prevent the model from overfitting. The initial learning rate is set to 1e-4, and a cosine annealing strategy is used to make the learning rate decrease periodically during training, which helps the model escape local optima and converge to a smoother extreme point, thereby improving generalization ability.

[0175] Next, the operation and maintenance equipment performs data augmentation, that is, during the training phase, it augments the feature vector of the input operation and maintenance data. Adding small Gaussian noise (mean 0, standard deviation 0.01) simulates fluctuations in sensor data collected in real-world environments, enhancing the model's robustness.

[0176] Finally, the operation and maintenance device adopts an early stopping mechanism for the student model, that is, a validation set is set, and training is stopped when the loss function value on the validation set no longer decreases for 10 consecutive batches to avoid overfitting.

[0177] For example, in the training initialization phase: the maintenance device loads the pre-trained teacher model and sets it to evaluation mode, freezing the model parameters so they do not participate in training; initialize the student model parameters, set the loss weight α=0.7 and the temperature parameter τ=5.0, select the cross-entropy loss function to calculate the classification loss, select AdamW as the optimizer and set the learning rate 1e-4.

[0178] Next, in the single-batch training phase: the batch feature vectors and corresponding fault labels of the input maintenance data are used to first obtain the gradient-free soft output probability distribution through the teacher model; then the same data is input into the student model to obtain the soft output probability distribution, and the hard output of the student model is obtained at the same time to calculate the classification loss; the distillation loss and cross-entropy loss are calculated separately, and the total loss is obtained by fusion according to the α weight; the parameters of the student model are updated through the backpropagation algorithm to complete the single-batch training.

[0179] Next, in the full training phase: simulate the generation of batch training data for operation and maintenance failures, set the batch size to 32 and the sequence length to 128, and execute a total of 100 rounds of training, with each round containing 100 training batches. After each round of training, calculate the average loss, monitor the model training effect by the change in loss, and ensure model convergence.

[0180] Finally, the model saving stage: After training is completed, the parameter file of the student model is saved in binary format for subsequent edge deployment.

[0181] Thus, through the above steps, training and validation were performed on an operational dataset containing 100,000 samples, and the student model ultimately achieved an accuracy of 92.3% on the test set. This accuracy is a specific result achieved under a particular dataset and optimization strategy, rather than a general target.

[0182] The following section provides a detailed explanation of the quantification and deployment of the student model.

[0183] For example, to further reduce the resource consumption on the edge, the trained student model is quantized using INT8, converting the model weights from 32-bit floating-point numbers (FP32) to 8-bit integers (INT8). The quantization formula (6) is as follows:

[0184] (6)

[0185] in, σ is the original weight, μ is the weight mean, σ is the weight standard deviation, and round() is the rounding function.

[0186] In this way, the quantized model size is reduced to 4MB, memory usage is reduced by 75%, and inference time is further reduced to 5ms, making it deployable on edge terminal devices with limited computing power.

[0187] In this embodiment of the application, the terminal device can be designed with a collaborative mechanism between the edge-side lightweight inference engine and the cloud-based large-scale analysis engine to reduce the end-to-end latency of operation and maintenance tasks to less than 50ms, thereby meeting the real-time operation and maintenance requirements.

[0188] In some embodiments, combined with Figure 2 ,like Figure 7 As shown, after step 203 above, the operation and maintenance method provided in this application embodiment further includes the following steps 501 and 502.

[0189] Step 501: When it is determined that the maintenance equipment is a cloud-based device and an abnormal state of the cloud-based device is detected, the maintenance device controls the terminal device to identify the fault based on the maintenance data and the first local emergency strategy stored in the terminal device, and obtain the maintenance result.

[0190] In some embodiments, the first local emergency response strategy is sent to the terminal device when the cloud device is in a normal state.

[0191] In some embodiments, the above-mentioned abnormal state may correspond to any of the following: network latency greater than a preset threshold, heartbeat packet not responding, communication not replying within a preset time period, etc.

[0192] For example, the operation and maintenance device can detect that the cloud device is in an abnormal state as follows: the operation and maintenance device detects the communication status (network latency, heartbeat response) with the cloud at a period of 100ms. When the network latency is detected to be >500ms and lasts for 3 seconds, it is determined to be a network interruption; when the cloud heartbeat response is detected to be non-responsive and lasts for 5 seconds, it is determined to be a cloud failure.

[0193] Step 502: Upon detecting that the cloud device has returned to normal status, the maintenance device sends the maintenance results to the cloud device and forwards the second local emergency policy sent by the cloud device to the terminal device.

[0194] In some embodiments, the maintenance device can send non-privacy maintenance data to cloud devices.

[0195] For example, the following provides a detailed explanation of the scenario where the cloud device is in an abnormal state, as provided in the embodiments of this application.

[0196] Step 10: The operation and maintenance device performs local resource pre-configuration for the terminal equipment.

[0197] For example, the operation and maintenance device can have a basic operation and maintenance strategy library (containing execution rules for 50+ basic operation and maintenance actions), a lightweight fault diagnosis model (already deployed locally, inference does not require cloud support), and a local storage module (capable of storing at least 7 days of operation and maintenance data) built into the terminal device. All resources that complete basic operation and maintenance are deployed locally and do not depend on the cloud.

[0198] Step 11: The maintenance equipment performs anomaly detection on cloud devices.

[0199] For example, the maintenance device can detect the communication status (network latency, heartbeat response) with the cloud device at a period of 100ms. When a network latency of >500ms is detected and lasts for 3 seconds, it is determined to be a network interruption; when a cloud heartbeat response is detected and lasts for 5 seconds, it is determined to be a cloud device failure.

[0200] Step 12: The operation and maintenance device control terminal equipment performs operation and maintenance processing.

[0201] For example, when the maintenance device detects a network latency greater than 500ms, it controls the terminal device to switch to "offline mode" so that the maintenance device stops uploading data and retains only local processing capabilities. It can independently complete basic fault diagnosis (such as insufficient disk space or abnormal network port). After the network is restored, it uploads non-sensitive data from the offline period in batches.

[0202] For example, when a cloud device fails, the terminal device cannot receive the cloud policy. The operation and maintenance device controls the terminal device to automatically enable the "local emergency policy" (such as performing data collection and fault handling according to the historical policy). The local emergency policy is the latest policy issued when the cloud is working normally. The terminal side will synchronize and save it in real time. When a fault occurs, the policy is directly called without the need for new instructions from the cloud. After the cloud recovers, the local operation and maintenance data is synchronized to achieve seamless switching.

[0203] In this embodiment, the operation and maintenance device can achieve adaptive fault handling capabilities. In abnormal scenarios such as network interruption and cloud failure, the terminal device can independently complete basic operation and maintenance tasks (such as local device restart and temporary alarm blocking) to ensure the continuity of operation and maintenance.

[0204] In some embodiments, combined with Figure 2 ,like Figure 8 As shown, after step 203 above, the operation and maintenance method provided in this application embodiment further includes the following steps 601 and 602.

[0205] Step 601: When it is determined that the maintenance equipment is a terminal device and the terminal device is detected to be in an abnormal state, the maintenance device controls the cloud device to perform fault identification based on the maintenance data and obtain the maintenance result.

[0206] In some embodiments, the operation and maintenance device can control the cloud device to perform fault identification through the teacher model in the cloud device and obtain the operation and maintenance results.

[0207] Step 602: When the terminal device is detected to have returned to normal status, the maintenance device sends the maintenance results to the terminal device.

[0208] In this embodiment, the maintenance device can control the maintenance equipment to perform maintenance processing when it detects that the terminal device is in an abnormal state, thereby improving the timeliness of maintenance processing.

[0209] The operation and maintenance methods of this application are described below through specific embodiments.

[0210] like Figure 9 As shown, the implementation process of the operation and maintenance method provided in this application embodiment includes the following S1 to S10.

[0211] S1. The operation and maintenance device collects operation and maintenance data.

[0212] S2, the maintenance equipment performs data preprocessing.

[0213] For example, the above data preprocessing may include data cleaning and operation and maintenance data feature extraction.

[0214] S3. The operation and maintenance device performs a classification and grading of operation and maintenance data.

[0215] S4. When the maintenance data level is the first privacy level, the maintenance device will identify the terminal device as the maintenance device.

[0216] S5. When the maintenance data level is the second privacy level, the maintenance device performs data anonymization processing on the privacy data in the maintenance data, and determines the maintenance device based on the status of the cloud device and the maintenance allocation decision value.

[0217] S6. When the maintenance data level is the third privacy level, the maintenance device determines the maintenance equipment based on the status of the cloud equipment and the maintenance allocation decision value.

[0218] S7. The operation and maintenance device monitors in real time whether the cloud devices are in an abnormal state.

[0219] S8. When the maintenance device detects that the cloud device is in an abnormal state, the maintenance device will switch the terminal device to offline state and control the terminal device to complete the maintenance process.

[0220] S9. When the maintenance device detects that the cloud device has returned to normal, the maintenance device will send the maintenance data and maintenance processing results to the cloud device.

[0221] S10. When the maintenance device detects that the cloud device is in a normal state, the maintenance device sends the maintenance data to the cloud device and controls the cloud device to complete the maintenance process.

[0222] Thus, by constructing a two-way collaborative architecture of "edge-side autonomous decision-making + cloud-side global optimization" and using a dynamic task scheduling algorithm, the utilization rate of edge-cloud resources can be increased to over 70%. Furthermore, a collaborative mechanism between the lightweight edge-side inference engine and the large-scale cloud-side analysis engine is designed to reduce end-to-end latency of maintenance tasks to less than 50ms, meeting real-time maintenance requirements. A data tiered processing and privacy protection mechanism is also established, with core sensitive data processed only on the edge and non-sensitive data uploaded to the cloud, complying with data security compliance requirements. Finally, adaptive fault handling capabilities are achieved, enabling the edge to independently complete basic maintenance tasks (such as local device restarts and temporary alarm masking) in abnormal scenarios such as network interruptions and cloud failures, ensuring maintenance continuity.

[0223] It should be noted that the descriptions of each step S1 to S10 in this embodiment can be found in the descriptions in the above embodiments, and will not be repeated here.

[0224] It should be noted that the above-described method embodiments, or the various possible implementations of the method embodiments, can be executed individually, or, provided there is no conflict, they can be combined with each other. The specific implementation can be determined according to actual usage requirements, and this application embodiment does not impose any restrictions on this.

[0225] Figure 10 This is a schematic diagram of the structure of an operation and maintenance system provided in an embodiment of this application. Figure 10 As shown, the operation and maintenance system 800 may include: a cloud-based global scheduling center 801, which includes: a resource management module 8010, a complex task analysis module 8011, and a collaborative scheduling module 8012; a secure communication gateway 802; and N lightweight intelligent agents 803, where N is a positive integer. Each lightweight intelligent agent 803 includes: a data acquisition module 8031, a data preprocessing module 8032, a lightweight inference module 8032, and a local execution module 8033.

[0226] The data acquisition module 8031 ​​is used to collect operation and maintenance data (metric data such as CPU / memory utilization, log data such as system error logs, and alarm data such as device offline alarms) through sensors and application programming interfaces (APIs), and supports more than 1,000 device protocols (SNMP, Modbus, HTTP). It is applied to step 201 above and related steps.

[0227] The data preprocessing module 8032 is used to clean (remove outliers) and extract features (such as calculating the 5-minute average of CPU utilization) of the operation and maintenance data. It is applied to the above step 202 and related steps.

[0228] The lightweight inference module 8032 is used to perform basic fault identification (such as insufficient disk space or abnormal network port). It is applied to step 401 above, and related steps.

[0229] The local execution module 8033 is used to perform operation and maintenance operations (such as device restart and configuration distribution), supporting 50+ common operation and maintenance actions. It is applied to step 501 above, and related steps.

[0230] The secure communication gateway 802 is used to enable encrypted communication between terminal devices and cloud devices. It supports the TLS 1.3 protocol and has data filtering capabilities, allowing only non-sensitive data to be uploaded to the cloud. It is applied to step 502 above, and related steps.

[0231] The resource management module 8010 is used to monitor the computing load (CPU utilization, memory usage) and network status (bandwidth, latency) of all terminal devices in real time. It is applied to step 201 above and related steps.

[0232] The complex task analysis module 8011 is used to run large-scale deep learning models to complete complex operation and maintenance tasks (such as fault root cause localization and capacity prediction for the next 72 hours). It is applied to step 203 above, as well as related steps to step 201.

[0233] The collaborative scheduling module 8012 is used to dynamically allocate tasks based on the edge-cloud status and issue global operation and maintenance policies (such as data upload thresholds and fault handling priorities). It is applied to step 602 above and related steps.

[0234] It should be noted that for a detailed explanation of the steps performed by each module and their beneficial effects, please refer to the description in the above embodiments, which will not be repeated here.

[0235] As can be seen, the above mainly describes the solutions provided by the embodiments of this application from a methodological perspective. To achieve the above functions, the embodiments of this application provide corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the modules and algorithm steps of the various examples described in the embodiments disclosed herein, the embodiments of this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0236] This application embodiment can divide the operation and maintenance device into functional modules according to the above method example. For example, each function can be divided into a separate functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. Optionally, the module division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0237] In some embodiments, this application also provides an operation and maintenance device. This operation and maintenance device may include one or more functional modules for implementing the operation and maintenance methods of the above method embodiments.

[0238] For example, Figure 11 This is a schematic diagram of the structure of an operation and maintenance device provided in an embodiment of this application. Figure 11As shown, the operation and maintenance device 900 includes: an acquisition module 901, a processing module 902, and a determination module 903. The acquisition module 901 is used to acquire operation and maintenance data and first status information of the terminal device. The first status information includes: terminal computing load and terminal network bandwidth. The processing module 902 is used to calculate an operation and maintenance allocation decision value based on the terminal computing load, terminal network bandwidth, and characteristic information corresponding to the operation and maintenance data, when it detects that the second status information of the cloud device meets preset conditions. The second status information includes: cloud computing load and remaining cloud storage capacity. The characteristic information includes at least task complexity, data privacy, and real-time requirements. The determination module 903 is used to determine the operation and maintenance device that processes the operation and maintenance data based on the privacy level corresponding to the operation and maintenance data and the operation and maintenance allocation decision value. The operation and maintenance device is either a terminal device or a cloud device.

[0239] In the operation and maintenance device provided in this application, when the second state information of the cloud device is detected to meet the preset conditions, the operation and maintenance allocation decision value is calculated by using the first state information of the terminal device and the feature information corresponding to the operation and maintenance data. Then, based on the privacy level corresponding to the operation and maintenance data and the operation and maintenance allocation decision value, the operation and maintenance device for processing the operation and maintenance data is determined. That is, when the computing power resources of the cloud device are detected to meet the preset conditions, the terminal device can make autonomous decisions and the cloud device can be optimized globally through the operation and maintenance allocation decision algorithm and privacy level. This enables the computing power resources of the cloud device and the terminal device to be coordinated and fully utilized, thereby improving the resource utilization rate and operation and maintenance efficiency.

[0240] In some embodiments, the determining module 903 is specifically used to determine the terminal device as an operation and maintenance device when the privacy level is a first privacy level; or, when the privacy level is a second privacy level or a third privacy level and the operation and maintenance allocation decision value is greater than or equal to a preset decision threshold, the terminal device is determined as an operation and maintenance device; or, when the privacy level is a second privacy level or a third privacy level and the operation and maintenance allocation decision value is less than a preset decision threshold, the cloud device is determined as an operation and maintenance device; wherein, the first privacy level is greater than the second privacy level, and the second privacy level is greater than the third privacy level.

[0241] In other embodiments, the processing module 902 is further configured to calculate a privacy value based on the amount of privacy data contained in the operation and maintenance data; if the privacy value is greater than the maximum value of the first privacy range, determine the privacy level corresponding to the operation and maintenance data as the first privacy level; if the privacy value is within the first privacy range, determine the privacy level corresponding to the operation and maintenance data as the second privacy level; and if the privacy value is less than the minimum value of the first privacy range, determine the privacy level corresponding to the operation and maintenance data as the third privacy level.

[0242] In some other embodiments, the processing module 902 is further configured to determine the maintenance equipment for processing maintenance data based on the privacy level and maintenance allocation decision value corresponding to the maintenance data, and then, if the maintenance equipment is a terminal device, input the maintenance data into the maintenance inference model in the terminal device to perform fault identification on the maintenance data and obtain maintenance results; wherein, the maintenance inference model is trained based on cross-entropy loss and distillation loss.

[0243] In some other embodiments, the processing module 902 is further configured to, after determining the maintenance device for processing the maintenance data based on the privacy level and maintenance allocation decision value corresponding to the maintenance data, and in the case that the maintenance device is a cloud device and the cloud device is detected to be in an abnormal state, control the terminal device to perform fault identification based on the maintenance data and the first local emergency policy stored on the terminal device to obtain the maintenance result, wherein the first local emergency policy is to be sent to the terminal device when the cloud device is in a normal state; and in the case that the cloud device is detected to have returned to a normal state, send the maintenance result to the cloud device and forward the second local emergency policy sent by the cloud device to the terminal device.

[0244] In some other embodiments, the processing module 902 is further configured to determine the maintenance device for processing maintenance data based on the privacy level and maintenance allocation decision value corresponding to the maintenance data; and, if the maintenance device is determined to be a terminal device and the terminal device is detected to be in an abnormal state, control the cloud device to perform fault identification based on the maintenance data to obtain maintenance results; and if the terminal device is detected to have returned to normal state, send the maintenance results to the terminal device.

[0245] In some other embodiments, the processing module 902 is specifically used to perform a weighted summation of the terminal computing power load, terminal network bandwidth, task complexity, data privacy, and real-time requirements based on the weights corresponding to the terminal computing power load, the terminal network bandwidth, the task complexity, the data privacy, and the real-time requirements, to obtain an operation and maintenance allocation decision value; wherein the sum of the weights corresponding to the terminal computing power load, the terminal network bandwidth, the task complexity, the data privacy, and the real-time requirements is 1.

[0246] It should be noted that the operation and maintenance device can implement all the processes implemented in the above method embodiments and achieve the same beneficial effects. To avoid repetition, it will not be described again here.

[0247] In the case where the functions of the integrated modules described above are implemented in hardware, this application provides a possible structural schematic diagram of the electronic device involved in the above embodiments. For example... Figure 12As shown, the electronic device 90 includes: a processor 92, a communication interface 93, and a bus 94. Optionally, the electronic device 90 may also include a memory 91.

[0248] Processor 92 may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 92 may be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 92 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0249] Communication interface 93 is used to connect with other devices via a communication network. This communication network can be Ethernet, wireless access network, wireless local area network (WLAN), etc.

[0250] The memory 91 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto.

[0251] As one possible implementation, the memory 91 can exist independently of the processor 92. The memory 91 can be connected to the processor 92 via a bus 94 and is used to store instructions or program code. When the processor 92 calls and executes the instructions or program code stored in the memory 91, it can implement the operation and maintenance method provided in the embodiments of this application.

[0252] In another possible implementation, memory 91 can also be integrated with processor 92.

[0253] Bus 94 can be an Extended Industry Standard Architecture (EISA) bus, etc. Bus 94 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 12 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0254] Through the above description of the implementation methods, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the service calling device can be divided into different functional modules to complete all or part of the functions described above.

[0255] This application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the various processes of the above-described operation and maintenance method embodiments and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0256] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0257] This application also provides a readable storage medium storing programs or instructions, which, when executed by a computer, implement the operation and maintenance methods provided in the above embodiments. It is understood that all or part of the processes in the above method embodiments can be executed by computer instructions instructing related hardware; the readable storage medium can be any of the foregoing embodiments or memory; the readable storage medium can also be an external storage device of the service invocation device, such as a pluggable hard drive, Smart Media Card (SMC), Secure Digital (SD) card, flash card, etc., equipped on the service invocation device. Further, the readable storage medium can include both internal storage units of the service invocation device and external storage devices. The readable storage medium is used to store the computer program and other programs and data required by the service invocation device. The readable storage medium can also be used to temporarily store data that has been output or will be output.

[0258] This application also provides a computer program product, which is stored in a storage medium and implements the operation and maintenance method provided in the above embodiments when the computer program product is executed by a computer.

[0259] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0260] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0261] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. An operation and maintenance method, characterized in that, include: Obtain operation and maintenance data and the first status information of terminal devices, the first status information including: terminal computing load and terminal network bandwidth; When the second state information of the cloud device is detected to meet the preset conditions, the operation and maintenance allocation decision value is calculated based on the terminal computing power load, the terminal network bandwidth and the feature information corresponding to the operation and maintenance data. The second state information includes: cloud computing power load and cloud storage remaining capacity. The feature information includes at least: task complexity, data privacy and real-time requirements. Based on the privacy level corresponding to the operation and maintenance data and the operation and maintenance allocation decision value, the operation and maintenance equipment that processes the operation and maintenance data is determined, and the operation and maintenance equipment is the terminal device or the cloud device.

2. The operation and maintenance method according to claim 1, characterized in that, The step of determining the maintenance equipment to process the maintenance data based on the privacy level corresponding to the maintenance data and the maintenance allocation decision value includes: If the privacy level is set to the first privacy level, the terminal device will be identified as the maintenance device; or, If the privacy level is the second or third privacy level, and the operation and maintenance allocation decision value is greater than or equal to a preset decision threshold, the terminal device is identified as the operation and maintenance device; or, If the privacy level is the second privacy level or the third privacy level, and the operation and maintenance allocation decision value is less than the preset decision threshold, the cloud device is identified as the operation and maintenance device. Wherein, the first privacy level is greater than the second privacy level, and the second privacy level is greater than the third privacy level.

3. The operation and maintenance method according to claim 2, characterized in that, The method further includes: A privacy value is calculated based on the amount of privacy data contained in the operation and maintenance data; If the privacy value is greater than the maximum value of the first privacy range, the privacy level corresponding to the operation and maintenance data is determined as the first privacy level; If the privacy value is within the first privacy range, the privacy level corresponding to the operation and maintenance data is determined as the second privacy level; If the privacy value is less than the minimum value of the first privacy range, the privacy level corresponding to the operation and maintenance data is determined as the third privacy level.

4. The operation and maintenance method according to claim 1, characterized in that, After determining the maintenance equipment for processing the maintenance data based on the privacy level corresponding to the maintenance data and the maintenance allocation decision value, the method further includes: When the maintenance equipment is the terminal equipment, the maintenance data is input into the maintenance reasoning model in the terminal equipment, and the maintenance data is used to identify faults and obtain maintenance results. The operation and maintenance inference model is trained based on cross-entropy loss and distillation loss.

5. The operation and maintenance method according to claim 1, characterized in that, After determining the maintenance equipment for processing the maintenance data based on the privacy level corresponding to the maintenance data and the maintenance allocation decision value, the method further includes: If the maintenance equipment is determined to be the cloud equipment and the cloud equipment is detected to be in an abnormal state, the terminal equipment is controlled to perform fault identification based on the maintenance data and the first local emergency policy stored in the terminal equipment to obtain the maintenance result. The first local emergency policy is issued to the terminal equipment when the cloud equipment is in a normal state. Upon detecting that the cloud device has returned to normal, the operation and maintenance results are sent to the cloud device, and the second local emergency policy sent by the cloud device is forwarded to the terminal device.

6. The operation and maintenance method according to claim 1, characterized in that, After determining the maintenance equipment for processing the maintenance data based on the privacy level corresponding to the maintenance data and the maintenance allocation decision value, the method further includes: If the maintenance equipment is determined to be the terminal equipment and the terminal equipment is detected to be in an abnormal state, the cloud equipment is controlled to perform fault identification based on the maintenance data to obtain maintenance results. Upon detecting that the terminal device has returned to normal operation, the maintenance result is sent to the terminal device.

7. The operation and maintenance method according to claim 1, characterized in that, The step of calculating the operation and maintenance allocation decision value based on the terminal computing power load, the terminal network bandwidth, and the characteristic information corresponding to the operation and maintenance data includes: Based on the weights corresponding to the terminal computing power load, the terminal network bandwidth, the task complexity, the data privacy, and the real-time requirements, a weighted sum is performed on the terminal computing power load, the terminal network bandwidth, the task complexity, the data privacy, and the real-time requirements to obtain the operation and maintenance allocation decision value. The sum of the weights corresponding to the terminal computing power load, the terminal network bandwidth, the task complexity, the data privacy, and the real-time requirements is 1.

8. A maintenance device, characterized in that, include: Acquisition module, processing module, and determination module; The acquisition module is used to acquire operation and maintenance data and the first status information of the terminal device, the first status information including: terminal computing power load and terminal network bandwidth; The processing module is used to calculate the operation and maintenance allocation decision value based on the terminal computing power load, the terminal network bandwidth and the feature information corresponding to the operation and maintenance data when the second state information of the cloud device meets the preset conditions. The second state information includes: cloud computing power load and cloud storage remaining capacity. The feature information includes at least: task complexity, data privacy and real-time requirements. The determining module is used to determine the operation and maintenance equipment that processes the operation and maintenance data based on the privacy level corresponding to the operation and maintenance data and the operation and maintenance allocation decision value. The operation and maintenance equipment is the terminal device or the cloud device.

9. An electronic device, characterized in that, It includes a processor and a memory, the memory storing programs or instructions that can run on the processor, the programs or instructions being executed by the processor to implement the operation and maintenance method as described in any one of claims 1-7.

10. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions, which, when executed by a computer, implement the operation and maintenance method as described in any one of claims 1-7.

11. A computer program product, characterized in that, The computer program product is stored in a storage medium, and when the computer program product is executed by a computer, it implements the operation and maintenance method as described in any one of claims 1-7.

Images