Board card resource management method and device and electronic equipment

Abstract

The application discloses a board card resource management method and device and electronic equipment, and belongs to the technical field of artificial intelligence and transmission equipment management, to solve the problems of insufficient board card type identification accuracy and fixed idle rate threshold that cannot be dynamically adjusted in related technologies. The method comprises the following steps: determining the board card types of each board card in a transmission device according to physical characteristic information and / or electrical signal data of the board cards; the board card types comprise at least one of the following: a service board card, a system board card and a new type board card; determining the service load states of the board cards according to real-time load data of the board cards monitored in real time; the real-time load data comprises at least one of the following: processor utilization, memory occupancy and network bandwidth occupancy; and determining the idle rate thresholds of the board cards according to the board card types and the service load states of the board cards by using an intelligent adjustment algorithm; the intelligent adjustment algorithm is an improved genetic algorithm combined with a simulated annealing algorithm.

Description

Technical Field

[0001] This application belongs to the field of artificial intelligence and transmission equipment management technology, specifically relating to a board resource management method, device and electronic equipment. Background Technology

[0002] In electronic devices and systems, circuit boards, as critical hardware components, directly impact system resource utilization efficiency and overall performance due to their idle rate. Different types of circuit boards exhibit different characteristics under varying workloads and operating environments. For example, service circuit boards may experience higher loads during certain periods and lower loads at others, while system circuit boards typically require high stability and reliability, resulting in relatively smaller load fluctuations. Furthermore, with the continuous introduction of new circuit boards, their performance and characteristics differ from traditional boards, making idle rate management more complex. Therefore, how to rationally adjust the idle rate threshold of circuit boards based on their type and dynamic changes in workload to achieve optimal resource allocation has become a crucial task in transmission equipment management.

[0003] However, the relevant technologies still have many shortcomings in this field. First, in terms of board type identification, current methods mostly rely on manual judgment or simple labeling, which lacks accuracy, especially in distinguishing different board types that are similar in appearance or function. Furthermore, the relevant identification technologies lack universality, making it difficult to adapt to the diverse board types from different manufacturers and models. Moreover, when introducing new board types, identification methods need to be redeveloped, increasing the complexity of system maintenance and upgrades. At the same time, some identification methods have long response times, making it difficult to meet real-time requirements, and they depend on specific hardware or software environments, limiting their application scope.

[0004] Secondly, regarding the dynamic adjustment of idle rate thresholds, current solutions often employ fixed threshold settings, lacking flexibility and failing to adjust according to real-time changes in business load, leading to low resource utilization efficiency or impaired system performance. Furthermore, these solutions are mostly based on simple rules or empirical values, lacking intelligent analysis of the overall system status, resulting in delayed responses and difficulty in quickly adapting to business fluctuations. More importantly, these technologies struggle to achieve a practical balance between multiple objectives such as improving resource utilization, ensuring system performance, and reducing energy consumption, thus affecting the overall optimization effect of the system.

[0005] Therefore, how to improve the accuracy, versatility, and real-time performance of board type identification, and intelligently adjust the idle rate threshold based on the characteristics of different types of boards and dynamic business load, so as to achieve efficient resource utilization and stable system operation, is a key issue that urgently needs to be addressed in the current technology field. Summary of the Invention

[0006] This application provides a board resource management method, device, and electronic device that can solve the problems of insufficient accuracy in board type identification and fixed idle rate threshold that cannot be dynamically adjusted in related technologies.

[0007] In a first aspect, embodiments of this application provide a board resource management method, comprising: determining the board type of each board based on the physical characteristic information and / or electrical signal data of each board in a transmission device; the physical characteristic information includes at least one of the following: size information, weight information, port information, and heat dissipation structure information; the board type includes at least one of the following: service board, system board, and new type board; determining the service load status of each board based on real-time load data of each board monitored in real time; the real-time load data includes at least one of the following: processor utilization, memory occupancy, and network bandwidth occupancy; and determining an idle rate threshold for each board based on the board type and the service load status using an intelligent adjustment algorithm; the intelligent adjustment algorithm is an improved genetic algorithm combined with a simulated annealing algorithm.

[0008] Secondly, embodiments of this application provide a distributed data processing system. The system is based on a Hadoop distributed architecture and achieves separation of Driver and Hive Client and combination of Driver and Hive Service by improving the Hive architecture. The distributed data processing system is used to implement the steps of the board resource management method as described in the first aspect.

[0009] Thirdly, embodiments of this application provide a board resource management device, comprising: a first determining module, configured to determine the board type of each board based on the physical characteristic information and / or electrical signal data of each board in the transmission device; the physical characteristic information includes at least one of the following: size information, weight information, port information, and heat dissipation structure information; the board type includes at least one of the following: service board, system board, and new type board; a second determining module, configured to determine the service load status of each board based on real-time load data of each board monitored in real time; the real-time load data includes at least one of the following: processor utilization, memory occupancy, and network bandwidth occupancy; a third determining module, configured to determine the idle rate threshold of each board based on the board type and the service load status of each board using an intelligent adjustment algorithm; the intelligent adjustment algorithm is an improved genetic algorithm combined with a simulated annealing algorithm.

[0010] Fourthly, embodiments of this application provide an electronic device including a processor; and a memory arranged to store computer-executable instructions configured to be executed by the processor to implement the steps of the board resource management method as described in the first aspect.

[0011] Fifthly, embodiments of this application provide a computer-readable storage medium for storing computer-executable instructions, which, when executed by a processor, implement the steps of the board resource management method as described in the first aspect.

[0012] In a sixth aspect, embodiments of this application provide a computer program product, the computer program product including a computer program that, when executed by a processor, implements the steps of the board resource management method as described in the first aspect.

[0013] In a seventh aspect, embodiments of this application provide a chip, the chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to execute executable instructions to implement the steps of the board resource management method as described in the first aspect.

[0014] In this embodiment, the board type of each board is determined based on its physical characteristics (such as size, weight, port, and heat dissipation structure) and / or electrical signal data. This avoids the bias of relying on a single criterion and ensures accurate classification of service boards, system boards, and new types of boards. Furthermore, the inclusion of new types of boards in the identification scope facilitates rapid adaptation during transmission equipment upgrades, eliminating the need for frequent adjustments to the identification logic and helping the transmission equipment cope with technological updates. Based on real-time load data (such as processor utilization, memory usage, and network bandwidth usage) of each board, the service load status of each board is determined. Then, using an intelligent adjustment algorithm, the idle rate threshold of each board is determined based on its board type and service load status. This intelligent adjustment algorithm combines an improved genetic algorithm with a simulated annealing algorithm. This overcomes the problem of fixed idle rate thresholds that cannot be dynamically adjusted. Moreover, the combination of the improved genetic algorithm and the simulated annealing algorithm can quickly find the optimal threshold in different scenarios, which is more scientific, faster-responding, and adaptable to complex and changing load environments than manual settings. As can be seen, this technical solution provides a reliable foundation for dynamic threshold adjustment through accurate board type identification, avoiding threshold setting errors caused by type misjudgment. Dynamic threshold adjustment maximizes the value of various boards, forming a closed loop of accurate identification, intelligent adjustment and efficient operation, thereby improving the overall performance, stability and resource utilization of transmission equipment. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating a board resource management method provided in an embodiment of this application; Figure 2 This is a schematic diagram illustrating a method for determining the idle rate threshold adjustment amount provided in an embodiment of this application; Figure 3 This is a schematic diagram illustrating a method for monitoring the idle rate of circuit boards provided in an embodiment of this application; Figure 4 This is a schematic diagram of a distributed data processing system provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of a board resource management device provided in an embodiment of this application; Figure 6 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0017] 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.

[0018] In electronic devices and systems, the idle rate of circuit boards directly affects resource utilization efficiency. An excessively high idle rate can lead to resource waste, while an excessively low idle rate can affect system stability and reliability. Therefore, it is necessary to dynamically adjust the idle rate threshold based on the characteristics and requirements of different types of circuit boards. Different types of circuit boards experience varying load conditions under different business scenarios. For example, business circuit boards may have high loads during certain periods and low loads during others; system circuit boards typically require a certain level of stability and reliability, and their load fluctuations are relatively small. Dynamically adjusting the idle rate threshold based on changes in business load can better adapt to different business needs, improving system performance and stability, and achieving optimal resource allocation. Furthermore, new types of circuit boards may have different performance and characteristics, and their idle rate requirements may differ from traditional circuit boards. Therefore, it is necessary to dynamically adjust the idle rate threshold based on the characteristics of new circuit boards to fully leverage their advantages while avoiding resource waste and system instability.

[0019] Related technologies generally identify idle cards through the following points, including policy formulation for parameters such as data transmission status, data transmission direction, temperature detection, remote access status, and time thresholds. Regarding data transmission status, it can monitor whether the device containing the card has no data transmission for a continuous period. Regarding data transmission direction, it can monitor whether the card on the transmission device only receives data without sending, or only sends without receiving, which may indicate a specific state of the card. Regarding temperature detection, it can detect the card's temperature; when the temperature remains at a low level for an extended period, it may indicate that the card is idle. Regarding remote access status, it can determine whether the card can be configured or monitored through a remote access interface. In some cases, the inability to access remotely may indicate that the card is idle. Regarding time thresholds, a time threshold can be set; when the card has not been used for a certain period of time, it is considered idle.

[0020] However, the relevant technologies still have many shortcomings. Regarding board type identification, firstly, there is a lack of accuracy. Specifically, current methods may rely on manual judgment or simple label reading, which is prone to errors. For example, judging solely based on the label on the board may lead to misjudgment due to label wear, incorrect labeling, or human error. For boards with similar appearances or functions, it is difficult to accurately distinguish whether they belong to a business board, a system board, or a new type of board. Secondly, there is a lack of universality. Specifically, boards from different manufacturers and models may have different characteristics and identification methods, and existing identification technologies may not be applicable to multiple types of boards, resulting in poor universality. When a new type of board is introduced into the system, it may be necessary to redevelop the identification method, increasing maintenance and upgrade costs. Thirdly, there is a problem with poor real-time performance. Specifically, some identification methods may require a long time for analysis and judgment, which cannot meet the needs of systems with high real-time requirements. For example, in situations requiring rapid deployment or troubleshooting, the inability to determine the board type in a timely manner will affect the system's recovery speed. Furthermore, there is also the problem of dependence on specific environments. Specifically, some identification technologies may rely on specific hardware devices, software environments, or network connections, limiting their application in different scenarios. For example, specific card readers or interfaces are required to read board information, increasing system complexity and cost.

[0021] Regarding the dynamic adjustment of idle rate thresholds, firstly, there is the problem of rigidity and fixation. Specifically, related technologies may use fixed idle rate thresholds, unable to dynamically adjust according to actual conditions. This may lead to unreasonable resource utilization under different business loads, board types, or system states. Fixed thresholds may fail to adapt to dynamic changes in business, such as sudden business peaks or troughs, resulting in system performance degradation or resource waste. Secondly, there is the problem of a lack of intelligence. Specifically, current adjustment methods may be based on simple rules or empirical values, lacking a deep understanding and intelligent analysis of the overall system state. For example, adjustments may be made only based on board usage time or load conditions, without considering other system factors such as performance requirements and reliability requirements. Thirdly, there is the problem of response lag. Specifically, when the system state changes, related technologies may require a long time to detect and adjust the idle rate threshold. This may lead to unreasonable system resource utilization for a period of time after the change occurs, affecting system performance. In addition, there is the problem of difficulty in balancing multiple objectives. Specifically, in practical applications, it may be necessary to consider multiple objectives simultaneously, such as improving resource utilization, ensuring system performance, and reducing energy consumption. Current adjustment methods may struggle to find a balance among these objectives, resulting in the optimization of one objective at the expense of others.

[0022] In response, this application provides a board resource management method, device, and electronic device. By automatically identifying board types and dynamically adjusting idle rate thresholds through algorithms, the cost and error rate of manual intervention can be reduced, the efficiency and accuracy of system management can be improved, and the system can be better adapted to the needs of technological innovation, promoting continuous system upgrades and optimization.

[0023] The board resource management method, apparatus, and electronic equipment provided in this application will be described in detail below with reference to the accompanying drawings and through specific embodiments and application scenarios.

[0024] Figure 1 This application illustrates a board resource management method according to an embodiment. This method can be executed by an electronic device, which may include a server and / or a terminal device, such as an in-vehicle terminal or a mobile terminal. In other words, the method can be executed by software or hardware installed in the electronic device, and the method includes the following steps: Step 102: Determine the board type of each board based on the physical characteristics and / or electrical signal data of each board in the transmission equipment.

[0025] The physical characteristics information may include at least one of the following: size information, weight information, port information, and heat dissipation structure information. Size information may include length, width, and thickness. Port information may include the number and arrangement of ports. Heat dissipation structure information may include the size, shape, and location of the heat sink, as well as the number and type of fans. The board type may include at least one of the following: service board, system board, or new type board.

[0026] Step 104: Determine the service load status of each board based on the real-time load data of each board that is monitored in real time.

[0027] The real-time load data may include at least one of the following: processor utilization, memory usage, and network bandwidth usage.

[0028] In practice, business load status can be comprehensively assessed using metrics such as processor utilization, memory usage, and network bandwidth usage. By collecting this data in real time, a dynamic load profile can be created. For example, processor utilization reflects the use of computing resources, memory usage reveals memory resource consumption, and network bandwidth usage reflects data traffic load. Combining these metrics, the current business activity level of the board is determined based on a preset business response time threshold. If the response time exceeds this threshold, the business activity level is considered high. This multi-dimensional metric collection and analysis mechanism enhances the accuracy and real-time nature of load assessment and can adapt to the dynamic changes in board load under different business scenarios.

[0029] Step 106: Using an intelligent adjustment algorithm, determine the idle rate threshold of each board based on the board type and service load status; the intelligent adjustment algorithm is an improved genetic algorithm combined with a simulated annealing algorithm.

[0030] In this embodiment, the board type of each board is determined based on its physical characteristics (such as size, weight, port, and heat dissipation structure) and / or electrical signal data. This avoids the bias of relying on a single criterion and ensures accurate classification of service boards, system boards, and new types of boards. Furthermore, the inclusion of new types of boards in the identification scope facilitates rapid adaptation during transmission equipment upgrades, eliminating the need for frequent adjustments to the identification logic and helping the transmission equipment cope with technological updates. Based on real-time load data (such as processor utilization, memory usage, and network bandwidth usage) of each board, the service load status of each board is determined. Then, using an intelligent adjustment algorithm, the idle rate threshold of each board is determined based on its board type and service load status. This intelligent adjustment algorithm combines an improved genetic algorithm with a simulated annealing algorithm. This overcomes the problem of fixed idle rate thresholds that cannot be dynamically adjusted. Moreover, the combination of the improved genetic algorithm and the simulated annealing algorithm can quickly find the optimal threshold in different scenarios, which is more scientific, faster-responding, and adaptable to complex and changing load environments than manual settings. As can be seen, this technical solution provides a reliable foundation for dynamic threshold adjustment through accurate board type identification, avoiding threshold setting errors caused by type misjudgment. Dynamic threshold adjustment maximizes the value of various boards, forming a closed loop of accurate identification, intelligent adjustment and efficient operation, thereby improving the overall performance, stability and resource utilization of transmission equipment.

[0031] In one implementation, determining the board type of each board based on its physical characteristics and / or electrical signal data (i.e., step 102) can be performed as follows: steps A11-A14, and / or steps A21-A23. Step A11: Collect the size, weight, port, and heat dissipation structure information of each board in the transmission device to obtain the physical characteristic information of each board.

[0032] In practical implementation, sensors can be used to measure the length, width, and thickness of the circuit board to obtain its size information. Weighing sensors can be used to obtain the board's weight information. The number and arrangement of the board's ports can be measured to obtain its port information. The size, shape, and location of the heatsinks on the board, as well as the number and type of fans, can be checked to obtain the board's heat dissipation structure information.

[0033] Step A12: For each board, convert the physical feature information into a board image and generate the input feature map corresponding to the board image.

[0034] The collected physical feature information is converted into a board image, and an input feature map X (containing height h, width w, and number of channels c, such as c=3 for a color image) is generated as the algorithm input.

[0035] Step A13: The input feature map is processed by a convolutional neural network to obtain the output features.

[0036] Optionally, the algorithm function for a convolutional neural network is: X is the input feature map, representing the pixel information of the board image after preprocessing. h, w, and c represent the height, width, and number of channels of the input feature map, respectively. For example, for a color board image, c=3 (three RGB channels), and h and w depend on the image size. By inputting the feature map X, the overall feature information of the board can be obtained, providing a basis for the output features and input parameter data for board type identification.

[0037] K is the convolution kernel, a filter used to extract features. h and w are the height and width of the convolution kernel, determining the receptive field size for each convolution operation, i.e., the local image region of interest. c is the number of output channels, representing the number of different extracted features. c is the same as the number of channels in the input feature map, ensuring that convolution operations can be performed on each channel. The size and shape of the convolution kernel can be designed according to the characteristics of the board image. For recognizing boards with intricate circuit patterns (such as high-end FPGA boards), a smaller convolution kernel (such as 3) can be used. 3) To extract detailed features such as circuit patterns; while for identifying the overall layout of the board and large components (such as power modules, large heat sinks, etc.), a larger convolution kernel (such as 7) can be used. 7) To obtain macroscopic features. By using the convolution kernel algorithm, the convolution operation is performed by sliding on the input feature map, which greatly reduces the number of parameters and can effectively extract board features for board classification.

[0038] Y is the output feature map, representing the element value in the i-th row, j-th column, and c-th channel of the output feature map. This element value is obtained by sliding the convolution kernel across the input feature map and performing a weighted sum of the elements within the covered area. It is used to extract local features from the board image, such as the edges of chips and the texture of pins on the board.

[0039] Step A14: Determine the board type based on the output characteristics.

[0040] Changes in the number of channels in the output feature map can serve as a medium for identifying new and old circuit boards. For example, if the number of channels is not in the dataset, or if a new combination of features is extracted, the circuit board can be defined as a new type of circuit board. If the number of channels in the output feature map Y matches the known circuit board features in the dataset, the circuit board can be determined to be the corresponding model circuit board.

[0041] Step A21: Collect the real-time power consumption value of each board in the transmission equipment, as well as the signal strength and frequency of the electrical signals generated by each board, to obtain the electrical signal data of each board.

[0042] In practice, current sensors or power meters can be used to measure the real-time power consumption of the boards. Service boards typically consume more power when processing tasks, while fan boards have relatively stable power consumption.

[0043] Different types of circuit boards may generate different electrical signals. By monitoring the signal strength and frequency of these electrical signals, it is possible to distinguish the types of circuit boards.

[0044] Step A22: For each board, match the electrical signal data with a pre-established electrical signal feature library to obtain the matching result. The electrical signal feature library includes multiple sets of statistical information, each set of statistical information including board type, real-time power consumption range, signal strength range, and frequency range.

[0045] Optionally, each set of statistical information may also include information such as the mean and standard deviation of signal strength. In specific implementations, if the real-time power consumption of the board, the signal strength and frequency of the generated electrical signal all fall within the range recorded in a set of statistical information, then it can be determined that the electrical signal data of the board matches the set of statistical information.

[0046] Step A23: Determine the board type based on the matching results.

[0047] If the electrical signal data of a board matches a set of statistical information, the board type can be determined based on the board type recorded in the statistical information.

[0048] In specific implementation, if step 102 is executed as steps A11-A14 and A21-A23 as described above, the board type can be obtained if the board type determined in step A14 and step A23 is consistent; otherwise, an error is reported to indicate to the network administrator that there is a problem with the board type identification.

[0049] In this embodiment, the type of each board in the transmission device is determined as the basis. Specifically, the hardware attributes of the boards are obtained by collecting physical feature information of each board, including but not limited to size, weight, port, and heat dissipation structure information. These physical features reflect the appearance and structural characteristics of the boards, which are then used to distinguish different types of boards, such as service boards, system boards, and new types of boards. Next, the collected physical feature information is converted into a board image as input data, and a convolutional neural network is used for feature extraction, exemplarily. This convolutional neural network uses multiple layers of convolutional kernels to perform weighted summation of local features on the board image, thereby extracting detailed features such as circuit patterns, chip edges, and pin textures, while also obtaining the overall layout of the board and the macroscopic features of large components. The size of the convolutional kernels can be adjusted according to the recognition requirements; for example, a smaller 3×3 convolutional kernel can be used to capture circuit patterns in detail, or a larger 7×7 convolutional kernel can be used to obtain overall structural information. Through the output feature map, different board types with varying degrees of complexity can be distinguished, further determining whether the board belongs to a new type of board or a known model.

[0050] Furthermore, in addition to physical feature-based identification methods, this embodiment also exemplarily combines electrical signal data for board type determination. An electrical signal feature library is established by real-time acquisition of power consumption values ​​and signal strength and frequency characteristics of electrical signals from each board. This feature library contains multiple sets of statistical information, each covering the power consumption range, signal strength range, and frequency range corresponding to the board type. The acquired data of the board to be identified is matched with the feature library. If the power strength and signal characteristics match the statistical range of a service board, it is initially determined to be a service board; if they match the characteristics of a fan board, it is determined to be a fan board. This method can quickly assist in type identification without the need for complex algorithms and is more suitable for board types with clearly distinguishable power and signal characteristics.

[0051] In one implementation, an intelligent adjustment algorithm is used to determine the idle rate threshold of each board based on its board type and service load status (i.e., step 106), which can be executed as follows: Steps B1-B3: Step B1: Determine the service load of the transmission equipment based on the service load status of each board.

[0052] In practice, the business busyness index can be calculated by weighting and integrating indicators such as processor utilization, memory usage, and network bandwidth usage, and then divided into high, medium, and low levels to obtain the business busyness level.

[0053] Step B2: Using an improved genetic algorithm combined with a simulated annealing algorithm, determine the idle rate threshold adjustment amount for each board based on the board type and the service load of the transmission equipment.

[0054] In practice, different idle rate threshold adjustment strategies can be adopted according to the level of business activity. For example, when business activity is high, the idle rate threshold can be lowered to improve resource utilization, while when business activity is low, the threshold can be raised to save energy and costs.

[0055] Step B3: Adjust the idle rate threshold of the corresponding board according to the idle rate threshold adjustment amount.

[0056] In this embodiment, the service load status is used not only to determine the real-time load of each board, but also to estimate the overall service busy level of the transmission equipment. Different idle rate threshold adjustment strategies are adopted based on the service busy level. The intelligent adjustment algorithm dynamically calculates the adjustment amount based on the real-time service busy level, ensuring that the idle rate threshold is highly matched with service demand, avoiding resource waste or performance degradation.

[0057] In one implementation, such as Figure 2 As shown, by using an improved genetic algorithm combined with a simulated annealing algorithm, the idle rate threshold adjustment amount for each board is determined based on the board type and the service load of the transmission equipment (i.e., step B2), which can be executed as follows: steps B21-B27: Step B21: Generate multiple sets of initial idle rate threshold combinations based on the historical service load status of each board to obtain the initial population.

[0058] Each initial idle rate threshold combination includes all board types and the initial idle rate threshold corresponding to each board type. Each initial idle rate threshold combination constitutes an individual in the initial population. For example, an individual may be represented as a vector [25%, 30%, 45%], corresponding to the initial idle rate thresholds of service boards, system boards, and new type boards, respectively.

[0059] Understandably, maintaining population diversity during the optimization process of genetic algorithms can prevent the algorithm from falling into a single solution pattern. For training the adjustment of the board idle rate threshold, board usage is complex and diverse, influenced by various factors such as fluctuations in business traffic, equipment failures, and peak business periods at different times. If the population lacks diversity, the algorithm may only focus on one or a few typical board usage scenarios, ignoring other possible situations. For example, with good population diversity, individuals can be included in various combinations of board port states such as high load, medium load, low load, and idle, thus more comprehensively exploring the various possibilities of board idle rate.

[0060] Step B22: Calculate the first fitness of each individual in the initial population based on the pre-constructed fitness function.

[0061] The fitness function is determined based on the initial fitness function and the traffic load of the transmission equipment. In practice, the fitness of each individual board can be evaluated based on the effectiveness of board auditing. For example, indicators such as the identification accuracy and resource utilization of idle boards can be considered. Therefore, the fitness function... Where T represents the idle rate threshold, , , Let be the weight coefficients of each item, and satisfy . . . For accuracy, For recall rate, The score is the F1 score.

[0062] Accuracy methods: ,in, This indicates that at a given idle rate threshold The number of board resources (such as ports, fiber optic cables, etc.) actually occupied and correctly identified as occupied during board auditing is the number of true positives. For example, if a board has 10 ports actually transmitting services, the idle rate threshold is used to determine the number of true positives. During the judgment process, all 10 ports were determined to be in use. The value increases by 10. This refers to the idle rate threshold. The number of board resources that are actually idle but are mistakenly identified as occupied, i.e., the number of false positives. For example, if 5 free ports are judged to be in use, then... The value is 5. Accuracy measures the proportion of board resources that are judged to be occupied but are actually occupied, reflecting the accuracy of the audit judgment, that is, minimizing the situation where idle resources are mistakenly judged as occupied.

[0063] For example, when the business activity level is "high", the fitness function will focus on resource utilization (weight increased to 60%), penalizing threshold schemes with excessively high idle rates. When the business activity level is "low", the fitness function will focus more on energy consumption reduction (weight increased to 50%), prioritizing high idle rate thresholds.

[0064] Step B23: Based on the first fitness, perform selection, crossover, and mutation operations on the initial population using a genetic algorithm to generate the offspring population.

[0065] The selection operation involves using methods such as roulette wheel selection or tournament selection to choose individuals with high fitness from the current population as parents. The crossover operation uses methods such as single-point crossover or multi-point crossover to perform crossover on parent individuals to generate offspring. The mutation operation performs mutation on offspring individuals to increase population diversity.

[0066] Step B24: Select an individual from the offspring population as the initial solution for the simulated annealing algorithm; and determine the iterative environment data for each individual, including the annealing temperature and cooling rate.

[0067] Step B25: In each iteration of the simulated annealing algorithm, the current solution is randomly perturbed to obtain a new solution; the second fitness of the new solution is calculated according to the fitness function; based on the fitness relationship between the new solution and the current solution, it is determined whether the new solution satisfies the iteration termination condition; if yes, then proceed to step B26; if no, then proceed to step B27.

[0068] The iteration termination condition may include one or more of the following: reaching the maximum number of iterations, the fitness reaching a certain value, or the temperature decreasing to a certain level.

[0069] Step B26: Stop the iteration and determine the idle rate threshold adjustment amount of the board based on the new solution.

[0070] Step B27: Continue to update the iterative environment data and new solutions, and determine the idle rate threshold adjustment amount of the board based on the updated iterative environment data and new solutions, until the iteration termination condition is met.

[0071] The temperature parameter of the simulated annealing algorithm is incorporated into the genetic algorithm's operation. After crossover and mutation operations, the temperature parameter is used to control the acceptance probability of new individuals. For example, after crossover, the fitness difference between offspring and parents is calculated, and then the probability of accepting offspring is calculated based on the Metropolis criterion according to temperature. Similarly, in mutation, temperature is used to determine whether to accept mutated individuals. This approach allows the genetic algorithm's operation to be influenced by simulated annealing, thereby adjusting the search strategy.

[0072] In practical implementation, when combining genetic algorithms and simulated annealing algorithms, the two can share the same fitness function. This fitness function is defined based on the board audit results, taking into account factors such as audit precision, recall, and the accuracy of idle rate assessment. By sharing the fitness function, both algorithms use the same standard to measure the quality of individuals. Thus, during algorithm execution, both the selection operation in the genetic algorithm and the acceptance criterion in the simulated annealing algorithm can be judged based on this unified fitness function.

[0073] In this embodiment, the genetic algorithm, by simulating natural selection and genetic mutation, can quickly explore a large solution space, generating diverse solutions and avoiding getting trapped in local optima. However, the decrease in diversity in the later stages of the genetic algorithm can easily lead to local convergence. The simulated annealing algorithm, based on the principle of metal annealing, accepts inferior solutions with a certain probability and performs a fine search in local regions, avoiding getting trapped in local minima. When the two are combined, the genetic algorithm provides diversified initial solutions for the simulated annealing algorithm, broadening the search range, while the simulated annealing algorithm uses its local search capabilities to refine the solutions, ultimately achieving both global and local optimization effects. Secondly, the combined algorithm introduces an adaptive parameter adjustment mechanism, dynamically adjusting the crossover and mutation probabilities of the genetic algorithm based on the annealing temperature. This encourages diverse exploration at high temperatures and promotes solution space convergence at low temperatures, improving overall optimization efficiency. Furthermore, this combined algorithm, when applied in complex systems such as network management networks, can significantly reduce the number of misjudgments, reduce resource waste, and improve audit accuracy and resource utilization.

[0074] As an example, a network management system contains numerous audit cards used to monitor compliance in critical business processes such as network traffic and data storage. With dynamic changes in business volume, the idle rate of these cards fluctuates constantly, requiring real-time adjustments to the idle rate threshold to accurately determine whether cards are idle, while simultaneously optimizing audit strategies.

[0075] Initial stage: A genetic algorithm is used to generate an initial set of idle rate threshold combinations. Assuming the data center has 100 different types of audit boards, the genetic algorithm, based on analysis of historical data (such as board load at different times over the past week), generates 50 different idle rate threshold vectors, each corresponding to a board type. These threshold vectors constitute the initial population.

[0076] Meanwhile, parameters such as the initial temperature and cooling rate of the annealing algorithm are set. The initial temperature is set to 100, and the cooling rate is 0.95, which means that as the annealing process progresses, the probability of the algorithm accepting inferior solutions will gradually decrease, tending towards the search for local optima.

[0077] Iterative optimization phase: For each combination of idle rate thresholds generated by the genetic algorithm, annealing is used for local optimization. Taking a board responsible for network traffic auditing as an example, its initial idle rate threshold is set to 30%. Based on this, the annealing algorithm attempts to adjust the threshold with a certain probability, according to the current temperature and objective function (such as a balance function between audit accuracy and resource waste). If the adjusted threshold brings better audit results (such as reducing false positives and improving sensitivity to abnormal traffic), the new threshold is accepted; otherwise, according to the Metropolis criterion, a suboptimal solution is accepted with a certain probability, and the search continues.

[0078] In this process, the genetic algorithm continuously updates the population through selection, crossover, and mutation operations based on the optimization results fed back by the annealing algorithm. For example, if a certain threshold combination performs well in auditing performance after optimization by the annealing algorithm, its corresponding individuals have a higher probability of being selected in the selection phase of the genetic algorithm to participate in the reproduction of the next generation and pass on the superior genes.

[0079] Convergence and Decision-Making Phase: After multiple iterations, the algorithm gradually converged. Finally, an optimized set of idle rate threshold vectors was obtained and applied to the actual auditing system.

[0080] Actual results show that before adjusting the idle rate threshold using the combined algorithm, the audit system suffered from an average of 200 false positives per month (misclassifying normally functioning boards as idle or vice versa) due to unreasonable fixed threshold settings, and the resource waste rate reached 15% (i.e., the percentage of time idle boards were not used effectively). After applying the combined algorithm, the number of false positives decreased to 50 per month, the resource waste rate decreased to 5%, and both audit accuracy and resource utilization were significantly improved.

[0081] In one implementation, such as Figure 3 As shown, the idle rate of the board can be monitored through the following steps 301-306 to identify idle boards in a timely manner.

[0082] Step 301: Obtain the port bearer information of each board in the transmission device.

[0083] In practice, the port bearer information of each board in the transmission equipment can be obtained from the Network Management System (NMS).

[0084] Step 302: Select the target board from each board based on the port carrying information.

[0085] The target boards are those whose idle rate needs to be determined. In practice, a dataset of boards that do not require idle rate determination can be identified based on port carrying information. This dataset is then cleaned to remove boards that do not require idle rate determination, yielding the target boards. The dataset of boards that do not require idle rate determination can include portless boards and primary / backup boards. Portless boards (such as some power boards and management boards) do not have data transmission ports; their functionality does not rely on ports for service interaction, therefore, idle rate determination through port status is unnecessary. Primary / backup boards (such as primary and backup boards) employ a redundancy design. Backup boards are typically in standby mode, their idle state controlled by the system redundancy mechanism, not by regular service load.

[0086] Step 303: For the target board, iterate through all ports of the target board, and count the number of ports in low occupancy or unoccupancy state according to the preset service traffic threshold to obtain port idle information; detect the occupancy status of each optical fiber connected to the target board to obtain optical fiber idle information.

[0087] Step 304: Obtain comprehensive idle information of the target board by simulating the hardware interface.

[0088] The comprehensive idle information is used to characterize the current resource idle status of the target board. In specific implementations, the current load value and idle rate of the board can be obtained by simulating the hardware interface. The load value is generated by random numbers, ranging from 0% to 100%, to simulate the actual workload of the board. The idle rate is obtained by subtracting the current load value from 100%, representing the current resource idle status of the board.

[0089] Step 305: Determine the first idle rate of the target board based on the port idle information, fiber idle information, and comprehensive idle information.

[0090] In practice, the first idle rate can be output in the form of a report, including information such as the target board type, port idle rate, fiber optic idle rate, and overall idle rate.

[0091] Step 306: Monitor the first idle rate based on the idle rate threshold of the target board.

[0092] In practice, when the first idle rate of the target board exceeds the idle rate threshold, an alarm can be triggered to notify network administrators so that timely measures can be taken to optimize and adjust resources.

[0093] In addition, a scheduled task can be set up to repeat the above data collection and idle rate calculation steps at certain time intervals (such as daily or weekly) to track changes in the utilization of board resources in real time.

[0094] In this embodiment, fine-grained port and fiber optic load monitoring, combined with simulated hardware interface data, enables comprehensive awareness of board resource idleness. Based on dynamically adjusted idle rate thresholds, real-time control of resource utilization efficiency is achieved, allowing for timely detection of idle resources and potential waste. This provides precise data for system optimization, improving the overall operating efficiency and stability of the transmission equipment.

[0095] In one implementation, the board resource management method provided in this application embodiment can be implemented by training a large model.

[0096] In the initial stages of building a large model, the dataset can be divided into training and test sets. The model is trained using the training set and its performance is validated using the test set. Appropriate evaluation metrics, such as accuracy, precision, and recall, are then used to assess the model.

[0097] Key factors to consider when selecting an AI (Artificial Intelligence) decision tree model include: Factor 1: Model Type. Decision tree models suitable for idle rate threshold problems can be selected, such as CART (Classification and Regression Trees), RF (Random Forests), or GBDT (Gradient Boosting Decision Trees). These models can adapt to non-linear relationships and complex feature interactions.

[0098] Factor 2: Algorithm Performance. Considering the model's performance, robustness, and computational efficiency, XGBoost (eXtremeGradient Boosting) and LightGBM (Light Gradient Boosting Machine) are high-performing models that perform well with large-scale data and high-dimensional features.

[0099] Factor 3: Interpretability. Specifically, this requires considering both business needs and the interpretability of the decision tree model. Some scenarios may require a deeper understanding of the model's decision-making process; in such cases, models with better interpretability, such as AI decision trees, can be chosen.

[0100] This embodiment employs multiple metrics to evaluate the model, including accuracy, precision, recall, and F1 score. A comprehensive analysis of model performance is conducted using confusion matrices and visualization tools such as ROC (Receiver Operating Characteristic Curve) and PR (Precision-Recall Curve). The F1 score, a statistical metric used to evaluate the performance of binary classification models, ranges from 0 to 1 and is calculated as the harmonic mean of precision and recall. To ensure model interpretability, the system uses SHAP (SHapley Additive exPlanations) values ​​(a game-theory-based method for model interpretability that quantifies the contribution of each feature to model prediction) and feature importance analysis to help understand the model's decision-making process. Model optimization is an ongoing process, with training data updated periodically, and automated tools ensure efficient and consistent model updates. A feedback mechanism collects user feedback to promptly identify and address performance degradation issues. Model monitoring ensures stability and reliability in real-time environments, supporting the system's real-time response to data sources such as sensors.

[0101] This application provides a distributed data processing system based on the Hadoop distributed architecture. It improves the Hive architecture by separating the Driver and Hive Client, and combining the Driver and Hive Service, to implement the various functions of transmission device board resource management mentioned above. By dividing and optimizing component responsibilities, the system ensures high efficiency in data processing and system stability, supporting dynamic adjustment and accurate identification of resource management methods.

[0102] like Figure 4As shown, specifically, this distributed data processing system includes core Hadoop components such as HDFS (Hadoop Distributed File System), MapReduce (a parallel computing model used for massive data computation, processing data through mapping and reduction phases), and YARN (Yet Another Resource Negotiator, responsible for cluster resource scheduling and supporting various workloads) as the underlying data storage and computing framework. Based on this, the system improves the Hive (a Hadoop-based data warehouse tool) architecture by decoupling the traditional Hive Driver module from the Hive Client, forming independent query execution modules (Driver) and user interaction modules (Hive Client). The Driver module handles query processing tasks such as syntax parsing, logical plan generation, physical plan generation, and execution, while the Hive Client receives user query requests and displays query results. To achieve this separation between the Driver and Hive Client, the system defines abstract interfaces such as IQueryExecutor and IQueryClient, used for query execution and sending / receiving query requests and results, respectively. This decoupling design enhances the system's modularity and maintainability.

[0103] As you can understand, Hive is a data warehouse tool built on top of Hadoop, primarily used for processing and analyzing large-scale structured data. Its core services include HiveServer2 and Hive Metastore. The Hive Client is the entry point for users to interact with Hive, responsible for receiving user query requests and sending them to the Hive server. Common Client types include: CLI (Command Line Interface): The most commonly used command-line interface. Starting the CLI also starts a copy of Hive (local mode). JDBC / ODBC Client: Allows applications such as Java and Python to remotely access Hive through standard database connection protocols; requires a connection to HiveServer2. WebUI (Hive Web Interface): A simple graphical interface for accessing Hive through a browser. In short, the Client is the "front end," used for entering commands and viewing results.

[0104] Furthermore, the system integrates the Driver with Hive Service (a set of service components provided by Apache Hive for accessing, managing, and operating distributed data warehouses). It dynamically obtains Hive Service node information using service discovery mechanisms (such as integrating a Zookeeper client) and establishes a secure connection through a network communication library, completing protocol handshakes and authentication. After preprocessing user requests, the Driver module sends query requests to the Hive Service for metadata management, query optimization, and execution. Once processed, the Hive Service returns the results to the Driver. To ensure metadata consistency and real-time performance, the system employs an event-driven mechanism. The Hive Service registers metadata update event listeners and pushes metadata change events via message queues (such as Kafka). The Driver receives these events and synchronously updates its metadata cache. In addition, the system improves the Hive architecture, separating the Driver from the Hive Client and integrating it with the Hive Service to form the Hive Service. The Hive Client can use the Hive Service without requiring a full Hive installation, and it is compatible with both Hive CLI and HWI operation modes, significantly improving algorithm execution efficiency and system performance.

[0105] In the Hadoop ecosystem, Hive CLI and HWI are two different types of client interfaces provided by Hive for interacting with Hive services.

[0106] Specifically, the Hive CLI (Command Line Interface) is the command-line tool included with Hive, allowing users to directly enter HiveQL statements for data querying and management through the terminal or shell environment. Upon startup, it creates a local copy of Hive (i.e., embedded mode), suitable for standalone testing or development environments. It is suitable for users familiar with command-line operations and is often used for quickly executing queries and debugging scripts. In newer versions of Hive, the CLI has been gradually replaced by Beeline (a JDBC-based client for connecting to HiveServer2) because Beeline supports more secure and efficient remote connections.

[0107] HWI (Hive Web Interface) is a web-based graphical client that allows users to access Hive services and execute HiveQL queries through a browser. The HWI service needs to be started separately (default port 9999), and then accessed through a browser. It is suitable for users without command-line access or who prefer a graphical interface, and is primarily used in development and testing environments.

[0108] Through the above technical solutions, this distributed data processing system can efficiently process large amounts of board usage and business load data, supporting the training and inference process of AI decision tree models and ensuring the smooth implementation of data collection, preprocessing, model training, and real-time dynamic adjustment steps in resource management methods. The system's modular design and distributed architecture not only improve data processing speed but also enhance system stability and scalability, meeting the needs of energy saving and resource optimization for transmission equipment boards in large-scale network environments.

[0109] The distributed data processing system described in this embodiment can significantly improve the real-time monitoring and intelligent management capabilities of transmission equipment board resource status. Through efficient data processing and a flexible module communication mechanism, the system ensures the accuracy and real-time performance of dynamically adjusting idle rate thresholds and board type identification in resource management methods. This effectively reduces resource waste, minimizes misjudgments, improves the overall performance and stability of the system, and brings significant economic and social benefits.

[0110] Furthermore, the system's decoupled design allows the Driver module to run independently from the Hive Client and Hive Service, facilitating maintenance and upgrades, and supporting optimized query execution efficiency without affecting user interaction. Simultaneously, through service discovery and connection management, the system possesses strong adaptability, capable of handling dynamic node changes and improving system availability and fault tolerance. An event-driven metadata synchronization mechanism ensures data consistency and supports immediate responses to board status changes in resource management methods. Moreover, the design's compatibility with multiple operation interfaces reduces the complexity of system deployment and maintenance, making algorithm optimization and data processing more efficient and reliable.

[0111] It should be noted that the board resource management method provided in this application embodiment can be executed by a board resource management device or a control module within that board resource management device for executing the board resource management method. This application embodiment uses the execution of the board resource management method by a board resource management device as an example to illustrate the board resource management device provided in this application embodiment.

[0112] Figure 5 This is a schematic diagram of the structure of a board resource management device provided in an embodiment of this application. Figure 5 As shown, the board resource management device includes: a first determination module 510, a second determination module 520, and a third determination module 530.

[0113] The first determining module 510 is used to determine the board type of each board based on the physical characteristic information and / or electrical signal data of each board in the transmission equipment; the physical characteristic information includes at least one of the following: size information, weight information, port information, and heat dissipation structure information; the board type includes at least one of the following: service board, system board, and new type board; the second determining module 520 is used to determine the service load status of each board based on the real-time load data of each board monitored in real time; the real-time load data includes at least one of the following: processor utilization, memory occupancy, and network bandwidth occupancy; the third determining module 530 is used to determine the idle rate threshold of each board based on the board type and service load status of each board using an intelligent adjustment algorithm; the intelligent adjustment algorithm is an improved genetic algorithm combined with a simulated annealing algorithm.

[0114] In one implementation, the first determining module 510 includes: a first acquisition and determining unit and / or a second acquisition and determining unit.

[0115] The first acquisition and determination unit is used to acquire the size, weight, port, and heat dissipation structure information of each board in the transmission device to obtain the physical feature information of each board; for each board, the physical feature information is converted into a board image, and an input feature map corresponding to the board image is generated; the input feature map is processed by a convolutional neural network to obtain the output features; and the board type is determined based on the output features. And / or, The second acquisition and determination unit is used to acquire the real-time power consumption value of each board in the transmission equipment, as well as the signal strength and frequency of the electrical signals generated by each board, to obtain the electrical signal data of each board; for each board, the electrical signal data is matched with a pre-established electrical signal feature library to obtain the matching result; based on the matching result, the board type is determined; the electrical signal feature library includes multiple sets of statistical information, each set of statistical information including board type, real-time power consumption range, signal strength range, and frequency range.

[0116] In one implementation, the third determining module 530 includes: a first determining unit, a second determining unit, and an adjustment unit.

[0117] The first determining unit is used to determine the service load of the transmission equipment based on the service load status of each board; the second determining unit is used to determine the idle rate threshold adjustment amount of each board based on the board type and the service load of the transmission equipment by using an improved genetic algorithm combined with a simulated annealing algorithm; the adjusting unit is used to adjust the idle rate threshold of the corresponding board based on the idle rate threshold adjustment amount.

[0118] In one implementation, the second determining unit is specifically used to: generate multiple sets of initial idle rate threshold combinations based on the historical service load status of each board, to obtain an initial population; each set of initial idle rate threshold combinations includes all board types and the initial idle rate threshold corresponding to each board type; wherein each set of initial idle rate threshold combinations constitutes an individual in the initial population; calculate the first fitness of each individual in the initial population according to a pre-constructed fitness function; the fitness function is determined based on the initial fitness function and the service busyness of the transmission equipment; perform selection, crossover, and mutation operations on the initial population using a genetic algorithm based on the first fitness to generate a offspring population; and select from the offspring population... An individual is used as the initial solution for the simulated annealing algorithm; and the iterative environment data for each individual is determined, including the annealing temperature and cooling rate; in each iteration of the simulated annealing algorithm, the current solution is randomly perturbed to obtain a new solution; the second fitness of the new solution is calculated according to the fitness function; based on the fitness relationship between the new solution and the current solution, it is determined whether the new solution meets the iteration termination condition; if so, the iteration stops, and the idle rate threshold adjustment amount of the board is determined based on the new solution; if not, the iterative environment data and the new solution are updated, and the idle rate threshold adjustment amount of the board is determined based on the updated iterative environment data and the new solution, until the iteration termination condition is met.

[0119] In one implementation, the board resource management device further includes: a first acquisition module, a filtering module, a statistics and detection module, a second acquisition module, a fourth determination module, and an idle rate monitoring module.

[0120] The first acquisition module is used to acquire the port bearer information of each board in the transmission equipment; the filtering module is used to filter the target board from each board according to the port bearer information; the target board is the board whose idle rate needs to be determined; the statistics and detection module is used to traverse all ports of the target board, and according to the preset service traffic threshold, count the number of ports in a low-occupancy or unoccupancy state to obtain port idle information; and detect the occupancy status of each optical fiber connected to the target board to obtain optical fiber idle information; the second acquisition module is used to acquire the comprehensive idle information of the target board through a simulated hardware interface; the comprehensive idle information is used to characterize the current resource idle status of the target board; the fourth determination module is used to determine the first idle rate of the target board according to the port idle information, optical fiber idle information and comprehensive idle information; the idle rate monitoring module is used to monitor the first idle rate based on the idle rate threshold of the target board.

[0121] In this embodiment, the board type of each board is determined based on its physical characteristics (such as size, weight, port, and heat dissipation structure) and / or electrical signal data. This avoids the bias of relying on a single criterion and ensures accurate classification of service boards, system boards, and new types of boards. Furthermore, the inclusion of new types of boards in the identification scope facilitates rapid adaptation during transmission equipment upgrades, eliminating the need for frequent adjustments to the identification logic and helping the transmission equipment cope with technological updates. Based on real-time load data (such as processor utilization, memory usage, and network bandwidth usage) of each board, the service load status of each board is determined. Then, using an intelligent adjustment algorithm, the idle rate threshold of each board is determined based on its board type and service load status. This intelligent adjustment algorithm combines an improved genetic algorithm with a simulated annealing algorithm. This overcomes the problem of fixed idle rate thresholds that cannot be dynamically adjusted. Moreover, the combination of the improved genetic algorithm and the simulated annealing algorithm can quickly find the optimal threshold in different scenarios, which is more scientific, faster-responding, and adaptable to complex and changing load environments than manual settings. As can be seen, this technical solution provides a reliable foundation for dynamic threshold adjustment through accurate board type identification, avoiding threshold setting errors caused by type misjudgment. Dynamic threshold adjustment maximizes the value of various boards, forming a closed loop of accurate identification, intelligent adjustment and efficient operation, thereby improving the overall performance, stability and resource utilization of transmission equipment.

[0122] The board resource management device in this application embodiment can be a device, or a component, integrated circuit, or chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. For example, mobile electronic devices can be mobile phones, tablets, laptops, PDAs, in-vehicle electronic devices, wearable devices, ultra-mobile personal computers (UMPCs), netbooks, or personal digital assistants (PDAs), etc., while non-mobile electronic devices can be servers, network attached storage (NAS), personal computers (PCs), televisions (TVs), ATMs, or self-service machines, etc. This application embodiment does not impose specific limitations.

[0123] The board resource management device in this application embodiment can be a device with an operating system. This operating system can be Android, iOS, or other possible operating systems; this application embodiment does not specifically limit the specific operating system used.

[0124] The board resource management device provided in this application embodiment can achieve... Figures 1 to 3 The various processes implemented in the method embodiments are not described in detail here to avoid repetition.

[0125] Based on the same technical concept, embodiments of this application also provide an electronic device for executing the above-described board resource management method. Figure 6 This is a schematic diagram of the structure of an electronic device to implement various embodiments of this application. The electronic device can vary significantly due to differences in configuration or performance, and may include a processor 610, a communications interface 620, a memory 630, and a communication bus 640. The processor 610, communications interface 620, and memory 630 communicate with each other via the communication bus 640. The processor 610 can call a computer program stored in the memory 630 and executable on the processor 610 to perform the following steps: Based on the physical characteristics and / or electrical signal data of each board in the transmission equipment, determine the board type of each board; the physical characteristics include at least one of the following: size information, weight information, port information, and heat dissipation structure information; the board type includes at least one of the following: service board, system board, and new type board; based on the real-time load data of each board monitored in real time, determine the service load status of each board; the real-time load data includes at least one of the following: processor utilization, memory utilization, and network bandwidth utilization; using an intelligent adjustment algorithm, determine the idle rate threshold of each board based on the board type and service load status of each board; the intelligent adjustment algorithm is an improved genetic algorithm combined with a simulated annealing algorithm.

[0126] In this embodiment, the board type of each board is determined based on its physical characteristics (such as size, weight, port, and heat dissipation structure) and / or electrical signal data. This avoids the bias of relying on a single criterion and ensures accurate classification of service boards, system boards, and new types of boards. Furthermore, the inclusion of new types of boards in the identification scope facilitates rapid adaptation during transmission equipment upgrades, eliminating the need for frequent adjustments to the identification logic and helping the transmission equipment cope with technological updates. Based on real-time load data (such as processor utilization, memory usage, and network bandwidth usage) of each board, the service load status of each board is determined. Then, using an intelligent adjustment algorithm, the idle rate threshold of each board is determined based on its board type and service load status. This intelligent adjustment algorithm combines an improved genetic algorithm with a simulated annealing algorithm. This overcomes the problem of fixed idle rate thresholds that cannot be dynamically adjusted. Moreover, the combination of the improved genetic algorithm and the simulated annealing algorithm can quickly find the optimal threshold in different scenarios, which is more scientific, faster-responding, and adaptable to complex and changing load environments than manual settings. As can be seen, this technical solution provides a reliable foundation for dynamic threshold adjustment through accurate board type identification, avoiding threshold setting errors caused by type misjudgment. Dynamic threshold adjustment maximizes the value of various boards, forming a closed loop of accurate identification, intelligent adjustment and efficient operation, thereby improving the overall performance, stability and resource utilization of transmission equipment.

[0127] The specific execution steps can be found in the various steps of the above-described board resource management method embodiment, and can achieve the same technical effect. To avoid repetition, they will not be repeated here.

[0128] It should be noted that the electronic devices in the embodiments of this application include: servers, terminals, or other devices besides terminals.

[0129] The above electronic device structure does not constitute a limitation on the electronic device. An electronic device may include more or fewer components than illustrated, or combine certain components, or arrange them differently. For example, an input unit may include a Graphics Processing Unit (GPU) and a microphone, and a display unit may use a liquid crystal display (LCD), organic light-emitting diode (OLED), or other similar display panels. User input units include at least one of a touch panel and other input devices. A touch panel is also called a touchscreen. Other input devices may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, power buttons, etc.), trackballs, mice, and joysticks, which will not be elaborated further here.

[0130] Memory can be used to store software programs and various data. Memory can primarily include a first storage area for storing programs or instructions and a second storage area for storing data. The first storage area can store the operating system, application programs or instructions required for at least one function (such as sound playback, image playback, etc.). Furthermore, memory can include volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (Synchlink DRAM, SLDRAM), and direct memory bus RAM (DRRAM).

[0131] The processor may include one or more processing units; optionally, the processor integrates an application processor and a modem processor, wherein the application processor mainly handles operations related to the operating system, user interface, and applications, while the modem processor mainly handles wireless communication signals, such as a baseband processor. It is understood that the aforementioned modem processor may also not be integrated into the processor.

[0132] This application also provides a computer-readable storage medium for storing computer-executable instructions. When the computer-executable instructions are executed by a processor, they implement the various processes of the above-described board resource management method embodiments and achieve the same technical effects. To avoid repetition, they will not be described again here.

[0133] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0134] This application also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the various processes of the above-described board resource management method embodiments and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0135] 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 board resource management method embodiments and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0136] 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.

[0137] It should be understood that the training and prediction processes of the AI ​​models involved in the various embodiments of this application's specification all adhere to multiple legal and compliant principles, including legal data sources, compliant data content, compliant data governance, compliant training objectives and schemes, compliant training processes, compliant training environments and tools, and compliant ethical verification of training results, thus meeting the requirements of Article 5 of the Patent Law. Among them: Data source legitimacy: All datasets used for AI model training were obtained through legal means, covering three categories: publicly authorized data, data authorized by partners, and self-collected compliant data. Publicly authorized data originates from compliant data sources following open-source licenses such as Apache 2.0, with complete copyright attribution and authorization scope clearly marked, and contains no unauthorized open-source code or data reuse. Data authorized by partners has been subject to formal data usage agreements, clearly defining the scope, duration, and confidentiality obligations, and possessing a complete authorization chain. For self-collected data involving personal information, strict informed consent procedures have been followed, and anonymization processes (including but not limited to field masking, feature anonymization, and differential privacy technology applications) have been implemented to remove personally identifiable information, fully complying with the requirements of relevant laws and regulations such as the "Interim Measures for the Administration of Generative Artificial Intelligence Services" and the "Personal Information Protection Law."

[0138] Data content compliance: The AI ​​model's dataset undergoes multiple screenings and cleaning processes to remove all content that may violate social morality or harm public interests. It contains no obscene, pornographic, violent, discriminatory, or information that endangers national or public safety, nor does it involve the illegal acquisition or use of genetic resources. For data in sensitive fields (such as healthcare and finance), an additional privacy-preserving computation module (including federated learning and secure multi-party computation technologies) ensures that the data is "usable but not visible," avoiding compliance risks during the original data transmission process and ensuring that the data application scenarios and uses comply with public order and good morals and industry regulatory requirements.

[0139] Data governance norms: A complete data traceability system is established during the AI ​​model training process to automatically record the source, collection time, annotation process, cleaning rules, and permission allocation of training data, generating traceable compliance reports to ensure that the data is verifiable throughout its entire lifecycle. The dataset annotation process for AI models is completed by a professional human R&D team, clearly defining the proportion of human creative contributions and avoiding reliance on AI-generated data that has not undergone substantial human modification, thus meeting the examination requirements for "human main contributions" in AI patent applications.

[0140] Training objectives and plans are compliant: The AI ​​model training objective focuses on intelligent voice interaction. The training scheme and the final output results do not violate any mandatory provisions of laws and administrative regulations, do not harm the public interest or the legitimate rights and interests of others, and do not pose any potential risks of being used for illegal activities, infringing on privacy, or disrupting public safety. The model strictly adheres to the ethical principle of "intelligent for good".

[0141] Training process compliance: A closed-loop training framework is adopted to ensure compliance and controllability of the training process. The specific process is as follows: First, training samples are obtained through compliant data sources. After the aforementioned data cleaning and desensitization, they are input into the neural network model to generate preliminary training results. Second, an expert system is introduced to verify the preliminary results. Based on preset rules and human expert experience, the feasibility of the results is evaluated, and outputs that may pose ethical risks or compliance hazards are corrected (such as removing decision-making logic that violates public order and good morals, and adjusting model parameters that do not comply with safety regulations). Finally, the loss function weights are dynamically optimized based on expert system feedback to strengthen the model's learning of compliant results, avoid overfitting errors or non-compliant labels, and form a closed-loop control of "data input - model training - expert verification - parameter optimization - result feedback" to ensure that the entire training process complies with A5 ethical review requirements.

[0142] Training environment and tool compliance: AI model training is implemented using nationally licensed chips and a compliant training platform. All open-source frameworks and components used in the training process have obtained their corresponding licenses, and copyright statements and patent citation information are fully retained, with no instances of infringement or reuse. The training environment is built using virtual devices (containers / virtual machines) with fixed random seeds and initial parameter configurations to ensure the reproducibility of the training process. Furthermore, through access control and operation log recording, risks such as data leakage and parameter tampering during training are prevented, ensuring the security and compliance of the training process.

[0143] Training results ethical verification compliance: After the model is trained, it undergoes additional third-party ethical compliance assessment and algorithm filing review to verify that the model output does not violate social morality or harm public interests. For potentially sensitive scenarios (such as public services and intelligent decision-making), a special result verification mechanism is established to ensure that the model always complies with Article 5 of the Patent Law and relevant laws and regulations in practical applications.

[0144] In summary, the data and training process used in the AI ​​model of this application strictly comply with the relevant provisions of Article 5 of the Patent Law and the Patent Examination Guidelines (2023 Edition), and there is no violation of laws, social ethics, public interests, or illegal use of genetic resources. Therefore, it fully meets the compliance requirements for patent authorization.

[0145] 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.

[0146] 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 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, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0147] 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. A method for managing board resources, characterized in that, include: Based on the physical characteristics and / or electrical signal data of each board in the transmission equipment, determine the board type of each board; The physical characteristic information includes at least one of the following: size information, weight information, port information, and heat dissipation structure information; the board type includes at least one of the following: service board, system board, and new type board; Based on the real-time load data of each board, the service load status of each board is determined; the real-time load data includes at least one of the following: processor utilization, memory usage, and network bandwidth usage. Using an intelligent adjustment algorithm, the idle rate threshold of each board is determined based on the board type and the service load status of each board; the intelligent adjustment algorithm is an improved genetic algorithm combined with a simulated annealing algorithm.

2. The method according to claim 1, characterized in that, The step of determining the board type of each board based on the physical characteristic information and / or electrical signal data of each board in the transmission device includes: The system collects the size, weight, port, and heat dissipation structure information of each board in the transmission device to obtain the physical feature information of each board; for each board, the physical feature information is converted into a board image, and an input feature map corresponding to the board image is generated; the input feature map is processed by a convolutional neural network to obtain output features; and the board type of the board is determined based on the output features. And / or, The real-time power consumption values ​​of each board in the transmission device, as well as the signal strength and frequency of the electrical signals generated by each board, are collected to obtain electrical signal data for each board. For each board, the electrical signal data is matched with a pre-established electrical signal feature library to obtain a matching result. Based on the matching result, the board type of the board is determined. The electrical signal feature library includes multiple sets of statistical information, each set of statistical information including board type, real-time power consumption range, signal strength range, and frequency range.

3. The method according to claim 1, characterized in that, The method of using an intelligent adjustment algorithm to determine the idle rate threshold of each board based on the board type and the service load status includes: The service load status of the transmission device is determined based on the service load status of each of the aforementioned boards; By using an improved genetic algorithm combined with a simulated annealing algorithm, the idle rate threshold adjustment amount of each board is determined based on the board type and the service load of the transmission equipment. Adjust the idle rate threshold of the corresponding board according to the idle rate threshold adjustment amount.

4. The method according to claim 3, characterized in that, The method of using an improved genetic algorithm combined with a simulated annealing algorithm to determine the idle rate threshold adjustment amount for each board based on the board type and the service load of the transmission equipment includes: Multiple initial idle rate threshold combinations are generated based on the historical service load status of each board to obtain an initial population; each initial idle rate threshold combination includes all board types and the initial idle rate threshold corresponding to each board type; wherein, each initial idle rate threshold combination constitutes an individual in the initial population; The first fitness of each individual in the initial population is calculated based on a pre-constructed fitness function; the fitness function is determined based on the initial fitness function and the traffic load of the transmission device. Based on the first fitness, the initial population is subjected to selection, crossover, and mutation operations using a genetic algorithm to generate a offspring population; Select an individual from the offspring population as the initial solution for the simulated annealing algorithm; and determine the iterative environment data for each individual, the iterative environment data including annealing temperature and cooling rate; In each iteration of the simulated annealing algorithm, the current solution is randomly perturbed to obtain a new solution; the second fitness of the new solution is calculated according to the fitness function; and the fitness relationship between the new solution and the current solution is used to determine whether the new solution satisfies the iteration termination condition. If yes, then stop the iteration and determine the idle rate threshold adjustment amount of the board based on the new solution; if no, then continue to update the iteration environment data and the new solution, and determine the idle rate threshold adjustment amount of the board based on the updated iteration environment data and the new solution, until the iteration termination condition is met.

5. The method according to claim 1, characterized in that, The method further includes: Obtain the port bearer information of each board in the transmission device; Based on the port carrying information, target boards are selected from each of the boards; the target boards are those whose idle rate needs to be determined. For the target board, all ports of the target board are traversed, and the number of ports in a low-occupancy or unoccupancy state is counted according to a pre-set service traffic threshold to obtain port idle information; the occupancy status of each optical fiber connected to the target board is detected to obtain optical fiber idle information. By simulating the hardware interface, the comprehensive idle information of the target board is obtained; the comprehensive idle information is used to characterize the current resource idle status of the target board. Based on the port idle information, the fiber idle information, and the comprehensive idle information, determine the first idle rate of the target board; The first idle rate is monitored based on the idle rate threshold of the target board.

6. A distributed data processing system, characterized in that, The system is based on a Hadoop distributed architecture, and achieves separation of Driver and Hive Client and combination of Driver and Hive Service by improving the Hive architecture; the distributed data processing system is used to implement the board resource management method as described in any one of claims 1 to 5.

7. A board resource management device, characterized in that, include: The first determining module is used to determine the board type of each board based on the physical characteristic information and / or electrical signal data of each board in the transmission device; the physical characteristic information includes at least one of the following: size information, weight information, port information, and heat dissipation structure information; the board type includes at least one of the following: service board, system board, and new type board; The second determining module is used to determine the service load status of each board based on the real-time load data of each board monitored in real time; the real-time load data includes at least one of the following: processor utilization, memory usage, and network bandwidth usage. The third determining module is used to determine the idle rate threshold of each board based on the board type and the service load status of each board using an intelligent adjustment algorithm; the intelligent adjustment algorithm is an improved genetic algorithm combined with a simulated annealing algorithm.

8. An electronic device, characterized in that, include: processor; as well as A memory configured to store computer-executable instructions configured to be executed by the processor to implement the board resource management method as described in any one of claims 1-5.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store computer-executable instructions, which, when executed by a processor, implement the board resource management method as described in any one of claims 1-5.

10. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the board resource management method as described in any one of claims 1-5.

Images