Security protection task scheduling method, device, computer equipment and storage medium
By converting task execution data into frequency domain data, predicting the execution time of security protection tasks, and sorting and scheduling based on this, the problem of unreasonable resource allocation in the existing technology is solved, system efficiency and security are improved, and dynamic networks are adapted to environment.
Patent Information
- Application Number
- CN202411161668.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-23
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-08-23
Smart Images

Figure CN118689612B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of network security technology, and in particular to a security protection task scheduling method, device, computer equipment and storage medium. Background Art
[0002] At present, when carrying out security protection, various security protection devices are usually deployed, such as firewalls, website application-level intrusion prevention systems (WAFs), IDPS, etc., to achieve timely and effective handling of network security incidents. Due to the large variety of security protection tasks, different network security incidents require different security protection tasks. However, the current task scheduling of security protection tasks is often based on fixed rules or experience, which is difficult to adapt to the dynamically changing network environment, resulting in unreasonable resource allocation and affecting system efficiency and security. Summary of the invention
[0003] Based on this, it is necessary to provide a security protection task scheduling method, device, computer equipment and storage medium to address the above technical problems, so as to solve at least one problem existing in the above-mentioned prior art.
[0004] In a first aspect, the embodiment of the present application is implemented as follows: a security protection task scheduling method is provided, comprising:
[0005] Collecting task execution data, wherein the task execution data includes time domain data;
[0006] Converting the time domain data into frequency domain data;
[0007] Acquire protection task characteristics of the task execution data and statistical characteristics of the frequency domain data;
[0008] Based on the protection task characteristics and the statistical characteristics, predicting the execution time of the first security protection task;
[0009] Based on the execution time of the first security protection task, the security protection tasks are sorted and scheduled.
[0010] In one embodiment, after converting the time domain data into frequency domain data, the method further comprises:
[0011] Extracting the amplitude of each frequency component from the frequency domain data to generate an amplitude spectrum;
[0012] Based on the amplitude spectrum, the amplitudes of different frequency components are compared to determine a periodic pattern and a trend, wherein the periodic pattern is used to represent the frequency component corresponding to the peak value of the amplitude spectrum, and the trend is used to represent the rising or falling trend of the amplitude spectrum within the target frequency range;
[0013] Based on the periodic pattern and the trend, the demand for security protection tasks is predicted, so that the security protection tasks are scheduled based on the demand for the security protection tasks.
[0014] In one embodiment, converting the time domain data into frequency domain data includes:
[0015] Construct a finite Fourier transform model;
[0016] Based on the time domain data, obtaining a time domain signal sequence, the number of sampling points of the time domain data, and a time domain data index;
[0017] Based on the time domain signal sequence, the number of sampling points of the time domain data and the time domain data index, the time domain data is converted into frequency domain data through the finite Fourier transform model.
[0018] In one embodiment, predicting the execution time of the first security protection task based on the protection task feature and the statistical feature includes:
[0019] Build a function approximation model through gradient boosting regression tree;
[0020] Inputting the protection task characteristics and the statistical characteristics into the function approximation model to obtain the security protection task execution time predicted by each decision tree;
[0021] A weighted average is calculated for the predicted security protection task execution time to obtain a final security protection task execution time as the first security protection task execution time.
[0022] In one embodiment, the ordering and scheduling of the security protection tasks based on the execution time of the first security protection tasks includes:
[0023] Obtain attribute information of security protection tasks;
[0024] Based on the attribute information of the safety protection task, predicting the execution time of the second safety protection task;
[0025] Determining a target security protection task execution time based on the second security protection task execution time and the first security protection task execution time;
[0026] Based on the execution time of the security target protection tasks, the security protection tasks are sorted and scheduled.
[0027] In one embodiment, the ordering and scheduling of the security protection tasks based on the execution time of the first security protection tasks includes:
[0028] Determining a scheduling plan for all security protection tasks based on the execution time of the first security protection task;
[0029] The global minimum and local minimum algorithms are used to optimize the situation in order to obtain the optimal security protection task scheduling solution.
[0030] In one embodiment, the ordering and scheduling of the security protection tasks based on the execution time of the first security protection tasks includes:
[0031] Based on the execution time of the first security protection task, the security protection tasks are sorted according to a preset sorting strategy to obtain a security protection task priority;
[0032] Determine currently available resource information;
[0033] Based on the currently available resource information and the security protection task priority, the security protection task is scheduled according to the preset optimization target conditions.
[0034] In a second aspect, a security protection task scheduling device is provided, comprising:
[0035] A task execution data collection unit, used for collecting task execution data, wherein the task execution data includes time domain data;
[0036] A data conversion unit, used for converting the time domain data into frequency domain data;
[0037] A feature acquisition unit, used to acquire the protection task features of the task execution data and the statistical features of the frequency domain data;
[0038] A first security protection task execution time prediction unit, used for predicting the first security protection task execution time based on the protection task characteristics and the statistical characteristics;
[0039] A task scheduling unit is used to sort and schedule the security protection tasks based on the execution time of the first security protection tasks.
[0040] In a third aspect, a computer device is provided, comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein the processor implements the steps of the security protection task scheduling method as described above when executing the computer-readable instructions.
[0041] In a fourth aspect, a readable storage medium is provided, wherein the readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the security protection task scheduling method as described above are implemented.
[0042] The above-mentioned security protection task scheduling method, device, computer equipment and storage medium, and its method implementation include: collecting task execution data, the task execution data includes time domain data; converting the time domain data into frequency domain data; obtaining the protection task characteristics of the task execution data and the statistical characteristics of the frequency domain data; predicting the execution time of the first security protection task based on the protection task characteristics and the statistical characteristics; sorting and scheduling the security protection tasks based on the execution time of the first security protection task. In the embodiment of the present application, by converting the time domain signal into the frequency domain signal so as to be used for the subsequent prediction of the execution time of the protection task, thereby providing important reference information for task scheduling, it is possible to reduce the waste of resources during task scheduling and improve system efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.
[0044] Figure 1 It is a schematic diagram of a process of a security protection task scheduling method in one embodiment of the present invention;
[0045] Figure 2 This is a schematic diagram of an implementation scenario of a perceptron prediction method in one embodiment of the present invention;
[0046] Figure 3 It is a structural schematic diagram of a security protection task scheduling device in one embodiment of the present invention;
[0047] Figure 4 is a schematic diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION
[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0049] In one embodiment, if Figure 1 As shown, a security protection task scheduling method is provided, comprising the following steps:
[0050] In step S110, task execution data is collected, and the task execution data includes time domain data;
[0051] In the embodiments of the present application, it is possible to determine which systems or applications can generate relevant task execution data, such as logs of security protection systems, records of task scheduling systems, outputs of network monitoring tools, etc. The historical task execution data of the past period of time, such as one month, can be exported from these systems or applications by API calls, log file parsing, or extraction from databases. The collected historical execution data can then be integrated together and stored in a central storage database for unified analysis.
[0052] Furthermore, the collected historical task execution data can be preprocessed, such as data cleaning and data standardization. Data cleaning refers to removing problematic data and processing invalid values, such as removing duplicate, invalid or erroneous data records, and filling missing values by interpolation, deletion or estimation. Data standardization can be used to ensure uniform data formats, such as timestamp format, data units, etc., and perform normalization or standardization to eliminate the impact of dimensions.
[0053] It should be noted that the anti-DDoS cleaning protection tasks and anti-DDoS suppression protection tasks in different network planes and network layers can be used as basic task data for collection.
[0054] In step S120, the time domain data is converted into frequency domain data;
[0055] In an embodiment of the present application, the time domain data can be converted into frequency domain data by a Fourier transform method, such as a finite Fourier transform. Based on the frequency domain data, the amplitude of each frequency component can be extracted to form an amplitude spectrum, and the main periodic patterns and trends of the frequency domain data can be determined based on the amplitude spectrum, which can be used to predict future security protection task requirements, so as to schedule security protection tasks based on security protection requirements, and ensure that there are sufficient resources to cope with the security protection task requirements during peak periods.
[0056] In step S130, the protection task characteristics of the task execution data and the statistical characteristics of the frequency domain data are obtained;
[0057] In the embodiment of the present application, the protection task features may specifically include: task type, such as backbone cleaning tasks, cloud cleaning tasks, overseas suppression, other network suppression, local network suppression, etc. Task complexity, such as the number of tasks to be processed, network traffic size, etc. Optionally, the protection task feature extraction can be performed through feature engineering, or feature extraction can be performed through feature extraction models, such as ResNet, VGG, etc.
[0058] In the embodiment of the present application, the statistical characteristics of frequency domain data refer to useful information extracted from the signal frequency distribution, which may specifically include: power in a specific frequency range, peak frequency, etc.
[0059] In step S140, based on the protection task characteristics and the statistical characteristics, the execution time of the first security protection task is predicted;
[0060] In the embodiment of the present application, a gradient boosting regression tree model can be used to construct a function approximation to predict the execution time of the first safety protection task. Specifically, the protection task characteristics and statistical characteristics can be input into the above-mentioned function approximation model for prediction, and the residual of the previous model can be corrected by iteratively adding more decision trees, that is, each decision tree can try to predict the residual of the prediction results of all previous decision trees, and then the prediction results of all decision trees are weighted summed to obtain the final prediction result, that is, the execution time of the first safety protection task.
[0061] It should be noted that the prediction performance of the model can be optimized by adjusting the model's hyperparameters, such as learning rate, maximum tree depth, subsample ratio, etc.
[0062] In step S150, the security protection tasks are sorted and scheduled based on the execution time of the first security protection tasks.
[0063] In the embodiment of the present application, there are multiple tasks that need to be executed in the system at the same time, so the execution time of the first security protection task may include the execution time of each security protection task. Based on the execution time of the obtained security protection tasks, they can be sorted in the manner of shortest time first or longest time first, so as to obtain the priority of the security protection tasks. At the same time, the currently available resources, such as CPU, memory, bandwidth, etc., can be obtained, and the task scheduling can be performed with the goal of maximizing resource utilization and ensuring that high-priority tasks can be executed first.
[0064] In an embodiment of the present application, a method for scheduling security protection tasks is provided, including: collecting task execution data, the task execution data including time domain data; converting the time domain data into frequency domain data; obtaining the protection task characteristics of the task execution data and the statistical characteristics of the frequency domain data; predicting the execution time of a first security protection task based on the protection task characteristics and the statistical characteristics; and sorting and scheduling security protection tasks based on the execution time of the first security protection task. In an embodiment of the present application, by converting the time domain signal into the frequency domain signal so as to be used for the subsequent prediction of the execution time of the protection task, important reference information is provided for task scheduling, which can reduce the waste of resources during task scheduling and improve system efficiency.
[0065] In an embodiment of the present application, after converting the time domain data into frequency domain data, the method includes:
[0066] Extracting the amplitude of each frequency component from the frequency domain data to generate an amplitude spectrum;
[0067] Based on the amplitude spectrum, the amplitudes of different frequency components are compared to determine a periodic pattern and a trend, wherein the periodic pattern is used to represent the frequency component corresponding to the peak value of the amplitude spectrum, and the trend is used to represent the rising or falling trend of the amplitude spectrum within the target frequency range;
[0068] Based on the periodic pattern and the trend, the demand for security protection tasks is predicted, so that the security protection tasks are scheduled based on the demand for the security protection tasks.
[0069] Specifically, the amplitude of each frequency component can be extracted from the finite Fourier transform result, that is, the spectrum data, to obtain the amplitude spectrum. The amplitude spectrum shows the strength of each frequency component in the spectrum signal. According to the peak value in the amplitude spectrum, the periodic pattern can be determined. By comparing the amplitudes of different frequency components, the main periodic patterns and trends in the signal can be determined. Based on the periodic pattern and trend, the demand for security protection tasks within a certain period of time in the future, such as within 3 days or 1 hour, can be predicted. For example, the anti-DDoS cleaning protection task can be called first.
[0070] Among them, the periodic pattern is specifically manifested as an obvious peak in the amplitude spectrum, and the peak may correspond to the main frequency component of the signal. Therefore, by analyzing the position and height of each peak, the periodic pattern can be identified.
[0071] Among them, trend refers to the overall shape and trend of the amplitude spectrum, which can be used to provide signal trend information. For example, if the amplitude spectrum gradually rises or falls within a certain frequency range, this may indicate a long-term trend of the signal.
[0072] In an embodiment of the present application, converting the time domain data into frequency domain data includes:
[0073] Construct a finite Fourier transform model;
[0074] Based on the time domain data, obtaining a time domain signal sequence, the number of sampling points of the time domain data, and a time domain data index;
[0075] Based on the time domain signal sequence, the number of sampling points of the time domain data and the time domain data index, the time domain data is converted into frequency domain data through the finite Fourier transform model.
[0076] Specifically, the time domain data may include a time domain signal sequence, the number of sampling points of the time domain data, and a time domain data index, which can be input into the constructed finite Fourier transform model to output frequency domain data.
[0077] The finite Fourier transform model can be specifically expressed as:
[0078] ;
[0079] Among them, X(k) represents the complex value of the kth frequency component in the frequency domain. In the context of security protection tasks, X(k) can be understood as the signal strength at a specific frequency, identifying periodic network attack patterns or system behaviors.
[0080] x(n) represents the signal sequence in the time domain, which corresponds to the data points collected in chronological order in the security protection task, such as network traffic, number of system log events, CPU usage, etc. These data points reflect the status or activity of the system or network at a specific point in time.
[0081] N represents the length of the sequence, that is, the number of sampling points of the time domain data. In security protection tasks, it represents the number of samples collected during the time period of the analyzed data.
[0082] n represents the index of the time domain data, indicating the nth sampling point. In the context of security protection task data, it corresponds to the position of the data points arranged in time order.
[0083] k represents the index in the frequency domain, indicating the kth frequency component. In security protection task analysis, it is necessary to determine which specific frequency components dominate the data, thereby revealing periodic patterns or abnormal behaviors.
[0084] Represents the complex exponential function, which is used to convert time domain data into frequency domain data. In security protection tasks, the role of this function is to map each time domain data point to its corresponding frequency domain representation, thereby revealing the frequency characteristics in the data.
[0085] j represents the imaginary unit, which is used to construct the complex exponential function in the finite Fourier transform formula so that the transformation can capture the phase information of the signal.
[0086] By inputting the time domain data into the finite Fourier transform model for processing, a finite Fourier transform result, namely, frequency domain data, can be obtained.
[0087] In an embodiment of the present application, predicting the execution time of the first security protection task based on the protection task feature and the statistical feature includes:
[0088] Build a function approximation model through gradient boosting regression tree;
[0089] Inputting the protection task characteristics and the statistical characteristics into the function approximation model to obtain the security protection task execution time predicted by each decision tree;
[0090] A weighted average is calculated for the predicted security protection task execution time to obtain a final security protection task execution time as the first security protection task execution time.
[0091] Specifically, the function approximation can be constructed by the gradient boosting regression tree model, which can be specifically expressed as:
[0092] ;
[0093] Among them, x represents the input feature vector, including protection task features and statistical features represents the final prediction function, which is used to predict the execution time of the protection task. represents the initial model, which is usually set to the mean of the target variable in the training set. represents the decision tree added in the mth iteration, v is the learning rate, which controls the contribution of each tree to the final prediction result, and M is the number of iterations, that is, the total number of decision trees constructed.
[0094] A gradient boosting regression tree model can be used to construct a function approximation to predict the execution time of the first safety protection task. Specifically, the protection task features and statistical features are input into the above function approximation model for prediction processing, and the residual of the previous model can be corrected by iteratively adding more decision trees, that is, each decision tree can try to predict the residual of the prediction results of all previous decision trees, and then the prediction results of all decision trees are weighted summed to obtain the final prediction result, that is, the execution time of the first safety protection task.
[0095] It should be noted that the prediction performance of the model can be optimized by adjusting the model's hyperparameters, such as learning rate, maximum tree depth, subsample ratio, etc.
[0096] In an embodiment of the present application, the ordering and scheduling of the security protection tasks based on the execution time of the first security protection tasks includes:
[0097] Obtain attribute information of security protection tasks;
[0098] Based on the attribute information of the safety protection task, predicting the execution time of the second safety protection task;
[0099] Determining a target security protection task execution time based on the second security protection task execution time and the first security protection task execution time;
[0100] Based on the execution time of the security target protection tasks, the security protection tasks are sorted and scheduled.
[0101] See also Figure 2 Specifically, a protection task scheduling model can be constructed, which may include a perceptron and a multi-layer network algorithm model. The perceptron can be used to predict whether a task will be completed within a certain time threshold, so as to help understand which tasks take longer, thereby affecting the scheduling decision. The perceptron can be composed of two layers of neurons. After the input layer receives the input signal, it can be passed to the output layer, and the output layer can be an MP neuron. It can be specifically expressed by the following formula:
[0102] ;
[0103] Where: f(x) is the output, representing the predicted classification result, for example, 1 means the task is expected to be completed in time, -1 means the task may be delayed, It is the input feature, corresponding to various attributes in the security protection task, such as task complexity, data volume, etc. is the weight, which needs to be determined through training, and b is the bias term.
[0104] The training process of the perceptron is as follows: a training data set can be constructed, and the weights (i=1,2,...,n) and the bias term b can be obtained through learning. The bias term b can be regarded as a "dumb node" and the corresponding connection weight , so that the learning of weights and thresholds can be unified into the learning of weights. For the training example (x, y), if the output of the current perceptron is , then the perceptron weight is adjusted to:
[0105] ;
[0106] ;
[0107] in, It is called the learning rate. If the perceptron predicts the training sample (x, y) correctly, that is, , the perceptron does not change, otherwise the weight will be adjusted according to the degree of error.
[0108] For the multi-layer network algorithm model, it can handle more complex nonlinear relationships, so the predicted task execution time can be made more accurate, which can be specifically expressed as:
[0109] ;
[0110] in: is the task execution time predicted by the neural network. It is the input feature, corresponding to the relevant attributes of the security protection task. and are the weights and biases of the hidden layers respectively. and are the weights and biases of the output layer respectively. n is the number of neurons in the hidden layer. m is the number of input features.
[0111] The relevant attributes of the security protection task may specifically include the type of security protection task, complexity, amount of data to be processed, etc. By training a neural network to minimize the difference between the predicted time and the actual time, the accuracy of the prediction can be improved.
[0112] It should be noted that the execution time of the first security protection task predicted by function approximation is predicted based on historical data and may include the execution time of each task. Based on the initial prediction, the second security protection task execution time is predicted again using the perceptron and multi-layer network algorithm model, which can further capture and simulate more influencing factors of the task execution time, and can take into account more variables and nonlinear relationships. The two predicted times can complement each other and more comprehensively understand the possible range of task execution time, thereby obtaining the target security protection task execution time, and scheduling the security protection task based on the target security protection task execution time.
[0113] The prediction process of the perceptron model can be understood as an optimization problem, that is, minimizing the classification error by finding the best hyperplane. The goal of this optimization process is to find the global optimal solution, that is, the hyperplane that can correctly classify all samples.
[0114] In an embodiment of the present application, the ordering and scheduling of the security protection tasks based on the execution time of the first security protection tasks includes:
[0115] Determining a scheduling plan for all security protection tasks based on the execution time of the first security protection task;
[0116] The global minimum and local minimum algorithms are used to optimize the situation in order to obtain the optimal security protection task scheduling solution.
[0117] Specifically, in task scheduling, a task allocation method is first found through the perceptron model, and then various performance indicators are met in a globally optimal way through global minimum and local minimum algorithms, such as response time, system utilization, the severity of the attack suffered by the customer, whether the business is interrupted, the effectiveness of the existing protection tasks, task execution time and resource occupancy, etc., so as to continuously adjust and optimize the objective function to obtain the optimal security protection task scheduling plan.
[0118] Among them, global minimum refers to finding the optimal solution in the entire solution space, while local minimum refers to finding the optimal solution in a local area, thereby optimizing resource allocation and ensuring the task scheduling solution for timely and effective execution of tasks. The global minimum and local minimum algorithms can be specifically expressed by the following formula:
[0119] right , ,like ,but:
[0120] ;
[0121] If established, is a local minimum solution; if for any ( ) ,but is the global minimum solution. The local minimum solution is a point in the parameter space whose error function value of the neighboring points is not less than the function value of the point. The global minimum solution refers to the point where the error function value of all points in the parameter space is not less than the error function value of the point. They are called the local minimum and global minimum of the error function respectively.
[0122] in, Indicates the connection right, as well as represents the threshold value, Indicates the parameters, This statement is true for all individuals. represents the error function value of the neighborhood point, Represents the function value of a point. In the parameter space, if the condition is met, that is, the parameter > 0, it is a local minimum solution. If the parameter is arbitrary, it is a global minimum solution.
[0123] It should be noted that during the execution of the security protection scheduling model, the global minimum and local minimum algorithms are used to continuously search and evaluate the objective function values under various parameter combinations to obtain the optimal security protection task scheduling solution.
[0124] In an embodiment of the present application, the ordering and scheduling of the security protection tasks based on the execution time of the first security protection tasks includes:
[0125] Based on the execution time of the first security protection task, the security protection tasks are sorted according to a preset sorting strategy to obtain a security protection task priority;
[0126] Determine currently available resource information;
[0127] Based on the currently available resource information and the security protection task priority, the security protection task is scheduled according to the preset optimization target conditions.
[0128] Specifically, after the task execution time is predicted, different sorting strategies can be selected according to the actual situation to sort the tasks, thereby determining the execution priority of the security protection tasks, for example, the shortest execution time first or the longest execution time first.
[0129] Among them, the currently available resource information may include CPU, memory, bandwidth, etc., which can maximize resource utilization and ensure that high-priority tasks can be executed first. Resources are allocated according to the execution priority of the security protection task and the currently available resource information. This optimizes resource allocation and ensures that tasks are executed in a timely and effective manner.
[0130] In the embodiment of the present application, by combining finite Fourier transform and function approximation technology, the time domain signal is converted into a frequency domain signal to predict the execution time of the protection task, thereby achieving the accuracy of the task execution time. The prediction of the task execution time through the perceptron and the multi-layer network can intelligently predict the task execution time in the future based on historical data and the current system state, thereby providing an important reference for task scheduling. The protection scheduling model is constructed using the global minimum and local minimum optimization strategy. In the task scheduling process, not only the execution efficiency of the current task should be considered, but also the security and stability of the entire system should be taken into account. Therefore, this method realizes efficient and safe scheduling of tasks by finding the balance point of the global minimum and the local minimum, thereby improving resource utilization and system response speed. And it can continuously collect new task execution data and update the function approximation model to adapt to changes in the network environment. Through intelligent task scheduling strategies, resource waste can be reduced and system efficiency can be improved.
[0131] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.
[0132] In one embodiment, a security protection task scheduling device is provided, and the security protection task scheduling device corresponds to the security protection task scheduling method in the above embodiment. Figure 3 As shown, the security protection task scheduling device includes a task execution data acquisition unit 10, a data conversion unit 20, a feature acquisition unit 30, a first security protection task execution time prediction unit 40 and a task scheduling unit 50. Each functional module is described in detail as follows:
[0133] The task execution data collection unit 10 is used to collect task execution data, wherein the task execution data includes time domain data;
[0134] A data conversion unit 20, configured to convert the time domain data into frequency domain data;
[0135] A feature acquisition unit 30, used to acquire the protection task features of the task execution data and the statistical features of the frequency domain data;
[0136] A first security protection task execution time prediction unit 40, configured to predict the first security protection task execution time based on the protection task characteristics and the statistical characteristics;
[0137] The task scheduling unit 50 is used to sort and schedule the security protection tasks based on the execution time of the first security protection tasks.
[0138] In one embodiment of the present application, the device further includes: a periodic pattern and trend acquisition unit, which is used to:
[0139] Extracting the amplitude of each frequency component from the frequency domain data to generate an amplitude spectrum;
[0140] Based on the amplitude spectrum, the amplitudes of different frequency components are compared to determine a periodic pattern and a trend, wherein the periodic pattern is used to represent the frequency component corresponding to the peak value of the amplitude spectrum, and the trend is used to represent the rising or falling trend of the amplitude spectrum within the target frequency range;
[0141] Based on the periodic pattern and the trend, the demand for security protection tasks is predicted, so that the security protection tasks are scheduled based on the demand for the security protection tasks.
[0142] In an embodiment of the present application, the data conversion unit 20 is further used for:
[0143] Construct a finite Fourier transform model;
[0144] Based on the time domain data, obtaining a time domain signal sequence, the number of sampling points of the time domain data, and a time domain data index;
[0145] Based on the time domain signal sequence, the number of sampling points of the time domain data and the time domain data index, the time domain data is converted into frequency domain data through the finite Fourier transform model.
[0146] In an embodiment of the present application, the first security protection task execution time prediction unit 40 is further used to:
[0147] Build a function approximation model through gradient boosting regression tree;
[0148] Inputting the protection task characteristics and the statistical characteristics into the function approximation model to obtain the security protection task execution time predicted by each decision tree;
[0149] A weighted average is calculated for the predicted security protection task execution time to obtain a final security protection task execution time as the first security protection task execution time.
[0150] In one embodiment of the present application, the task scheduling unit 50 is further configured to:
[0151] Obtain attribute information of security protection tasks;
[0152] Based on the attribute information of the safety protection task, predicting the execution time of the second safety protection task;
[0153] Determining a target security protection task execution time based on the second security protection task execution time and the first security protection task execution time;
[0154] Based on the execution time of the security target protection tasks, the security protection tasks are sorted and scheduled.
[0155] In one embodiment of the present application, the task scheduling unit 50 is further configured to:
[0156] Determining a scheduling plan for all security protection tasks based on the execution time of the first security protection task;
[0157] The global minimum and local minimum algorithms are used to optimize the situation in order to obtain the optimal security protection task scheduling solution.
[0158] In one embodiment of the present application, the task scheduling unit 50 is further configured to:
[0159] Based on the execution time of the first security protection task, the security protection tasks are sorted according to a preset sorting strategy to obtain a security protection task priority;
[0160] Determine currently available resource information;
[0161] Based on the currently available resource information and the security protection task priority, the security protection task is scheduled according to the preset optimization target conditions.
[0162] In the embodiment of the present application, by combining finite Fourier transform and function approximation technology, the time domain signal is converted into a frequency domain signal to predict the execution time of the protection task, thereby achieving the accuracy of the task execution time. The prediction of the task execution time through the perceptron and the multi-layer network can intelligently predict the task execution time in the future based on historical data and the current system state, thereby providing an important reference for task scheduling. The protection scheduling model is constructed using the global minimum and local minimum optimization strategy. In the task scheduling process, not only the execution efficiency of the current task should be considered, but also the security and stability of the entire system should be taken into account. Therefore, this method realizes efficient and safe scheduling of tasks by finding the balance point of the global minimum and the local minimum, thereby improving resource utilization and system response speed. And it can continuously collect new task execution data and update the function approximation model to adapt to changes in the network environment. Through intelligent task scheduling strategies, resource waste can be reduced and system efficiency can be improved.
[0163] For the specific definition of the security protection task scheduling device, please refer to the definition of the security protection task scheduling method above, which will not be repeated here. Each module in the above-mentioned security protection task scheduling device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.
[0164] In one embodiment, a computer device is provided. The computer device may be a terminal device, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a readable storage medium. The readable storage medium stores computer-readable instructions. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer-readable instructions are executed by the processor, a security protection task scheduling method is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.
[0165] In an embodiment of the present application, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, and when the processor executes the computer-readable instructions, the steps of the security protection task scheduling method as described above are implemented.
[0166] In an embodiment of the application, a readable storage medium is provided, wherein the readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the security protection task scheduling method as described above are implemented.
[0167] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through computer-readable instructions, and the computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When the computer-readable instructions are executed, they may include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0168] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.
[0169] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.
Claims
1. A security protection task scheduling method, It is characterized in that The method comprises: Collecting task execution data, wherein the task execution data includes time domain data; Converting the time domain data into frequency domain data; Acquire protection task characteristics of the task execution data and statistical characteristics of the frequency domain data; Based on the protection task characteristics and the statistical characteristics, predicting the execution time of the first security protection task; Based on the execution time of the first security protection task, the security protection tasks are sorted and scheduled; wherein, after the time domain data is converted into frequency domain data, the method includes: Extracting the amplitude of each frequency component from the frequency domain data to generate an amplitude spectrum; Based on the amplitude spectrum, the amplitudes of different frequency components are compared to determine a periodic pattern and a trend, wherein the periodic pattern is used to represent the frequency component corresponding to the peak value of the amplitude spectrum, and the trend is used to represent the rising or falling trend of the amplitude spectrum within the target frequency range; Based on the periodic pattern and the trend, the demand for security protection tasks is predicted, so that the security protection tasks are scheduled based on the demand for the security protection tasks.
2. The security protection task scheduling method according to claim 1, It is characterized in that The converting the time domain data into frequency domain data comprises: Construct a finite Fourier transform model; Based on the time domain data, obtaining a time domain signal sequence, the number of sampling points of the time domain data, and a time domain data index; Based on the time domain signal sequence, the number of sampling points of the time domain data and the time domain data index, the time domain data is converted into frequency domain data through the finite Fourier transform model.
3. The security protection task scheduling method according to claim 1, It is characterized in that The predicting the execution time of the first security protection task based on the protection task feature and the statistical feature includes: Build a function approximation model through gradient boosting regression tree; Inputting the protection task characteristics and the statistical characteristics into the function approximation model to obtain the security protection task execution time predicted by each decision tree; A weighted average is calculated for the predicted security protection task execution time to obtain a final security protection task execution time as the first security protection task execution time.
4. The security protection task scheduling method according to claim 1, It is characterized in that The sorting and scheduling of the security protection tasks based on the execution time of the first security protection task includes: Obtain attribute information of security protection tasks; Based on the attribute information of the safety protection task, predicting the execution time of the second safety protection task; Determining a target security protection task execution time based on the second security protection task execution time and the first security protection task execution time; Based on the target security protection task execution time, the security protection tasks are sorted and scheduled.
5. The security protection task scheduling method according to claim 1, It is characterized in that The sorting and scheduling of the security protection tasks based on the execution time of the first security protection task includes: Determining a scheduling plan for all security protection tasks based on the execution time of the first security protection task; The global minimum and local minimum algorithms are used to optimize the situation in order to obtain the optimal security protection task scheduling solution.
6. The security protection task scheduling method according to claim 1, It is characterized in that The sorting and scheduling of the security protection tasks based on the execution time of the first security protection task includes: Based on the execution time of the first security protection task, the security protection tasks are sorted according to a preset sorting strategy to obtain a security protection task priority; Determine currently available resource information; Based on the currently available resource information and the security protection task priority, the security protection task is scheduled according to the preset optimization target conditions.
7. A security protection task scheduling device, It is characterized in that The device comprises: A task execution data collection unit, used for collecting task execution data, wherein the task execution data includes time domain data; A data conversion unit, used for converting the time domain data into frequency domain data; A feature acquisition unit, used to acquire the protection task features of the task execution data and the statistical features of the frequency domain data; A first security protection task execution time prediction unit, used to predict the first security protection task execution time based on the protection task characteristics and the statistical characteristics; A task scheduling unit, used for sorting and scheduling the security protection tasks based on the execution time of the first security protection tasks; The device further includes: a periodic pattern and trend acquisition unit, which is used to: Extracting the amplitude of each frequency component from the frequency domain data to generate an amplitude spectrum; Based on the amplitude spectrum, the amplitudes of different frequency components are compared to determine a periodic pattern and a trend, wherein the periodic pattern is used to represent the frequency component corresponding to the peak value of the amplitude spectrum, and the trend is used to represent the rising or falling trend of the amplitude spectrum within the target frequency range; Based on the periodic pattern and the trend, the demand for security protection tasks is predicted, so that the security protection tasks are scheduled based on the demand for the security protection tasks.
8. A computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer-readable instructions, the steps of the security protection task scheduling method as described in any one of claims 1 to 6 are implemented.
9. A readable storage medium storing computer-readable instructions. It is characterized in that When the computer-readable instructions are executed by a processor, the steps of the security protection task scheduling method as described in any one of claims 1 to 6 are implemented.
Citation Information
Patent Citations
Workshop crane control intelligent monitoring method and system for heat exchanger internal preassembled body
CN117311283A
Intelligent computing power calling management system and method based on data characteristics
CN118484313A
Processing time-domain and frequency-domain representations of EEG data
US20220101997A1
Method and device for estimating a pattern in a signal
US9208799B2
