Data migration scheduling method based on equipment anomaly detection

Abstract

The invention relates to the technical field of information, and discloses a data migration scheduling method based on equipment anomaly detection, which comprises the following steps of: S1, modeling through the health state of equipment, monitoring and estimating the health state of the equipment in real time, and judging whether the equipment is in an abnormal state or not based on an estimated value; s2, when the equipment is in an abnormal state, triggering data migration scheduling; s3, designing an optimization scheduling target, minimizing data migration delay, equipment load fluctuation and bandwidth resource consumption, and calculating a data migration path and a scheduling strategy based on an equipment health state and bandwidth constraints; and S4, solving an optimal scheduling problem with constraints by adopting a Lagrange multiplier method. Modeling and real-time estimation are carried out on the equipment health state through the stochastic differential equation and the Kalman filtering algorithm, accurate abnormal state detection is achieved, and the problems that in the prior art, response to equipment state changes is not timely, and accuracy is insufficient are effectively solved.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a data migration scheduling method based on device anomaly detection. Background Art

[0002] With the development of information technology, large-scale equipment networks and complex systems (such as data centers, IoT equipment networks, etc.) are increasingly widely used. The reliability of the equipment operating status in these systems is directly related to the overall performance and stability of the system. However, the operating environment and load conditions of the equipment are complex and changeable, resulting in significant randomness in the dynamic changes in the health status of the equipment, and increasing the risk of equipment failure or performance degradation. In order to reduce the system performance degradation or data loss caused by failures, data migration scheduling technology has gradually become a key means in modern equipment management.

[0003] Existing device data migration scheduling methods usually rely on the following technical means: fixed threshold anomaly detection mechanism. The detection of device abnormal status is usually based on a simple fixed threshold judgment method, such as triggering an alarm when parameters such as temperature and current exceed the preset range. However, this method has significant shortcomings in scenarios where the device health status fluctuates greatly. Static migration path and strategy design. In the existing technology, data migration path and strategy design are usually based on static analysis or historical experience, and lack the ability to adapt to the dynamic changes of device health status and system resources in real time. This fixed strategy is often inefficient when facing complex and dynamically changing device operating environments. The singleness of the optimization method. Many existing methods adopt simple optimization goals, such as minimizing migration time or reducing resource consumption, but fail to comprehensively consider the balance of multiple key performance indicators (such as migration delay, device health status fluctuations, and bandwidth resource consumption). Lack of closed-loop optimization for dynamic scheduling. The traditional migration scheduling process is often a single static execution, lacking real-time monitoring, feedback, and dynamic adjustment capabilities. Once the device operating environment changes (such as tight bandwidth resources or further deterioration of device health status), the scheduling strategy cannot effectively adapt, resulting in reduced migration efficiency or even system operation interruption. In summary, the existing equipment data migration scheduling methods have obvious deficiencies in terms of anomaly detection accuracy, dynamic scheduling adaptability, comprehensive optimization objectives, and closed-loop scheduling process. These problems are particularly prominent in complex equipment operating environments, which directly affect the efficiency of data migration and the reliability of the system. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a data migration scheduling method based on device anomaly detection, which solves the problems that the existing device data migration scheduling method has obvious shortcomings in terms of anomaly detection accuracy, dynamic scheduling adaptability, comprehensiveness of optimization objectives and closed-loop nature of the scheduling process.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a data migration scheduling method based on device anomaly detection comprises the following steps: S1. Monitor and estimate the health status of the device in real time through device health status modeling, and determine whether the device is in an abnormal state based on the estimated value; S2. When the device is in an abnormal state, data migration scheduling is triggered; S3. Design optimization scheduling objectives to minimize data migration delays, device load fluctuations, and bandwidth resource consumption, and calculate data migration paths and scheduling strategies based on device health status and bandwidth constraints; S4. Use the Lagrange multiplier method to solve the optimal scheduling problem with constraints and determine the amount of data migration between each device; S5. Use the random optimal control method to model the randomness of the equipment health status, so as to dynamically adjust the scheduling strategy to ensure the optimal data migration path of the equipment in normal or abnormal conditions; S6. Apply the obtained optimal migration scheduling strategy to the system in real time.

[0006] Preferably, the health status modeling step of the device includes: Obtain various monitoring data of the equipment and convert them into the initial value of the equipment health status; The change of equipment health status is modeled using stochastic differential equations, which can be expressed as: in, For equipment At the moment health status, For floating items, is the fluctuation term, is a random disturbance; Through observation data Compare with the health status model to obtain a real-time estimate of the equipment health status; By comparing the estimated value with the preset threshold, it is determined whether the device is in a normal or abnormal state, and a basis is provided for subsequent anomaly detection and data migration decisions.

[0007] Preferably, the health status estimation step of the device includes: Get the observation data of the device And assume that it is related to the health status of the device There is a linear relationship; The Kalman filter algorithm is used to estimate the health status of the equipment, and the health status of the equipment at each moment is obtained through recursive calculation. Estimated health status ; Estimated health status The preset normal health threshold Make a comparison to determine whether the device is in an abnormal state; When the device health status exceeds the threshold When the data migration schedule is triggered.

[0008] Preferably, the step of designing an optimized scheduling target includes: Determine the objective function of data migration ,in This includes minimizing data migration delays, fluctuations in device health, and system resource consumption; Set weight coefficient , used to balance the relative importance of migration delay, health status fluctuation, and bandwidth resource consumption; By calculating the amount of data migration between each device , combined with the maximum load capacity of the equipment and remaining available resources , optimize the objective function; Under bandwidth constraints, the optimal data migration path between computing devices is calculated to meet the system performance optimization goal.

[0009] Preferably, the step of optimizing the scheduling objective function includes: By minimizing data migration delays , Fluctuations in equipment health status and bandwidth resource consumption , construct the objective function; Assign weight coefficients to each indicator in the objective function , so that migration delay, health status fluctuation and resource consumption are optimized according to the predetermined proportion; Based on the remaining available resources of the device and bandwidth constraints, adjusting the amount of data migration between devices and ensure that the migration process does not exceed bandwidth limitations; Solve the optimization problem and obtain the optimal data migration scheduling strategy.

[0010] Preferably, the step of solving the optimal scheduling problem with constraints by using the Lagrange multiplier method includes: Construct the Lagrangian function, combine the objective function with the bandwidth constraint and the equipment health status constraint, and express it as: in, represents the Lagrangian function, represents the optimization objective function, Indicates the device Lagrange multiplier for health state constraints, Indicates the device At the moment The rate of change of health status, Indicates the device The drift term of health state change, The Lagrange multiplier representing the bandwidth constraint between devices, Indicates the device To the device At the moment The amount of data migration, Indicates the device and equipment The bandwidth limit between Indicates that all devices in the system The summation operation, is the total number of devices in the system; The Lagrangian function is used to calculate the health status of the equipment and data migration volume Find partial derivatives; The optimal data migration amount is obtained according to the Lagrange equation and the device health status evolution path ,to meet system performance and resource constraints; Based on the solution results, determine the optimal data migration path and scheduling strategy between each device.

[0011] Preferably, the step of using the stochastic optimal control method comprises: The equipment health status is modeled by establishing the Hamilton-Jacobi-Bellman equation, taking into account the random changes in the equipment health status; Defining state variables For equipment Health status, control variables For data migration strategies, the goal is to minimize the risk of equipment failure and migration delays; Solve the Hamilton-Jacobi-Bellman equation to obtain the optimal control strategy , that is, the optimal strategy for data migration between devices; Based on the optimal control strategy ,Dynamically adjust the data migration path between devices to ensure that the system always maintains the optimal scheduling during the device status change process.

[0012] Preferably, the step of applying the obtained optimal migration scheduling strategy to the system in real time includes: The optimal migration scheduling strategy Transmit the data to the scheduling execution module to ensure that each device migrates data along the optimal path; Dynamically adjust the scheduling strategy based on the real-time monitored equipment health status and system load changes; During the data migration process, the bandwidth and device resource utilization are continuously monitored to ensure that the scheduling strategy always meets the bandwidth limit and device health status constraints in actual operation; Update the scheduling strategy in time according to the changes in equipment status to ensure that the system is always in the optimal operating state.

[0013] Preferably, the step of solving the optimization problem of data migration scheduling by adopting dynamic programming includes: Based on the optimal control strategy and scheduling objective function , discretize the scheduling problem into sub-problems at several moments; Through the dynamic programming algorithm, the optimal data migration amount between devices at each moment is gradually calculated , and consider the constraints of bandwidth and device health status.

[0014] The present invention also provides a data migration scheduling system based on device anomaly detection, comprising: Health status monitoring module, which is used to monitor the health status of each device in real time and estimate the health status of the device through random process modeling; An anomaly detection module, used to determine whether the device is in an abnormal state based on the health status estimation value; The scheduling decision module is used to generate data migration scheduling strategies based on changes in device health status and bandwidth constraints; The optimization calculation module is used to calculate the optimal data migration path and scheduling strategy based on the Lagrange multiplier method and the stochastic optimal control method; The real-time scheduling execution module is used to adjust the data migration path and scheduling strategy in the system according to the optimization results.

[0015] The present invention provides a data migration scheduling method based on device anomaly detection. It has the following beneficial effects: 1. The present invention models and estimates the health status of the equipment in real time through stochastic differential equations and Kalman filtering algorithms, realizes accurate abnormal state detection, and effectively solves the problems of untimely response to equipment state changes and insufficient accuracy in the prior art.

[0016] 2. The present invention comprehensively optimizes migration delay, equipment load fluctuation and bandwidth resource consumption, constructs a multi-objective optimization function and combines health status and resource constraints to provide an efficient data migration path and scheduling strategy, thereby solving the problems of low migration efficiency and low resource utilization in the prior art.

[0017] 3. The present invention dynamically adjusts the scheduling strategy by utilizing a random optimal control method, and can always maintain the optimality of the migration path when the equipment health status and system resource conditions change, thereby solving the problem of insufficient adaptability of fixed scheduling strategies in the prior art to complex dynamic environments.

[0018] 4. The present invention ensures that the system can efficiently complete data migration when the equipment is in an abnormal state through the collaborative work of real-time health status monitoring, optimization calculation and scheduling execution modules, solving the problem of lack of closed-loop optimization of scheduling strategies under abnormal conditions in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a flow chart of the method of the present invention; Figure 2 It is a system architecture diagram of the present invention. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] Please see attached Figure 1 , an embodiment of the present invention provides a data migration scheduling method based on device anomaly detection, comprising the following steps: S1. Monitor and estimate the health status of the device in real time through device health status modeling, and determine whether the device is in an abnormal state based on the estimated value; S2. When the device is in an abnormal state, data migration scheduling is triggered; S3. Design an optimization scheduling goal to minimize data migration delay, device load fluctuation, and bandwidth resource consumption, and calculate the data migration path and scheduling strategy based on device health status and bandwidth constraints; S4. Use the Lagrange multiplier method to solve the optimal scheduling problem with constraints and determine the amount of data migration between each device; S5. Use the random optimal control method to model the randomness of the equipment health status, so as to dynamically adjust the scheduling strategy to ensure the optimal data migration path of the equipment in normal or abnormal conditions; S6. Apply the obtained optimal migration scheduling strategy to the system in real time.

[0022] For step S1, the present invention provides a data migration scheduling method based on device anomaly detection, which is used to trigger data migration scheduling when device anomalies are detected, and dynamically adjust the data migration path and strategy through optimization algorithms to ensure system performance and efficient use of resources. Compared with the prior art, the method of the present invention can achieve efficient and reliable data migration management by real-time modeling and estimation of the health status of the equipment, combined with optimization scheduling objectives and constraints, and has significant advantages, especially in scenarios where the operating status of the equipment changes randomly.

[0023] In this embodiment, first, by modeling the health status of the device, a random process is used to describe the dynamic changes of the health status of the device. The health status modeling is based on stochastic differential equations, which are specifically expressed as: in, Indicates the device At the moment health status, is the drift term, which indicates the average change trend of the health status of the equipment. is the fluctuation term, reflecting the fluctuation intensity of health status, is the standard Brownian motion, and represents the random disturbance term. Through this model, the random evolution process of the equipment health status between normal operation and abnormal state can be characterized.

[0024] Next, obtain the real-time monitoring data of the equipment, including operating parameters (such as temperature, vibration frequency, current, etc.) and historical data records. Through data preprocessing, the raw data is converted into the initial value of the health status. , as the input condition of the stochastic differential equation. It should be noted that the accuracy of the initial value has an important impact on the accuracy of subsequent health state estimation and anomaly detection.

[0025] Specifically, by using the real-time observation data of the equipment The health state is estimated by fusion calculation with the health state model using the Kalman filter algorithm. The state equation and observation equation of the Kalman filter are: in, For the device The health status at a moment, is the corresponding observed value, and are process noise and observation noise respectively, which are assumed to obey zero-mean normal distribution.

[0026] Based on the above model and real-time monitoring data, an estimated value of the equipment health status is obtained. By comparing the estimated value with the preset normal health status threshold, it can be determined whether the equipment is in an abnormal state. When the estimated value exceeds the threshold, the condition is met. ,The system will immediately trigger an abnormal alarm and enter the data migration ,scheduling phase.

[0027] In one implementation, the health status threshold can be dynamically adjusted according to the type of device and the operating environment. For example, for a server with high-load tasks, its health status threshold may be set lower to allow data migration at an early stage of potential failure. For low-priority devices, the threshold can be appropriately increased to reduce unnecessary scheduling frequency.

[0028] It should be noted that in order to improve the accuracy of health status estimation, the present invention can introduce a multi-model filtering method. By constructing multiple hypothetical models (such as different health status drift rates and fluctuation intensities), a weighted fusion technique is used to integrate multiple estimation results, thereby generating a more robust health status prediction value.

[0029] In an exemplary embodiment, the method of the present invention also supports long-term prediction of the trend of changes in the health status of the equipment. For example, by using the time series of health status estimates, combined with a time series analysis algorithm, It can predict the changing trend of future health status and provide more forward-looking decision support for anomaly detection and data migration scheduling.

[0030] The above health status modeling and real-time estimation process lays the foundation for data migration scheduling. The core lies in using stochastic differential equations and Kalman filtering algorithms to effectively capture the dynamic changes in the health status of equipment and its random characteristics, thereby achieving accurate judgment of abnormal equipment status.

[0031] For step S2, during the data migration scheduling process, accurately judging whether the device is in an abnormal state is a key prerequisite for the effectiveness of the scheduling strategy. In order to achieve this goal, the present invention accurately detects the abnormal state of the device through modeling and real-time estimation of the device health status, combined with a threshold judgment mechanism, thereby providing a reliable basis for triggering data migration. Specifically, the core of step S2 is to use stochastic differential equations and real-time monitoring data to build a modeling and estimation framework for the device health status. In this embodiment, the health status modeling process is based on the form of stochastic differential equations, which can be described as follows: in, Indicates the device At the moment health status, is the drift term, representing the trend change of health status, is the fluctuation term, indicating the fluctuation intensity of health status, It is a standard Brownian motion, reflecting the impact of random disturbances on the health state. This model can better describe the state changes of the equipment during operation, and is particularly suitable for the nonlinear and random characteristics of the health state of the equipment.

[0032] Specifically, the modeling steps include the following: First, obtain the real-time monitoring data of the equipment, such as temperature, power consumption, vibration frequency, current, etc., and preprocess the data according to its characteristics and standardize it into a unified range to eliminate the dimensional differences between different physical quantities.

[0033] Next, the model parameters are determined based on the historical data and domain knowledge of the equipment. and These parameters can be estimated by statistical methods or machine learning algorithms. For example, the drift term is calculated by using the mean change rate of the historical health status. , using the variance of the health state to estimate the fluctuation term .

[0034] In order to improve the accuracy of health state estimation, the Kalman filter algorithm is used to update the model in real time. The Kalman filter algorithm includes state equations and observation equations, and its specific form is as follows: in, Indicates that the device is Health status at all times, represents the observed data, and are the state transfer matrix and the observation matrix respectively, and are process noise and observation noise, which are assumed to obey zero-mean normal distribution.

[0035] In one possible implementation, by observing the data and modeling health status The estimated error is calculated in real time and the estimated value of the health state is dynamically corrected based on the Kalman gain to obtain a health state prediction value that is more in line with the actual situation. .

[0036] It should be noted that the real-time estimation result of health status is consistent with the preset health threshold. Comparison is used to determine whether the device is in an abnormal state. The specific judgment conditions are: like , the device is in normal state; like , the device is in an abnormal state.

[0037] The above threshold It can be adjusted dynamically based on the importance of the device and the operating environment. For example, for mission-critical devices, the threshold can be set lower to detect abnormal conditions earlier; for non-critical devices, the threshold can be appropriately relaxed to reduce false alarm rates.

[0038] As an option, in order to enhance the robustness of health status modeling, the present invention also supports multi-model filtering technology. By simultaneously constructing multiple hypothetical models (for example, different combinations of drift terms and fluctuation terms), the prediction results of each model are weighted averaged, thereby improving the accuracy of the estimated value.

[0039] In an exemplary embodiment, the present invention also provides a health status prediction function. By analyzing the health status time series data, using methods such as autoregressive integral moving average model or long short-term memory network, the future health status change trend of the equipment can be predicted. This prediction capability provides important support for triggering data migration in advance and formulating scheduling strategies.

[0040] Through the above health status modeling and estimation method, the present invention can not only accurately detect the abnormal state of the equipment, but also capture the random change characteristics of the health status, laying a solid foundation for subsequent optimization scheduling.

[0041] For step S3, it involves designing an optimization scheduling objective to minimize data migration delay, device load fluctuation and bandwidth resource consumption, and calculating the data migration path and scheduling strategy based on the device health status and bandwidth constraints. This process effectively guides the solution of the optimal scheduling strategy in the subsequent steps by reasonably setting the optimization objective function and related constraints. In the overall approach, the design of the optimal scheduling target needs to comprehensively consider the balance between multiple performance indicators to ensure that data migration can be performed efficiently while meeting the device health status and system resource constraints. This is an important prerequisite for achieving dynamic optimization of scheduling strategies.

[0042] In this embodiment, the design of the optimization scheduling target and related steps are as follows: First, determine the form of the objective function to comprehensively reflect the impact of migration delay, device health status fluctuation, and bandwidth resource consumption on system performance. The objective function can be expressed as: in: Indicates the device and equipment The migration delay between Indicates the impact of fluctuations in device health status on migration. Indicates the consumption of bandwidth resources between devices. are the weight coefficients of migration delay, health status fluctuation and bandwidth resource consumption respectively.

[0043] As an option, the system can dynamically adjust the weight coefficient according to actual needs .

[0044] Specifically, the following constraints need to be considered when designing the objective function: Equipment health status constraints: To ensure that the equipment will not fail due to deterioration of its health status during the migration process, the health status change of the equipment needs to meet certain conditions. The health status is modeled by stochastic differential equations, and its constraints can be expressed as: Indicates the device At the moment .

[0045] Bandwidth resource constraint: The amount of data migration must not exceed the available bandwidth limit between devices. This constraint is expressed as: in, limit.

[0046] Device carrying capacity constraint: The amount of migrated data cannot exceed the storage and processing capacity of the target device. This constraint is expressed as: in, For equipment Maximum carrying capacity.

[0047] In one possible implementation, the data migration path and scheduling strategy between devices can be determined by comprehensive optimization of the objective function and constraints. For example, an optimization algorithm based on the gradient descent method can be used to iteratively solve the objective function, and the amount of data migration between devices can be updated in each iteration. , to approach the optimal solution.

[0048] It should be noted that in order to improve the efficiency and stability of the optimization process, heuristic algorithms, such as particle swarm optimization or simulated annealing, can also be combined to avoid falling into the local optimum while searching for the global optimal solution.

[0049] In some embodiments, the device health status fluctuation part in the objective function can be Detailed modeling. For example, the fluctuation of equipment health status can be regarded as a dynamic time series, and a prediction model can be built using historical data to predict potential health status changes in advance during the optimization process and make corresponding adjustments. This approach can further improve the accuracy of the scheduling strategy.

[0050] The design of the optimal scheduling target is the core part of the data migration scheduling method. Its rationality and completeness directly determine the quality of the scheduling strategy. By comprehensively considering factors such as data migration delay, fluctuations in device health status, and bandwidth resource consumption, and performing optimization calculations under constraints, it can ensure that the system always maintains an efficient and reliable operating state under different operating conditions. At the same time, this design provides clear theoretical guidance for subsequent optimization algorithms.

[0051] For step S4, in order to accurately solve the balance problem between system constraints and objectives in the optimization scheduling, the present invention proposes an optimization solution step based on the Lagrange multiplier method, which is suitable for solving nonlinear optimal scheduling problems with constraints. This step combines multiple constraints with the objective function by constructing a Lagrangian function, which theoretically ensures the global optimality and feasibility of the scheduling scheme.

[0052] In this embodiment, the following steps are first used to construct a Lagrangian function, which comprehensively considers multiple factors such as data migration delay, device health status constraints, and bandwidth limitations. The form of the Lagrangian function is expressed as: in: represents the Lagrangian function, To optimize the objective function, the specific design is to minimize data migration delay, device load fluctuation and bandwidth consumption. Equality constraints related to the health status of the equipment are used to ensure that the equipment is operating within a safe range. is the inequality expression of bandwidth resource constraint, and are the Lagrange multipliers for health state and bandwidth limit, respectively, and Represent the number of constraints respectively.

[0053] In the above formula, the change of equipment health status is modeled by stochastic differential equations, specifically: .

[0054] It should be noted that the bandwidth limit between devices is described in the form of inequality constraints: in , .

[0055] Next, by taking partial derivatives of the Lagrangian function with respect to the decision variables and the Lagrangian multiplier, we obtain the following necessary conditions: These conditions ensure that the optimization results reach the optimal solution under all constraints.

[0056] Specifically, in one implementation, the optimization problem of the Lagrangian function can be solved by an iterative algorithm. Initially, the Lagrangian multiplier is set to zero and gradually adjusted so that the amount of data migration between devices is Meet bandwidth constraints while ensuring health constraints To improve computational efficiency, the Lagrange multiplier can be dynamically adjusted in combination with the gradient descent method.

[0057] As an option, to further ensure the global optimality of the solution, the Lagrangian solution can be combined with the stochastic optimal control method. For example, when the randomness of the equipment health state is strong, the impact of random disturbances can be modeled through the optimal control method to improve the robustness of the system.

[0058] It should be noted that the evaluation results of the device health status are directly related to the allocation of bandwidth resources. Therefore, during the calculation process, the weight coefficient can be dynamically adjusted to make the balance between health status fluctuations and bandwidth resource consumption more in line with actual needs. For example, when the device status is close to abnormal, the weight of the health status constraint can be increased to prioritize the safety of device operation.

[0059] In summary, the introduction of the Lagrange multiplier method not only achieves the effective combination of multiple constraints, but also ensures the optimality and stability of system resource utilization, laying a theoretical foundation for the dynamic adjustment of subsequent scheduling strategies.

[0060] For step S5, in order to further optimize the data migration scheduling under abnormal device status, the present invention proposes a method based on random optimal control, which aims to dynamically adjust the scheduling strategy by modeling the randomness of the device health status, so as to ensure the optimality of the data migration path under normal or abnormal device status. This method achieves global optimization by introducing the Hamilton-Jacobi-Bellman equation and combining the association between the device health status and the migration strategy.

[0061] In this embodiment, a stochastic model of the health status of the device is first established, and the following stochastic differential equation is used to describe the dynamic changes of the health status: in, The health status of the device. The drift term of the health status describes the deterministic change trend of the health status. is a fluctuation term, describing the randomness of the health status, This is a standard Brownian motion.

[0062] Based on the model, the data migration problem is regarded as an optimal control problem with random constraints. The objective function is defined as: in, , is the operating cost function, including migration delay, device health status fluctuation and bandwidth consumption, is the penalty coefficient of the migration strategy, which is used to balance the size of the migration amount. is the terminal cost function, which describes the performance of the equipment in the final state.

[0063] It should be noted that by minimizing the objective function , which can effectively reduce migration delays and suppress fluctuations in device health status.

[0064] In one possible implementation, the optimal control strategy is obtained by solving the Hamilton-Jacobi-Bellman equation. The Hamilton-Jacobi-Bellman equation is in the form of: in is the value function, which indicates that the system is in state .

[0065] As an option, a numerical iteration method can be used to solve the above Hamilton-Jacobi-Bellman equation. Initially, based on the current observation of the equipment health status .

[0066] Specifically, in some embodiments, the migration path between devices can be adjusted in real time according to the optimal strategy. For example, when the health status of the device fluctuates greatly, the migration amount It will be dynamically reduced, thereby reducing the pressure on bandwidth resources. At the same time, for devices whose health status is close to abnormal, the migration amount can be increased first to ensure the safe transfer of data.

[0067] It should be noted that in actual applications, the adjustment of migration strategies needs to comprehensively consider the system bandwidth limitation and the load distribution of the devices. For example, in the scenario of data migration between multiple devices, a priority mechanism can be introduced to allocate bandwidth resources according to the degree of deterioration of the health status to maximize the overall efficiency of the system.

[0068] Through the above method, this embodiment successfully solves the problem that the traditional method cannot handle random disturbances and nonlinear constraints. This solution can not only ensure the global optimality of data migration scheduling, but also has strong dynamic adaptability, and provides an effective solution for data migration scheduling under abnormal device conditions.

[0069] For step S6, in order to ensure the effectiveness and real-time performance of data migration scheduling under abnormal device status, the present invention further proposes a method for applying the optimal migration scheduling strategy to the system in real time. The core of this method is to ensure that the system can adapt to the dynamic changes of device health status and system resources through real-time strategy execution and monitoring, while effectively achieving the migration scheduling goals. Specifically, through the scheduling execution module, combined with the health status monitoring and feedback mechanism, the theoretical optimal strategy is converted into an actual scheduling operation.

[0070] In this embodiment, the optimal migration scheduling strategy is first transmitted to the scheduling execution module in the system. During this process, the scheduling execution module will parse the scheduling strategy generated by the optimization calculation module and convert it into operable execution instructions. As an option, these instructions can be transmitted to the corresponding device end node through the network to instruct the device to perform data transfer operations according to the preset optimal migration path.

[0071] Specifically, the execution of the scheduling strategy requires real-time collection of the health status, load distribution, and bandwidth utilization of the devices in the system. For example: The health status monitoring module continuously monitors the health status of the equipment and feeds real-time data back to the scheduling decision module.

[0072] After receiving the instruction, the scheduling execution module will dynamically adjust the execution order and migration amount according to the current health status and load changes of the device.

[0073] In one implementation, the dynamic adjustment mechanism of the scheduling strategy includes the following steps: First, based on the health status data fed back by the health status monitoring module, evaluate whether the device is still in an abnormal state.

[0074] Secondly, when the device load or bandwidth resources change significantly, the scheduling execution module will re-evaluate the feasibility of the migration path. For example, when the bandwidth utilization of a path is close to saturation, the migration traffic can be reallocated to other available paths.

[0075] It should be noted that during the execution process, the scheduling execution module is not only responsible for the real-time operation of data migration, but also needs to continuously monitor possible exceptions that may occur during the execution process. For example: When bandwidth resources are limited, the module will reduce the migration rate to avoid large fluctuations in system performance.

[0076] When the health status of the target device deteriorates, the current migration operation can be interrupted immediately to prioritize the protection of critical data.

[0077] As an extension, in one embodiment, the execution of the scheduling strategy can be further optimized in combination with intelligent algorithms. For example, the trend of device health status changes can be dynamically predicted based on a deep reinforcement learning model, so as to adjust the migration strategy in advance. In this case, the scheduling execution module is not only a passive execution tool, but also can predict potential risks through machine learning models and actively optimize the execution strategy.

[0078] In order to ensure that the scheduling strategy is always in the optimal state, the system also designs a closed-loop feedback mechanism. During the data migration process, the scheduling execution module will feed back the latest data on bandwidth resource utilization and device health status to the optimization calculation module in real time. The optimization calculation module updates the Lagrange multiplier or random control strategy parameters accordingly, so that the system's scheduling decisions can adapt to dynamic changes.

[0079] The advantage of this method is that it achieves a seamless connection from theoretical optimization to practical application. Through real-time policy execution and dynamic adjustment, it not only ensures the system performance during data migration, but also greatly improves the system's ability to cope with abnormal conditions. This method is particularly suitable for scenarios with large-scale device networks and complex bandwidth resource constraints, and provides a reliable and efficient solution for data migration scheduling under abnormal device conditions.

[0080] Please see attached Figure 2 The present invention also provides a data migration scheduling system based on device anomaly detection, including: Health status monitoring module, which is used to monitor the health status of each device in real time and estimate the health status of the device through random process modeling; An anomaly detection module, used to determine whether the device is in an abnormal state based on the health status estimation value; The scheduling decision module is used to generate data migration scheduling strategies based on changes in device health status and bandwidth constraints; The optimization calculation module is used to calculate the optimal data migration path and scheduling strategy based on the Lagrange multiplier method and the stochastic optimal control method; The real-time scheduling execution module is used to adjust the data migration path and scheduling strategy in the system according to the optimization results.

[0081] Health status monitoring module: The main function of this module is to monitor and estimate the health status of the equipment in real time through random process modeling, and provide basic data support for anomaly detection and scheduling strategy formulation.

[0082] Specific functions include: Collect real-time data of equipment operation, such as temperature, workload, running time and other status information.

[0083] The health state evolution of the equipment is modeled through stochastic processes, such as models based on stochastic differential equations, which describe the floating terms, fluctuation terms and random disturbances of the health state.

[0084] Calculate and provide the health status value of the device at the current moment as a basis for subsequent judgment of whether it is abnormal.

[0085] Ensure that the monitoring data is accurate and reliable, and provide real-time health status input values ​​for the anomaly detection module.

[0086] The anomaly detection module is used to determine whether the device is in an abnormal state based on the health status estimation value, so as to trigger data migration scheduling in time under abnormal circumstances.

[0087] Specific functions include: Compare the estimated value of the equipment health status with the set threshold range to distinguish between normal and abnormal conditions of the equipment.

[0088] When the health status exceeds the threshold, an abnormal alarm signal is generated and the start of the data migration scheduling strategy is triggered.

[0089] Provide clear abnormal equipment information and provide input for target setting of the scheduling decision module.

[0090] Ensure that the judgment of abnormal status is highly accurate and real-time to avoid false alarms or missed alarms.

[0091] Scheduling decision module: This module generates specific data migration scheduling strategies based on changes in device health status and bandwidth constraints to ensure reasonable allocation and efficient use of system resources.

[0092] Specific functions include: Consider health status information and bandwidth constraints comprehensively, and set the target device and migration path for data migration.

[0093] Develop preliminary data migration volume and scheduling strategies based on real-time health status and network resource conditions.

[0094] Clarify the migration data allocation and transmission paths between devices, and provide a feasible scheduling framework for optimizing computing modules.

[0095] Ensure that the generated scheduling policy takes into account latency minimization, load balancing, and resource consumption control.

[0096] The optimization calculation module calculates the optimal data migration path and scheduling strategy through mathematical optimization methods (such as Lagrange multiplier method and stochastic optimal control method) to ensure the optimal global performance of the system under constraints.

[0097] Specific functions include: Construct an objective function that combines data migration latency, fluctuations in device health status changes, and bandwidth resource consumption indicators.

[0098] The optimization objective function is combined with bandwidth constraints and equipment health status constraints to construct a Lagrangian function and solve the optimal migration amount.

[0099] Through the random optimal control method, we can dynamically respond to the randomness of changes in equipment health status and optimize data migration paths and strategies.

[0100] The optimal data migration amount and scheduling strategy between output devices provide precise guidance for the real-time scheduling execution module.

[0101] The real-time scheduling execution module applies the scheduling strategy output by the optimization calculation module to the system, dynamically adjusts the data migration path and strategy between devices, and ensures that the entire scheduling process is efficient and smooth.

[0102] Specific functions include: Receive optimization calculation results and distribute the optimal migration strategy to specific devices.

[0103] Dynamically adjust the migration path and migration amount based on the real-time monitored device health status and system load changes.

[0104] Continuously monitor bandwidth utilization and device resource usage during the migration process to ensure that the scheduling strategy meets various constraints in actual operation.

[0105] Respond promptly to changes in device health status or sudden anomalies, iteratively update strategies, and maintain the optimal state of overall system performance.

[0106] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A data migration scheduling method based on device anomaly detection, characterized in that: The following steps are involved: S1. Monitor and estimate the health status of the device in real time through device health status modeling, and determine whether the device is in an abnormal state based on the estimated value; S2. When the device is in an abnormal state, data migration scheduling is triggered; S3. Design optimization scheduling objectives to minimize data migration delays, device load fluctuations, and bandwidth resource consumption, and calculate data migration paths and scheduling strategies based on device health status and bandwidth constraints; S4. Use the Lagrange multiplier method to solve the optimal scheduling problem with constraints and determine the amount of data migration between each device; S5. Use the random optimal control method to model the randomness of the equipment health status, so as to dynamically adjust the scheduling strategy to ensure the optimal data migration path of the equipment in normal or abnormal conditions; S6. Apply the obtained optimal migration scheduling strategy to the system in real time.

2. The data migration scheduling method based on device anomaly detection according to claim 1 is characterized in that: The health status modeling step of the device includes: Obtain various monitoring data of the equipment and convert them into the initial value of the equipment health status; The change of equipment health status is modeled using stochastic differential equations, which can be expressed as: in, For equipment At the moment health status, For floating items, is the fluctuation term, is a random disturbance; Through observation data Compare with the health status model to obtain a real-time estimate of the equipment health status; By comparing the estimated value with the preset threshold, it is determined whether the device is in a normal or abnormal state, and a basis is provided for subsequent anomaly detection and data migration decisions.

3. The data migration scheduling method based on device anomaly detection according to claim 2 is characterized in that: The health status estimation step of the device comprises: Get the observation data of the device And assume that it is related to the health status of the device There is a linear relationship; The Kalman filter algorithm is used to estimate the health status of the equipment, and the health status of the equipment at each moment is obtained through recursive calculation. Estimated health status ; Estimated health status The preset normal health threshold Make a comparison to determine whether the device is in an abnormal state; When the device health status exceeds the threshold When the data migration schedule is triggered.

4. The data migration scheduling method based on device anomaly detection according to claim 1 is characterized in that: The step of designing an optimized scheduling target comprises: Determine the objective function of data migration ,in This includes minimizing data migration delays, fluctuations in device health, and system resource consumption; Set weight coefficient ,used to balance the relative importance of migration delay, health status fluctuation, and bandwidth resource consumption; By calculating the amount of data migration between each device , combined with the maximum load capacity of the equipment and remaining available resources , optimize the objective function; Under bandwidth constraints, the optimal data migration path between computing devices is calculated to meet the system performance optimization goal.

5. The data migration scheduling method based on device anomaly detection according to claim 4 is characterized in that: The step of optimizing the scheduling objective function comprises: By minimizing data migration delays , Fluctuations in equipment health status and bandwidth resource consumption , construct the objective function; Assign weight coefficients to each indicator in the objective function , so that migration delay, health status fluctuation and resource consumption are optimized according to the predetermined proportion; Based on the remaining available resources of the device and bandwidth constraints, adjusting the amount of data migration between devices and ensure that the migration process does not exceed bandwidth limitations; Solve the optimization problem and obtain the optimal data migration scheduling strategy.

6. The data migration scheduling method based on device anomaly detection according to claim 1 is characterized in that: The steps of solving the optimal scheduling problem with constraints by using the Lagrange multiplier method include: Construct the Lagrangian function, combine the objective function with the bandwidth constraint and the equipment health status constraint, and express it as: in, represents the Lagrangian function, represents the optimization objective function, Indicates the device Lagrange multiplier for health state constraints, Indicates the device At the moment The rate of change of health status, Indicates the device The drift term of health state change, Lagrange multiplier representing the bandwidth constraint between devices, Indicates the device To the device At the moment The amount of data migration, Indicates the device and equipment The bandwidth limit between Indicates that all devices in the system The summation operation, is the total number of devices in the system; The Lagrangian function is used to calculate the health status of the equipment and data migration volume Find partial derivatives; The optimal data migration amount is obtained according to the Lagrange equation and the device health status evolution path , to meet system performance and resource constraints; Based on the solution results, determine the optimal data migration path and scheduling strategy between each device.

7. The data migration scheduling method based on device anomaly detection according to claim 1 is characterized in that: The steps of utilizing the stochastic optimal control method include: The equipment health status is modeled by establishing the Hamilton-Jacobi-Bellman equation, taking into account the random changes in the equipment health status; Defining state variables For equipment Health status, control variables For data migration strategies, the goal is to minimize the risk of equipment failure and migration delays; Solve the Hamilton-Jacobi-Bellman equation to obtain the optimal control strategy , i.e., the optimal strategy for data migration between devices; Based on the optimal control strategy , dynamically adjust the data migration path between devices to ensure that the system always maintains optimal scheduling during device status changes.

8. The data migration scheduling method based on device anomaly detection according to claim 1 is characterized in that: The step of applying the obtained optimal migration scheduling strategy to the system in real time includes: The optimal migration scheduling strategy Transmit the data to the scheduling execution module to ensure that each device migrates data along the optimal path; Dynamically adjust the scheduling strategy based on the real-time monitored equipment health status and system load changes; During the data migration process, the bandwidth and device resource utilization are continuously monitored to ensure that the scheduling strategy always meets the bandwidth limit and device health status constraints in actual operation; Update the scheduling strategy in time according to the changes in equipment status to ensure that the system is always in the optimal operating state.

9. The data migration scheduling method based on device anomaly detection according to claim 1 is characterized in that: The steps of solving the optimization problem of data migration scheduling using dynamic programming include: Based on the optimal control strategy and scheduling objective function , discretize the scheduling problem into sub-problems at several moments; Through the dynamic programming algorithm, the optimal data migration amount between devices at each moment is gradually calculated , and consider the constraints of bandwidth and device health status.

10. A data migration scheduling system based on device anomaly detection, adopting a data migration scheduling method based on device anomaly detection as claimed in any one of claims 1 to 9, characterized in that: include: Health status monitoring module, which is used to monitor the health status of each device in real time and estimate the health status of the device through random process modeling; An anomaly detection module, used to determine whether the device is in an abnormal state based on the health status estimation value; The scheduling decision module is used to generate data migration scheduling strategies based on changes in device health status and bandwidth constraints; The optimization calculation module is used to calculate the optimal data migration path and scheduling strategy based on the Lagrange multiplier method and the stochastic optimal control method; The real-time scheduling execution module is used to adjust the data migration path and scheduling strategy in the system according to the optimization results.

Images