Data processing method and device based on gating circulation unit, equipment and medium

Abstract

The invention discloses a data processing method and device based on a gating circulation unit, equipment and a medium, and belongs to the technical field of data processing. The method comprises the following steps: acquiring system state data of a target server system from a preset file position in the target server system; preprocessing the system state data based on a trained gating cycle unit model to obtain target state data; inputting the target state data into the gating circulation unit model, and generating predicted performance data of the target server system by the gating circulation unit model based on the input target state data; and carrying out resource allocation on the target server system based on the predicted performance data. According to the embodiment of the invention, the system is dynamically adjusted according to the environment and the load.

Description

Technical Field

[0001] This application belongs to the field of data processing technology, specifically relating to a data processing method based on a gated loop unit, a data processing device based on a gated loop unit, an electronic device, and a readable storage medium. Background Technology

[0002] With the rapid development of cloud computing and big data applications, performance monitoring and optimization of cloud computing infrastructure have become increasingly important. In cloud computing environments, especially in scenarios involving large amounts of data read and write operations, the performance stability of cloud disks directly affects the overall service quality. Therefore, real-time monitoring and prediction of disk performance can effectively prevent service interruptions or degradation caused by performance bottlenecks.

[0003] Traditional performance monitoring methods often rely on static monitoring strategies, which cannot be dynamically adjusted according to changes in the environment and workload. Summary of the Invention

[0004] The purpose of this application is to provide a data processing method based on a gated loop unit, which can solve the problem that the current service system cannot dynamically adjust according to the environment and load.

[0005] To solve the above-mentioned technical problems, this application is implemented as follows: In a first aspect, embodiments of this application provide a data processing method based on a gated loop unit, the method comprising: Obtain the system status data of the target server system from a preset file location in the target server system; The system state data is preprocessed based on the trained gated recurrent unit model to obtain the target state data; The target state data is input into the gated loop unit model, and the gated loop unit model generates the predicted performance data of the target server system based on the input target state data. Resource allocation is performed on the target server system based on the predicted performance data.

[0006] Optionally, the resource allocation to the target server system based on the predicted performance data includes: Based on the predicted performance data, determine whether the target server system exhibits any abnormal behavior; When the target server system exhibits abnormal behavior, a preset adaptive strategy for the abnormal behavior is obtained; The target server system is allocated resources according to the adaptive strategy described above.

[0007] Optionally, the step of allocating resources to the target server system according to the adaptive strategy includes: The priority information of each task to be processed in the target server system is determined according to the adaptive strategy. The target task is determined from the tasks to be processed based on the priority information; Based on the predicted performance data, determine whether one or more nodes in the target server system used to process the target task are all abnormal nodes; When it is determined that not all one or more nodes in the target server system used to process the target task are abnormal nodes, the predicted load information of one or more nodes of the target task is determined based on the predicted performance data. Based on the predicted load information, resources are allocated to one or more nodes of the target task.

[0008] Optionally, it also includes: Obtain real-time performance data for the target server system; The gated loop unit model is adjusted based on the real-time performance data.

[0009] Optionally, determining whether there are any abnormal nodes in one or more nodes of the target server system used to process the target task based on the predicted performance data includes: Determine the range of data values ​​corresponding to the type of the predicted performance data; Determine whether the predicted performance data is within the range of the data values; When the predicted performance data is not within the range of the data value, one or more nodes in the target server system used to process the target task are determined to be abnormal nodes. When the predicted performance data is within the range of the data value, it is determined that one or more nodes in the target server system used to process the target task are not abnormal nodes.

[0010] Optionally, the preprocessing may include any of the following: Cleaning, normalization, and filling missing values.

[0011] Optionally, it also includes: If it is determined that one or more nodes in the target server system used to process the target task are abnormal nodes, the target task will be transferred to a preset waiting queue.

[0012] Secondly, embodiments of this application provide a data processing apparatus based on a gated loop unit, the apparatus comprising: The status data acquisition module is used to acquire system status data of the target server system from a preset file location in the target server system; The data preprocessing module is used to preprocess the system state data based on the trained gated recurrent unit model to obtain the target state data; The performance prediction module is used to input the target state data into the gated loop unit model, and the gated loop unit model generates predicted performance data of the target server system based on the input target state data. The resource allocation module is used to allocate resources to the target server system based on the predicted performance data.

[0013] Thirdly, embodiments of this application provide an electronic device including a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the method described in the first aspect.

[0014] Fourthly, embodiments of this application provide a readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method described in the first aspect.

[0015] Fifthly, embodiments of this application provide a chip, the chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run programs or instructions to implement the method as described in the first aspect.

[0016] In this embodiment, system status data of the target server system is obtained from a preset file location in the target server system; the system status data is preprocessed based on a trained gated recurrent unit model to obtain target status data; the target status data is input into the gated recurrent unit model, which generates predicted performance data of the target server system based on the input target status data; and resource allocation is performed on the target server system based on the predicted performance data. This embodiment uses a gated recurrent unit model for performance prediction, thereby achieving resource allocation and enabling dynamic system adjustment based on real-time environment and load. Attached Figure Description

[0017] Figure 1a This is a flowchart illustrating a data processing method based on a gated loop unit in an embodiment of this application; Figure 1b This is a schematic diagram of a GRU model training process in an embodiment of this application; Figure 1cThis is a flowchart illustrating an adaptive protection strategy in an embodiment of this application; Figure 1d This is a schematic diagram of a resource allocation process in an embodiment of this application; Figure 2 This is a flowchart illustrating a data processing method based on a gated loop unit in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a data processing device based on a gated loop unit in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0020] The data processing method based on a gated loop unit provided in this application will be described in detail below with reference to the accompanying drawings, through specific embodiments and application scenarios.

[0021] Reference Figure 1a This paper illustrates a data processing method based on a gated loop unit in an embodiment of this application. The method includes the following steps: Step S101: Obtain system status data of the target server system from a preset file location in the target server system; In practical applications, the target server system can be a system consisting of one or more servers capable of performing certain functions. The target server system can collect system status data from preset file locations such as log files, database records, and time-series databases. This system status data mainly includes system status data collected every minute, every second, or higher, uptime, request processing time, etc. This system status data is used to capture the trend of system changes over time and help analyze the system's performance under different loads.

[0022] Step S102: Preprocess the system state data based on the trained gated recurrent unit model to obtain the target state data; Among them, the Gated Recurrent Unit (GRU) is a common variant of the Recurrent Neural Network (RNN). The GRU excels at processing sequential data. It controls the flow of input information by introducing "update gates" and "reset gates," making it suitable for learning tasks involving long sequences.

[0023] The gated loop unit model in this embodiment is a model for predicting system performance data based on input system state data. The gated loop unit model can typically be trained using historically collected system state and performance data. Furthermore, the gated loop unit model in this embodiment can be updated in real-time based on the latest state and performance data of the server system.

[0024] In practical applications, the collected data often contains noise, missing values, or inconsistencies. At this stage, the system can perform preprocessing operations such as cleaning, normalization, and filling in missing values ​​to ensure data quality and make it suitable for subsequent model training. The preprocessing may include any of the following: Cleaning, normalization, and filling missing values.

[0025] Among them, cleaning can remove invalid data from system status data; normalization can convert all values ​​into data within a preset range to reduce the difference in data units; and filling in missing values ​​can increase the effective sample size.

[0026] Step S103: Input the target state data into the gated loop unit model, and generate the predicted performance data of the target server system based on the input target state data. After obtaining the target state data, the target state data can be sent to the gated recurrent unit model. In the gated recurrent unit model, the predicted performance data of the target server system can be generated based on the model parameters obtained through pre-training. The predicted performance data can include data such as hard disk response time and load conditions.

[0027] In practical applications, preprocessed data is used to train a GRU (Gated Recurrent Unit) model. GRU is a type of neural network suitable for processing time-series data, capable of capturing long-term dependencies. At this stage, the model learns patterns from historical data to predict future performance. The GRU model training process is as follows: Figure 1b As shown.

[0028] The core idea of ​​GRU is to selectively retain historical information and forget irrelevant information by introducing a "gate" mechanism, thereby enabling the network to learn long-range dependencies. It achieves this goal through two key gating units: the reset gate and the update gate. Figure 1b The process described illustrates the internal workings of the GRU model, including how the reset and update gates help the model balance historical information with current new inputs. Through continuous training and weight updates, the GRU model can gradually improve its accuracy in predicting future performance.

[0029] This application's embodiments utilize a designed GRU neural network, employing reset and update gates to achieve a balance between long-term dependencies and current information, effectively handling long- and short-term dependencies in time series data. This improves performance prediction accuracy; by introducing the GRU model, the prediction accuracy for system performance is significantly enhanced.

[0030] Meanwhile, the GRU model can predict system performance metrics (such as CPU utilization and network latency) and detect potential performance problems in a timely manner by identifying abnormal fluctuations in the prediction results. Utilizing the anomaly detection mechanism in the prediction results, the system can promptly identify performance issues, reduce system crashes or failures, and improve system reliability.

[0031] Step S104: Allocate resources to the target server system based on the predicted performance data.

[0032] After obtaining the predicted performance data, the predicted performance data can be based on the current state of the target server. Then, the resources in the target server can be reasonably allocated based on the predicted performance data so that the target server can perform the tasks to be processed with better performance.

[0033] In one embodiment of this application, the step of allocating resources to the target server system based on the predicted performance data includes: determining whether the target server system exhibits abnormal behavior based on the predicted performance data; when the target server system exhibits abnormal behavior, obtaining a preset adaptive strategy for the abnormal behavior; and allocating resources to the target server system according to the adaptive strategy.

[0034] In practical applications, after obtaining the predicted performance data, it is possible to determine whether the target server exhibits abnormal behavior based on the predicted performance data. Abnormal behavior can be behavior that exceeds or deviates from the expected behavior of the target server. Specifically, it can be abnormal behavior that is predefined in the target server and associated with the predicted performance data.

[0035] If abnormal behavior is detected, the adaptive policy in the target server system can be triggered. The adaptive policy can be a pre-configured policy in the target server system used to dynamically adjust resource allocation.

[0036] In one embodiment of this application, the resource allocation to the target server system according to the adaptive strategy includes the following sub-steps: Sub-step S11: Determine the priority information of each task to be processed in the target server system according to the adaptive strategy; In practical applications, after determining the adaptive strategy, the adaptive strategy can be a resource priority adjustment strategy, that is, resource adjustment can be achieved according to priority. Then the target server system can determine the priority of each task to be processed. The task to be processed is the case that needs to be processed but has not been processed in the target server system. The priority information of the task can be preset, and there is no restriction on this in the embodiments of this application.

[0037] In this embodiment, the system dynamically adjusts its resource allocation strategy based on prediction results to optimize system performance. It can also utilize feedback loops to continuously train and improve the model. Through the adaptive resource allocation adjustment mechanism, the system can dynamically optimize resource usage, avoiding waste while ensuring performance efficiency. The feedback loop mechanism allows the system to continuously acquire the latest performance data for training and updating the model, maintaining the effectiveness and real-time nature of performance prediction and adjustment strategies.

[0038] Sub-step S12: Determine the target task from the tasks to be processed based on the priority information; After determining the priority information, multiple tasks to be processed in the queue can be sorted from high to low according to the priority information, and then the target task can be selected according to the sorting. The target task can be the task with the highest priority at present.

[0039] In this embodiment of the application, the target task is determined according to priority, which can ensure that important tasks (i.e., high-priority tasks) can be processed first in the event of abnormal behavior, thus ensuring that important tasks can be carried out smoothly.

[0040] Sub-step S13: Based on the predicted performance data, determine whether one or more nodes in the target server system used to process the target task are all abnormal nodes; In practical applications, predictive performance data can also be used to determine whether one or more nodes in the target server system that are used to process the target task are abnormal nodes. Specifically, it can be determined whether all nodes are abnormal nodes based on the predicted load data of each node in the predictive performance data.

[0041] Sub-step S14: When it is determined that not all one or more nodes in the target server system used to process the target task are abnormal nodes, the predicted load information of one or more nodes of the target task is determined based on the predicted performance data. When all nodes are abnormal, all nodes are overloaded and cannot be rebalanced. However, when not all nodes are abnormal (i.e., some nodes are abnormal or all nodes are normal), load balancing can be performed based on the predicted load information of each node. This allows each node to reach its expected operating state and improves its data processing efficiency. For example, some tasks from a currently high-load node can be migrated to a low-load node, thereby reducing the load on the high-load node and transitioning it from an abnormal state to a normal state.

[0042] Sub-step S15: Allocate resources to one or more nodes of the target task based on the predicted load information.

[0043] Once the predicted load information is determined, resources can be allocated in advance based on the predicted load information to avoid some nodes becoming overloaded and causing abnormal behavior in the future.

[0044] For example, the current highest priority target task includes task node A, task node B, and task node C. If, based on the predicted load information and the current resource allocation, it is predicted that task node A and task node B will be overloaded, while task node C will be able to operate normally, then some of the resources allocated to task node C can be pre-allocated to task node A and task node B, thereby ensuring that task node A, task node B, and task node C can all operate normally.

[0045] The resource allocation method can be determined based on the specific predicted load information of each node. The ultimate goal of resource allocation is to ensure that each node can execute tasks within its load range and avoid becoming an abnormal node.

[0046] In the embodiments of this application, the core idea of ​​the adaptive guarantee strategy is to optimize the resource allocation process through feedback mechanisms and dynamic adjustments. In practical applications, traditional allocation is usually random, and the Cinder system's node allocation is more focused on stability, with insufficient consideration for performance. When the system detects an anomaly, it dynamically adjusts the resource allocation strategy based on the current performance status. This means the system decides when and how to perform tasks to avoid system overload or resource waste. Since the core function of the adaptive performance guarantee strategy is to dynamically allocate and schedule resources based on real-time feedback, the strategy that can be used in this embodiment is resource priority adjustment: when the target server system resources are limited, the adaptive strategy can prioritize allocating resources to high-priority tasks, thereby ensuring the performance guarantee of critical tasks.

[0047] The specific process is as follows: Figure 1c As shown, when a new task is available, it can be arranged in the waiting queue. This allows us to determine whether each node fails to meet the performance metrics. If each node fails to meet the performance metrics (i.e., all nodes are abnormal), it continues to wait in the waiting queue. If all nodes meet the performance metrics or partially meet them, the nodes can be sorted from low to high based on the predicted load results from the training model, thus deriving the node allocation strategy.

[0048] In one embodiment of this application, the system can also optimize the allocation of resource nodes based on performance prediction and guarantee strategies. The weights of each resource allocation strategy are adjusted based on the accuracy of historical predictions to improve response flexibility. When allocating nodes, both node performance prediction results and specific dynamic resource optimization strategies need to be considered simultaneously, adjusting task node load and dynamic weights to ensure that the performance of each server node exceeds a preset standard. In one embodiment of this application, it is possible to determine whether one or more nodes in the target server system used to process the target task are abnormal nodes based on the predicted performance data, including: determining the data value range corresponding to the type of the predicted performance data; determining whether the predicted performance data is within the data value range; when the predicted performance data is not within the data value range, determining that one or more nodes in the target server system used to process the target task are abnormal nodes; when the predicted performance data is within the data value range, determining that one or more nodes in the target server system used to process the target task are not abnormal nodes.

[0049] Based on performance prediction, the target server system will perform anomaly detection. If abnormal behavior is detected that deviates from the preset normal range of the target server's performance (such as a sudden drop in performance or abnormal fluctuations), the target server system will mark these anomalies and notify the target server system's maintenance personnel that there may be a problem with the current target server, so that the maintenance personnel of the target server system can perform maintenance in a timely manner.

[0050] In one embodiment of this application, when it is determined that one or more nodes in the target server system used to process the target task are all abnormal nodes, each node is in an abnormal state. Therefore, all nodes are unable to perform dynamic resource allocation. Consequently, the target task can be transferred to a preset waiting queue for processing. The system can continuously monitor the nodes of the target server. When it detects that some or all nodes of the target task are abnormal, the target task can be removed from the waiting queue. The target server system can then perform resource balancing scheduling and optimization based on one or more task nodes of the target task to ensure efficient completion of the target task. The waiting queue can be sorted according to the priority of the target tasks, with higher priority tasks being removed first and lower priority tasks being removed later.

[0051] In practical applications, when a target task is re-entered into the waiting queue, the current node may be too saturated to schedule resources in a short period of time. Therefore, a cooldown period can be set for the target task. During the cooldown period, the target task may not be allowed to leave the queue, and the tasks following the target task in the waiting queue will be processed first. This avoids low task processing efficiency due to repeated processing of the target task.

[0052] Reference Figure 1d Here is a schematic diagram of a resource allocation process in one embodiment of this application, including the following steps: (1) Data collection: The system collects data from log files, database records, and time-series databases. This data mainly includes system status data collected every minute, every second, or at a higher frequency, uptime, request processing time, etc., which are used to capture the trend of system changes over time and help analyze the system's performance under different loads.

[0053] (2) Preprocess the collected data; (3) Train the GPU model using historical information and preprocessed data; (4) Performance prediction is performed using a trained GPU model; (5) Allocate resources based on performance predictions to optimize resource results; (6) Perform anomaly detection based on performance prediction results, and trigger an adaptive protection strategy when anomalies are found.

[0054] (7) Based on the trigger-based adaptive guarantee strategy, resource allocation is performed to obtain optimized resource results.

[0055] In this embodiment, system state data of the target server system is obtained from a preset file location in the target server system; the system state data is preprocessed based on a trained gated recurrent unit model to obtain target state data; the target state data is input into the gated recurrent unit model, which generates predicted performance data of the target server system based on the input target state data; and resource allocation is performed on the target server system based on the predicted performance data. This embodiment uses a gated recurrent unit model for performance prediction, thereby achieving resource allocation and enabling dynamic system adjustment based on environment and load.

[0056] Reference Figure 2 This paper illustrates a data processing method based on a gated loop unit in an embodiment of this application. The method includes the following steps: Step S201: Obtain the system status data of the target server system from a preset file location in the target server system; In practical applications, the target server system can be a system consisting of one or more servers capable of performing certain functions. The target server system can collect system status data from preset file locations such as log files, database records, and time-series databases. This system status data mainly includes system status data collected every minute, every second, or higher, uptime, request processing time, etc. This system status data is used to capture the trend of system changes over time and help analyze the system's performance under different loads.

[0057] Step S202: Preprocess the system state data based on the trained gated recurrent unit model to obtain the target state data; In practical applications, the target server system can be a system consisting of one or more servers capable of performing certain functions. The target server system can collect system status data from preset file locations such as log files, database records, and time-series databases. This system status data mainly includes system status data collected every minute, every second, or higher, uptime, request processing time, etc. This system status data is used to capture the trend of system changes over time and help analyze the system's performance under different loads.

[0058] Step S102: Preprocess the system state data based on the trained gated recurrent unit model to obtain the target state data; Among them, the Gated Recurrent Unit (GRU) is a common variant of the Recurrent Neural Network (RNN). The GRU excels at processing sequential data. It controls the flow of input information by introducing "update gates" and "reset gates," making it suitable for learning tasks involving long sequences.

[0059] The gated loop unit model in this embodiment is a model for predicting system performance data based on input system state data. The gated loop unit model can typically be trained using historically collected system state and performance data. Furthermore, the gated loop unit model in this embodiment can be updated in real-time based on the latest state and performance data of the server system.

[0060] In practical applications, the collected data often contains noise, missing values, or inconsistencies. At this stage, the system can perform preprocessing operations such as cleaning, normalization, and filling in missing values ​​to ensure data quality and make it suitable for subsequent model training. The preprocessing may include any of the following: Cleaning, normalization, and filling missing values.

[0061] Among them, cleaning can remove invalid data from system status data; normalization can convert all values ​​into data within a preset range to reduce the difference in data units; and filling in missing values ​​can increase the effective sample size.

[0062] Step S203: Input the target state data into the gated loop unit model, and generate the predicted performance data of the target server system based on the input target state data. After obtaining the target state data, the target state data can be sent to the gated recurrent unit model. In the gated recurrent unit model, the predicted performance data of the target server system can be generated based on the model parameters obtained through pre-training. The predicted performance data can include data such as hard disk response time and load conditions.

[0063] In practical applications, preprocessed data is used to train a GRU (Gated Recurrent Unit) model. GRU is a type of neural network suitable for processing time-series data, capable of capturing long-term dependencies. At this stage, the model learns patterns from historical data to predict future performance. The GRU model training process is as follows: Figure 1b As shown.

[0064] The core idea of ​​GRU is to selectively retain historical information and forget irrelevant information by introducing a "gate" mechanism, thereby enabling the network to learn long-range dependencies. It achieves this goal through two key gating units: the reset gate and the update gate. Figure 1b The process described illustrates the internal workings of the GRU model, including how the reset and update gates help the model balance historical information with current new inputs. Through continuous training and weight updates, the GRU model can gradually improve its accuracy in predicting future performance.

[0065] This application's embodiments utilize a designed GRU neural network, employing reset and update gates to achieve a balance between long-term dependencies and current information, effectively handling long- and short-term dependencies in time series data. This improves performance prediction accuracy; by introducing the GRU model, the prediction accuracy for system performance is significantly enhanced.

[0066] Meanwhile, the GRU model can predict system performance metrics (such as CPU utilization and network latency) and detect potential performance problems in a timely manner by identifying abnormal fluctuations in the prediction results. Utilizing the anomaly detection mechanism in the prediction results, the system can promptly identify performance issues, reduce system crashes or failures, and improve system reliability.

[0067] Step S204: Allocate resources to the target server system based on the predicted performance data.

[0068] After obtaining the predicted performance data, the predicted performance data can be based on the current state of the target server. Then, the resources in the target server can be reasonably allocated based on the predicted performance data so that the target server can perform the tasks to be processed with better performance.

[0069] In one embodiment of this application, the step of allocating resources to the target server system based on the predicted performance data includes: determining whether the target server system exhibits abnormal behavior based on the predicted performance data; when the target server system exhibits abnormal behavior, obtaining a preset adaptive strategy for the abnormal behavior; and allocating resources to the target server system according to the adaptive strategy.

[0070] In practical applications, after obtaining the predicted performance data, it is possible to determine whether the target server exhibits abnormal behavior based on the predicted performance data. Abnormal behavior can be behavior that exceeds or deviates from the expected behavior of the target server. Specifically, it can be abnormal behavior that is predefined in the target server and associated with the predicted performance data.

[0071] If abnormal behavior is detected, the adaptive policy in the target server system can be triggered. The adaptive policy can be a pre-configured policy in the target server system used to dynamically adjust resource allocation.

[0072] Step S205: Obtain real-time performance data for the target server system; In practical applications, the target server system can obtain real-time performance data of the target server system, and adjust the gated cyclic unit model based on the new real-time performance data feedback to improve the prediction accuracy of the gated cyclic unit model.

[0073] Step S206: Adjust the gated loop unit model according to the real-time performance data.

[0074] This application provides a continuous feedback loop. The system continuously updates and adjusts its model and strategies, and optimizes them based on real-time data, forming an adaptive resource management mechanism to ensure performance and stability.

[0075] In this embodiment, system state data of the target server system is obtained from a preset file location in the target server system; the system state data is preprocessed based on a trained gated recurrent unit model to obtain target state data; the target state data is input into the gated recurrent unit model, which generates predicted performance data of the target server system based on the input target state data; and resource allocation is performed on the target server system based on the predicted performance data. This embodiment uses a gated recurrent unit model for performance prediction, thereby achieving resource allocation and enabling dynamic system adjustment based on environment and load.

[0076] Reference Figure 3 The diagram shows a flowchart of a data processing device based on a gated loop unit according to an embodiment of this application. The device may include the following modules: The status data acquisition module 301 is used to acquire system status data of the target server system from a preset file location in the target server system; Data preprocessing module 302 is used to preprocess the system state data based on the trained gated recurrent unit model to obtain target state data; The performance prediction module 303 is used to input the target state data into the gated loop unit model, and the gated loop unit model generates predicted performance data of the target server system based on the input target state data. The resource allocation module 304 is used to allocate resources to the target server system based on the predicted performance data.

[0077] In this embodiment of the application, the resource allocation module 304 may include: An abnormal behavior judgment submodule is used to determine whether the target server system exhibits abnormal behavior based on the predicted performance data. The adaptive strategy acquisition submodule is used to acquire a preset adaptive strategy for the abnormal behavior when the target server system exhibits abnormal behavior. The resource allocation submodule is used to allocate resources to the target server system according to the adaptive strategy.

[0078] In one embodiment of this application, the resource allocation submodule may include: The priority information determination unit is used to determine the priority information of each task to be processed in the target server system according to the adaptive strategy. A target task determination unit is used to determine a target task from the tasks to be processed based on the priority information; An abnormal node determination unit is used to determine, based on the predicted performance data, whether one or more nodes in the target server system used to process the target task are all abnormal nodes. The predicted load information determination unit is used to determine the predicted load information of one or more nodes of the target task based on the predicted performance data when it is determined that not all of the nodes in the target server system used to process the target task are abnormal nodes. The resource allocation unit is used to allocate resources to one or more nodes of the target task based on the predicted load information.

[0079] In one embodiment of this application, the apparatus may further include: A real-time performance data acquisition module is used to acquire real-time performance data for the target server system. The model adjustment module is used to adjust the gated loop unit model according to the real-time performance data.

[0080] In one embodiment of this application, the abnormal node determination unit may include: The data value range determination subunit is used to determine the data value range corresponding to the type of the prediction performance data; A judgment subunit is used to determine whether the predicted performance data is within the range of the data value; An abnormal node determination subunit is used to determine one or more nodes in the target server system used to process the target task as abnormal nodes when the predicted performance data is not within the range of the data value. The non-abnormal node determination subunit is used to determine one or more nodes in the target server system used to process the target task as non-abnormal nodes when the predicted performance data is within the range of the data value.

[0081] In one embodiment of this application, the preprocessing may include any of the following: Cleaning, normalization, and filling missing values.

[0082] In one embodiment of this application, the apparatus further includes: The waiting queue module is used to transfer the target task to a preset waiting queue when it is determined that one or more nodes in the target server system used to process the target task are abnormal nodes.

[0083] In this embodiment, system state data of the target server system is obtained from a preset file location in the target server system; the system state data is preprocessed based on a trained gated recurrent unit model to obtain target state data; the target state data is input into the gated recurrent unit model, which generates predicted performance data of the target server system based on the input target state data; and resource allocation is performed on the target server system based on the predicted performance data. This embodiment uses a gated recurrent unit model for performance prediction, thereby achieving resource allocation and enabling dynamic system adjustment based on environment and load.

[0084] It should be noted that the data processing method based on a gated loop unit provided in this application embodiment can be executed by a data processing device based on a gated loop unit, or by a control module within that device for executing the data processing method based on a gated loop unit. This application embodiment uses the execution of the data processing method based on a gated loop unit by a data processing device as an example to illustrate the data processing method based on a gated loop unit provided in this application embodiment.

[0085] The data processing device based on the gated loop unit in this application embodiment can be a device, or a component, integrated circuit, or chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. For example, mobile electronic devices can be mobile phones, tablets, laptops, PDAs, in-vehicle electronic devices, wearable devices, ultra-mobile personal computers (UMPCs), netbooks, or personal digital assistants (PDAs), etc., while non-mobile electronic devices can be servers, network-attached storage (NAS), personal computers (PCs), televisions (TVs), ATMs, or self-service machines, etc. This application embodiment does not impose specific limitations.

[0086] The data processing device based on the gated loop unit in this application embodiment can be a device with an operating system. This operating system can be Android, iOS, or other possible operating systems; this application embodiment does not specifically limit it.

[0087] The data processing device based on a gated loop unit provided in this application embodiment can achieve... Figures 1a to 3 The various processes implemented by the data processing method based on the gated loop unit in the method embodiment will not be described again here to avoid repetition.

[0088] Optionally, this application embodiment also provides an electronic device, including a processor 1010, a memory 1009, and a program or instructions stored in the memory 1009 and executable on the processor 1010. When the program or instructions are executed by the processor 1010, they implement the various processes of the above-described data processing method embodiment based on the gated loop unit and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0089] It should be noted that the electronic devices in the embodiments of this application include the mobile electronic devices and non-mobile electronic devices described above.

[0090] Figure 4 A schematic diagram of the hardware structure of an electronic device to implement an embodiment of this application. The electronic device 1000 includes, but is not limited to, components such as: a radio frequency unit 1001, a network module 1002, an audio output unit 1003, an input unit 1004, a sensor 1005, a display unit 1006, a user input unit 1007, an interface unit 1008, a memory 1009, and a processor 1010. The memory 1009 includes applications and an operating system; the user input unit 1007 may include a touch panel 10071 and other input devices 100072; the input unit 1004 may include an image processor 10041 and a microphone 10042; and the display unit 1006 may include a display panel 10061.

[0091] Those skilled in the art will understand that the electronic device 1000 may also include a power supply (such as a battery) for supplying power to various components. The power supply may be logically connected to the processor 1010 through a power management system, thereby enabling functions such as managing charging, discharging, and power consumption through the power management system. Figure 4 The electronic device structure shown does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown, or combine certain components, or have different component arrangements, which will not be elaborated here. This application also provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of the above-described data processing method embodiment based on a gated loop unit and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0092] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0093] This application also provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the various processes of the above-described data processing method embodiment based on gated loop unit, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0094] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0095] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0096] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0097] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A data processing method based on a gated loop unit, characterized in that, The method includes: Obtain the system status data of the target server system from a preset file location in the target server system; The system state data is preprocessed based on the trained gated recurrent unit model to obtain the target state data; The target state data is input into the gated loop unit model, and the gated loop unit model generates the predicted performance data of the target server system based on the input target state data. Resource allocation is performed on the target server system based on the predicted performance data.

2. The method according to claim 1, characterized in that, The resource allocation to the target server system based on the predicted performance data includes: Based on the predicted performance data, determine whether the target server system exhibits any abnormal behavior; When the target server system exhibits abnormal behavior, a preset adaptive strategy for the abnormal behavior is obtained; The target server system is allocated resources according to the adaptive strategy described above.

3. The method according to claim 2, characterized in that, The step of allocating resources to the target server system according to the adaptive strategy includes: The priority information of each task to be processed in the target server system is determined according to the adaptive strategy. The target task is determined from the tasks to be processed based on the priority information; Based on the predicted performance data, determine whether one or more nodes in the target server system used to process the target task are all abnormal nodes; When it is determined that not all one or more nodes in the target server system used to process the target task are abnormal nodes, the predicted load information of one or more nodes of the target task is determined based on the predicted performance data. Based on the predicted load information, resources are allocated to one or more nodes of the target task.

4. The method according to claim 1, characterized in that, Also includes: Obtain real-time performance data for the target server system; The gated loop unit model is adjusted based on the real-time performance data.

5. The method according to claim 3, characterized in that, Based on the predicted performance data, determine whether one or more nodes in the target server system used to process the target task are abnormal, including: Determine the range of data values ​​corresponding to the type of the predicted performance data; Determine whether the predicted performance data is within the range of the data values; When the predicted performance data is not within the range of the data value, one or more nodes in the target server system used to process the target task are determined to be abnormal nodes. When the predicted performance data is within the range of the data value, it is determined that one or more nodes in the target server system used to process the target task are not abnormal nodes.

6. The method according to claim 1, characterized in that, The preprocessing may include any of the following: Cleaning, normalization, and filling missing values.

7. The method according to claim 3, characterized in that, Also includes: If it is determined that one or more nodes in the target server system used to process the target task are abnormal nodes, the target task will be transferred to a preset waiting queue.

8. A data processing device based on a gated loop unit, characterized in that, The device includes: The status data acquisition module is used to acquire system status data of the target server system from a preset file location in the target server system; The data preprocessing module is used to preprocess the system state data based on the trained gated recurrent unit model to obtain the target state data; The performance prediction module is used to input the target state data into the gated loop unit model, and the gated loop unit model generates predicted performance data of the target server system based on the input target state data. The resource allocation module is used to allocate resources to the target server system based on the predicted performance data.

9. An electronic device, characterized in that, It includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein when the program or instructions are executed by the processor, they implement the steps of the data processing method based on a gated loop unit as described in claims 1-7.

10. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions, which, when executed by a processor, implement the steps of the data processing method based on a gated loop unit as described in claims 1-7.

Images