Method and device for processing RapidIO protocol during fault injection and electronic equipment

Abstract

The invention relates to a processing method and device for a RapidIO protocol during fault injection and electronic equipment. The method comprises the following steps: acquiring performance index data of each node in a RapidIO network; analyzing the performance index data, and outputting a fault prediction result; setting multiple levels of fault thresholds according to the fault prediction result, dividing different levels according to different types of faults, and setting corresponding multiple levels of initial fault thresholds; adaptively adjusting the fault threshold value of each node based on the fault prediction result and the performance index data monitored in real time; and instructing the system to implement corresponding processing measures according to the fault level. According to the method, by fusing dynamic adjustment of the fault threshold and machine learning prediction, the system reliability during fault injection is improved conveniently, the timeliness and accuracy of fault processing are improved, the method adapts to a dynamically changing network environment, and the efficiency and flexibility of fault processing are improved through a stage processing mechanism.

Description

Technical Field

[0001] The present invention relates to the field of embedded systems, and in particular to a method, a device and an electronic device for processing a RapidIO protocol during fault injection. Background Art

[0002] The RapidIO protocol is widely used in high-performance computing and embedded systems. With its high bandwidth and low latency characteristics, it is an ideal choice for mission-critical applications. However, in the case of fault injection, the reliability of the system may be seriously affected. Currently, static threshold settings are mostly used to deal with such problems. This approach cannot adapt to the dynamically changing network environment, resulting in low fault handling efficiency and untimely handling when the error rate is high. In addition, there is a lack of effective fault prediction mechanism, making it difficult to take preventive measures in time. In addition, the fault handling mechanism is single and cannot be graded according to the severity of the fault. Therefore, it needs to be improved. Summary of the invention

[0003] Based on this, it is necessary to provide a RapidIO protocol processing method, device and electronic device during fault injection in order to address the problem that the traditional static threshold setting method is difficult to adapt to the dynamically changing network environment, resulting in inefficient fault handling, and prone to untimely handling when the error rate is high. In addition, there is a lack of effective fault prediction mechanism, making it difficult to take preventive measures in a timely manner. In addition, the fault handling mechanism is single and cannot be graded according to the severity of the fault.

[0004] The present invention provides a method for processing a RapidIO protocol during fault injection, the method comprising: Obtaining performance indicator data of each node in the RapidIO network, the performance indicator data including real-time monitored node signal strength indicators, error rate statistics, and data packet loss records; Analyze performance indicator data and output fault prediction results; According to the fault prediction results, multi-level fault thresholds are set, and different levels are divided according to different types of faults, and corresponding multi-level initial fault thresholds are set; Adaptively adjust the fault threshold of each node based on fault prediction results and real-time monitoring performance indicator data; According to the fault level instruction system implements corresponding treatment measures.

[0005] In one embodiment, analyzing the performance indicator data includes: Clean and format the acquired performance indicator data to obtain pre-processed data; Features useful for fault prediction are extracted from the preprocessed data, wherein the features include a change trend of signal strength, a change trend of error rate, and system load.

[0006] In one embodiment, the outputting of the fault prediction result includes: Obtain preprocessed performance indicator data; The performance indicator data is input into a machine learning model, and a fault prediction result is output. The machine learning model is trained using the performance indicator data of each node in the RapidIO network and a fault probability sample as training data.

[0007] In one embodiment, the machine learning model is a random forest model, a support vector machine model or a recurrent neural network model.

[0008] In one of the embodiments, the machine learning model is optimized using a grid search method or a Bayesian optimization method, and the prediction performance is evaluated using a cross-validation technique.

[0009] In one embodiment, the multi-level fault threshold is set by analyzing historical fault data and combining industry standards, the fault levels include minor faults, moderate faults and severe faults, and the threshold corresponding to each fault level is divided according to the error rate of the node.

[0010] In one embodiment, the instructing the system to implement corresponding processing measures according to the fault level includes: In response to the fault level being a minor fault, triggering a data packet retransmission mechanism; In response to the fault level being a medium fault, a load balancing mechanism is started and the data transmission task of the faulty node is allocated to a healthy node; In response to the fault level being a serious fault, an alarm message is sent to an administrator and the fault event is recorded.

[0011] The present invention also provides a RapidIO protocol processing device during fault injection, comprising: An acquisition module is used to acquire performance indicator data of each node in the RapidIO network, wherein the performance indicator data includes a real-time monitored node signal strength indicator, error rate statistics, and data packet loss records; Analysis module, used to analyze performance indicator data and output fault prediction results; A setting module, used to set multi-level fault thresholds according to the fault prediction results, divide different faults into different levels according to different types of faults, and set corresponding multi-level initial fault thresholds; An adjustment module is used to adaptively adjust the fault threshold of each node based on the fault prediction results and real-time monitoring performance indicator data; The instruction module is used to instruct the system to implement corresponding processing measures according to the fault level.

[0012] The present invention also provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the RapidIO protocol processing method during fault injection as described in any one of the above is implemented.

[0013] The present invention also provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the method for processing a RapidIO protocol during fault injection as described above is implemented.

[0014] The above-mentioned RapidIO protocol processing method, device and electronic device during fault injection can adapt to the dynamically changing network environment by dynamically adjusting the fault threshold, which is convenient for timely processing when the error rate is high. By analyzing the performance indicator data for fault prediction, it is convenient to take preventive measures in time, improve the system reliability during fault injection, improve the timeliness and accuracy of fault processing, and improve the efficiency and flexibility of fault processing through the mechanism of graded fault processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 It is a flowchart of a method for processing a RapidIO protocol during fault injection in an embodiment; Figure 2 A schematic diagram of a performance indicator data flow chart in one embodiment; Figure 3 A schematic diagram of a process flow of outputting a fault prediction result in one embodiment; Figure 4 A schematic diagram of a process flow of different fault level processing measures in one embodiment; Figure 5 A fault classification processing logic diagram in an embodiment; Figure 6 is a schematic diagram of the structure of a processing device of the RapidIO protocol during fault injection in an embodiment; Figure 7 FIG. 4 is a diagram showing the internal structure of an electronic device in one embodiment. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] Combine the following Figure 1-Figure 7 The invention describes a method, device and electronic device for processing a RapidIO protocol during fault injection.

[0019] like Figure 1 As shown, in one embodiment, a method for processing a RapidIO protocol during fault injection includes the following steps: Step S110, obtaining performance indicator data of each node in the RapidIO network, the performance indicator data including real-time monitored node signal strength indicators, error rate statistics, and data packet loss records.

[0020] By installing dedicated monitoring modules on each RapidIO node, these modules can accurately track the signal strength, error frequency, and packet loss of the node to ensure the real-time and accuracy of the data. In this way, detailed performance data of the system during the entire operation cycle can be collected, providing a solid data foundation for subsequent fault prediction and processing. The following are the detailed implementation steps of this process: 1. Deploy the monitoring module: Install the monitoring module on each RapidIO node to ensure that the module is compatible with the node and works properly.

[0021] 2. Data collection: After the monitoring module is started, the signal strength, error rate and transmission status of the node are collected in real time.

[0022] 3. Data recording: The collected data is recorded in real time in the local database for subsequent analysis. Implementation

[0023] First, technicians need to ensure the compatibility of the monitoring module with the RapidIO node and perform physical installation.

[0024] For example, standard interface design: Ensure that monitoring modules and RapidIO nodes use standardized interface design, which can reduce compatibility issues caused by interface mismatches.

[0025] Strict compliance with specifications: When designing the monitoring module, the specifications and standards of RapidIO technology are strictly followed to ensure the consistency of hardware and software specifications.

[0026] Compatibility testing: During the development process, full compatibility testing is performed, including unit testing, integration testing, and system testing, to ensure that the monitoring module can work properly on RapidIO nodes of different versions.

[0027] Secondly, after the installation is complete, start the monitoring module, which will automatically start collecting node performance data.

[0028] Finally, the collected data includes but is not limited to signal strength indicators, error rate statistics, and packet loss records, which will be transmitted to the database in real time for storage and analysis.

[0029] Step S120: Analyze the performance indicator data and output the fault prediction result.

[0030] Fault prediction based on machine learning models uses historical data to train machine learning models, analyzes real-time monitoring data, and predicts the possibility of faults.

[0031] Step S130, setting a multi-level fault threshold according to the fault prediction result, dividing different levels according to different types of faults, and setting corresponding multi-level initial fault thresholds.

[0032] By classifying possible faults according to the results of fault prediction, refined management of fault response can be achieved.

[0033] Step S140 , adaptively adjusting the fault threshold of each node based on the fault prediction result and the real-time monitored performance indicator data.

[0034] The system's responsiveness to fault injection is further improved by adopting a mechanism to dynamically adjust the fault threshold. Specifically, this step involves using real-time monitoring data and the prediction results of the machine learning model to adaptively adjust the fault threshold of each node. This adjustment mechanism allows the system to respond more flexibly to dynamic changes in the network environment.

[0035] To dynamically adjust the fault threshold: Based on continuous system monitoring and machine learning model predictions, we implement the following steps to dynamically adjust the fault threshold: 1. Real-time data analysis: The system will analyze the monitoring data of each node in real time, including but not limited to key indicators such as error rate and signal strength, and compare them with the current fault threshold.

[0036] 2. Threshold adjustment strategy: Based on the comparison results and the predicted output of the machine learning model, the system will decide whether the fault threshold needs to be adjusted.

[0037] The adjustment strategy is based on the following logic: If the prediction results show that the risk of failure is increasing and the current threshold does not reflect this change, the system will increase the threshold to increase sensitivity to potential failures.

[0038] Conversely, if the prediction results show that the risk of failure is reduced, the system will lower the threshold to reduce false alarms and improve the operating efficiency of the system.

[0039] The following are the implementation details of this paragraph: Implementation

[0040] The system compares real-time monitoring data with pre-set fault thresholds. This comparison is performed at fixed intervals to ensure timeliness.

[0041] When a deviation is found between the real-time data and the threshold, the system will adjust the threshold according to the following formula: , in, is the failure threshold at time t, is the fault threshold at the previous time point t-1, is the actual failure rate at time t, is the predicted failure probability at time t, is the learning rate, which controls how fast the threshold is adjusted.

[0042] Use the following formula to determine the failure level: , The following is a specific embodiment: Assume that the current error rate threshold of a node is 3%, and the machine learning model predicts that the error rate may rise to 4%.

[0043] After the system detects this change, it adjusts the threshold to 4.5% based on the formula.

[0044] This adjustment will enable the system to identify potential failures earlier and take appropriate measures.

[0045] Through this dynamic adjustment mechanism, the system can adapt to changes in the network environment more intelligently, thereby improving the accuracy and efficiency of fault handling.

[0046] Step S150, instructing the system to implement corresponding processing measures according to the fault level.

[0047] The RapidIO protocol's method for handling fault injection can adapt to dynamically changing network environments by dynamically adjusting fault thresholds, making it convenient to handle errors in a timely manner when the error rate is high. It also facilitates taking preventive measures in a timely manner by analyzing performance indicator data to predict faults, thereby improving system reliability during fault injection, improving the timeliness and accuracy of fault handling, and improving the efficiency and flexibility of fault handling through a mechanism for hierarchical fault handling.

[0048] In this embodiment, see Figure 2 , analyze performance indicator data, including: Step S121, cleaning and formatting the acquired performance indicator data to obtain pre-processed data.

[0049] Data collection: Collect historical data, including but not limited to system logs, performance indicators, fault records, etc.

[0050] Data cleaning: remove outliers, fill in missing values, eliminate duplicate data, and ensure data accuracy.

[0051] Data transformation: Convert time series data into a format suitable for machine learning models, for example, converting timestamps into periodic features.

[0052] Step S122, extracting features that are helpful for fault prediction from the preprocessed data, the features including the change trend of signal strength, the change trend of error rate and system load.

[0053] Parameter selection: Based on domain knowledge and data analysis, select features related to fault prediction, such as signal strength, error rate, system load, etc.

[0054] Feature engineering: Create new features or transform existing features to extract features that are helpful for fault prediction and enhance the predictive power of machine learning models. For example, use sliding windows to calculate features that are helpful for fault prediction.

[0055] In this embodiment, participating Figure 3 , output fault prediction results, including: Step S123, obtaining the pre-processed performance indicator data.

[0056] Step S124, inputting the performance indicator data into a machine learning model and outputting a fault prediction result, wherein the machine learning model is trained using the performance indicator data of each node in the RapidIO network and a fault occurrence probability sample as training data.

[0057] During the training process, the machine learning model continuously adjusts its internal parameters to achieve the best prediction results.

[0058] It should be noted that the machine learning model is a random forest model, a support vector machine model or a recurrent neural network model.

[0059] Among them, the random forest model is suitable for classification and regression problems; suitable for data with higher feature dimensions; able to handle situations where the data set contains missing values; and has no strict requirements for the distribution of the data set.

[0060] Support vector machine model: suitable for processing classification and regression problems, especially linearly inseparable data; suitable for data with low feature dimensions; requires data cleaning, removal or processing of missing values; relatively sensitive to the size of the data set, and may not perform as well as random forest on large data sets.

[0061] Recurrent Neural Networks: Particularly suitable for processing sequence data or time series data; suitable for situations where there is time dependency or sequence relationship between data; able to handle variable length input sequences; require a large amount of data to train to avoid overfitting.

[0062] Application examples: A recurrent neural network model is used for fault prediction, and its output can be expressed as: , in, is the predicted probability of failure occurrence by the recurrent neural network model, is the input feature vector, is the weight of the recurrent neural network model, is the i-th eigenvector of the input, is the i-th weight of the recurrent neural network model, n is the total number of input feature vectors and the total number of weights of the recurrent neural network model, b is the bias of the recurrent neural network model, is an activation function, such as the Sigmoid function.

[0063] Training process: Use the preprocessed data to train the machine learning model, including adjusting the hyperparameters of the machine learning model to achieve optimal performance.

[0064] Real-time data analysis: Real-time monitoring data is input into the trained machine learning model, which will output the probability prediction of failure occurrence.

[0065] Data input: The real-time monitoring data collected by the monitoring module is pre-processed and input into the trained machine learning model.

[0066] Prediction output: The machine learning model will output the probability of a failure occurring, as well as the possible failure type and time of occurrence.

[0067] In this embodiment, the machine learning model is optimized using a grid search method or a Bayesian optimization method, and the prediction performance is evaluated using a cross-validation technique.

[0068] Cross-validation: K-fold cross-validation is used to evaluate the generalization ability of the machine learning model to ensure that the machine learning model can make accurate predictions on unknown data.

[0069] Performance evaluation: using accuracy, recall, Score and other indicators are used to evaluate the performance of machine learning models.

[0070] Specific application examples: Technicians first need to preprocess historical data to remove outliers and fill missing values ​​so that the machine learning model can better learn data features.

[0071] Next, key features are extracted from the processed data, which will serve as important parameters for input into the machine learning model.

[0072] The machine learning model is trained using the preprocessed dataset. This process may require multiple iterations until the machine learning model achieves satisfactory prediction accuracy.

[0073] After training is completed, the prediction ability of the machine learning model is tested through the validation set to ensure that the machine learning model can accurately predict faults in actual applications.

[0074] Finally, the real-time monitoring data is input into the machine learning model, which predicts the possibility of failure at the current data point based on the learned knowledge, thereby providing a basis for subsequent fault threshold setting and fault handling.

[0075] In this embodiment, multi-level fault thresholds are set by analyzing historical fault data and combining industry standards. The fault levels include minor faults, moderate faults, and severe faults, and the threshold corresponding to each fault level is divided according to the error rate of the node.

[0076] For example, for a particular node, the error rate does not exceed The fault is minor, and the error rate is and The faults between are medium faults, and the error rate exceeds It is a serious fault.

[0077] Here is an example: Communication signal failure can be reflected by a variety of parameters. The following are some common parameters: Signal strength: refers to the received signal power level, usually used to determine whether the signal is attenuated or interrupted.

[0078] Signal-to-Noise Ratio (SNR): The ratio of signal power to noise power, used to assess signal quality.

[0079] Bit Error Rate (BER) or Error Rate: The proportion of errors that occur during data transmission, which can reflect the reliability of signal transmission.

[0080] Signal distortion: refers to the degree of change in the shape or waveform of a signal due to various reasons during transmission.

[0081] Signal delay: The time it takes for a signal to travel from the sender to the receiver. Excessive delay may be an indication of a malfunction.

[0082] Frequency offset: The deviation of a signal’s frequency from its original frequency, which may be caused by hardware problems or environmental interference.

[0083] Power instability: The degree of fluctuation of signal power. Instability may lead to a decrease in communication quality.

[0084] Modulation Error Rate (MER): A parameter that reflects the quality of the modulation signal and is used to measure the modulation quality of the received signal.

[0085] Eye closure: In digital communications, eye diagrams are used to observe the waveform of a signal in the time domain. The closure can reflect the stability and recognizability of the signal. Multi-level initial fault threshold setting: According to the fault level classification, a series of initial fault thresholds are set for each node. These thresholds are not fixed, but are dynamically adjusted according to the complexity of the network environment of the application scenario. Adjustment methods include but are not limited to: adjusting the upper and lower limits of the threshold based on historical data analysis, or making temporary corrections to the threshold according to real-time changes in the network environment.

[0086] The following is a detailed description of how this section is implemented: Implementation method: The technicians first need to determine the threshold of each fault level, which requires comprehensive consideration of the system's operating characteristics, historical fault data, and industry specifications; after determining the threshold, these thresholds are input into the monitoring system, which will monitor and judge the real-time data based on these thresholds; when the real-time data reaches or exceeds a certain threshold, the system will automatically identify the level of the fault and take corresponding processing measures. In order to adapt to the dynamic changes in the network environment, an adaptive adjustment mechanism is adopted, which can fine-tune the threshold according to the real-time data and prediction results to ensure the accuracy and timeliness of fault handling.

[0087] The following is a specific embodiment: Assume that the failure threshold is set , the fault threshold is .

[0088] If the error rate of a node rises to 2% during the monitoring period, the system identifies it as a moderate failure and triggers the corresponding retransmission mechanism.

[0089] If the error rate continues to rise to 6%, the system will upgrade it to a serious failure and take more stringent measures such as traffic redistribution.

[0090] In this way, we can ensure that the system can take the most appropriate response measures when facing different levels of failure, thereby improving the reliability and stability of the system.

[0091] In this embodiment, see Figure 4 , according to the fault level instruction system to implement corresponding processing measures, including: Step S151, in response to the fault level being a minor fault, triggering a data packet retransmission mechanism.

[0092] Step S152, in response to the fault level being a medium fault, starting a load balancing mechanism and allocating the data transmission task of the faulty node to a healthy node.

[0093] Step S153, in response to the fault level being a serious fault, sending an alarm message to an administrator and recording the fault event.

[0094] Retransmission mechanism for minor faults: When the monitoring system detects that the error rate of a node exceeds the preset minor fault threshold, the system will immediately start the data packet retransmission process. This process includes the following steps: Confirmation of data packet transmission failure; Resend failed packets; Acknowledges successful reception of a packet after retransmission.

[0095] Moderate failure load balancing mechanism: When the error rate rises to the moderate failure threshold, the system will achieve load balancing in the following ways: Identify healthy nodes and assign data transmission tasks from failed nodes to these nodes; Adjust network routing to ensure that data traffic is evenly distributed and avoid overloading a single node.

[0096] Serious failure alarm system: When the error rate exceeds the serious failure threshold, the system will do the following: Send real-time alert information to system administrators, including detailed information such as the identification of failed nodes and error rates; Record fault events for subsequent fault analysis and system optimization.

[0097] See also Figure 5 , the following is a specific embodiment: Assuming that the error rate of a node reaches the minor fault threshold of 3%, the system will retransmit the data packet of this node.

[0098] If the error rate continues to rise to the moderate failure threshold of 5%, the system will reallocate the traffic of the node to other nodes.

[0099] If the error rate deteriorates further and exceeds the critical failure threshold of 7%, the system will send an alarm to the administrator and record the failure event.

[0100] This hierarchical fault handling mechanism ensures that the RapidIO protocol can respond quickly and effectively when faced with a fault, thereby ensuring the stable operation of the system and the integrity of the data.

[0101] The following describes a processing device for the RapidIO protocol during fault injection provided by the present invention. The processing device for the RapidIO protocol during fault injection described below and the processing method for the RapidIO protocol during fault injection described above can be referred to each other.

[0102] like Figure 6 As shown, in one embodiment, a RapidIO protocol processing device during fault injection includes an acquisition module 610 , an analysis module 620 , a setting module 630 , an adjustment module 640 , and an instruction module 650 .

[0103] The acquisition module 610 is used to acquire performance indicator data of each node in the RapidIO network. The performance indicator data includes real-time monitored node signal strength indicators, error rate statistics, and data packet loss records.

[0104] The analysis module 620 is used to analyze the performance indicator data and output the fault prediction result.

[0105] The setting module 630 is used to set the multi-level fault threshold according to the fault prediction result, divide different faults into different levels according to different types of faults, and set the corresponding multi-level initial fault threshold.

[0106] The adjustment module 640 is used to adaptively adjust the fault threshold of each node based on the fault prediction result and the real-time monitored performance indicator data.

[0107] The instruction module 650 is used to instruct the system to implement corresponding processing measures according to the fault level.

[0108] In this embodiment, the performance indicator data is analyzed, specifically for: Clean and format the acquired performance indicator data to obtain pre-processed data; Features that are helpful for fault prediction are extracted from the preprocessed data, including the changing trend of signal strength, the changing trend of error rate, and system load.

[0109] In this embodiment, the fault prediction result is output, specifically used for: Obtain preprocessed performance indicator data; The performance indicator data is input into the machine learning model and the fault prediction result is output. The machine learning model is trained with the performance indicator data of each node in the RapidIO network and the fault probability samples as training data.

[0110] In this embodiment, the instruction module 650 is specifically used to: In response to the fault level being a minor fault, triggering a data packet retransmission mechanism; In response to the fault level being a medium fault, a load balancing mechanism is started and the data transmission task of the faulty node is allocated to a healthy node; In response to the fault level being a serious fault, an alarm message is sent to an administrator and the fault event is recorded.

[0111] The processing device of the RapidIO protocol during fault injection improves the reliability of the RapidIO protocol during fault injection through machine learning prediction and dynamic threshold adjustment methods, realizes the timeliness and accuracy of fault processing, and improves the efficiency and flexibility of fault processing through a hierarchical processing mechanism.

[0112] Figure 7 An example of a physical structure diagram of an electronic device is shown. The electronic device may be a smart terminal, and its internal structure diagram may be as follows: Figure 7 As shown. The electronic device includes a processor, a memory and a network interface connected via a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the electronic device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a processing method for the RapidIO protocol during fault injection is implemented, and the method includes: Obtain performance indicator data of each node in the RapidIO network, including real-time monitoring of node signal strength indicators, error rate statistics, and packet loss records; Analyze performance indicator data and output fault prediction results; According to the fault prediction results, multi-level fault thresholds are set, and different levels are divided according to different types of faults, and corresponding multi-level initial fault thresholds are set; Adaptively adjust the fault threshold of each node based on fault prediction results and real-time monitoring performance indicator data; According to the fault level instruction system implements corresponding treatment measures.

[0113] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the scheme of the present invention, and does not constitute a limitation on the electronic device to which the scheme of the present invention is applied. The specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0114] On the other hand, the present invention further provides a computer storage medium storing a computer program, wherein when the computer program is executed by a processor, a method for processing a RapidIO protocol during fault injection is implemented, the method comprising: Obtain performance indicator data of each node in the RapidIO network, including real-time monitoring of node signal strength indicators, error rate statistics, and packet loss records; Analyze performance indicator data and output fault prediction results; According to the fault prediction results, multi-level fault thresholds are set, and different levels are divided according to different types of faults, and corresponding multi-level initial fault thresholds are set; Adaptively adjust the fault threshold of each node based on fault prediction results and real-time monitoring performance indicator data; According to the fault level instruction system implements corresponding treatment measures.

[0115] In another aspect, a computer program product or a computer program is provided, the computer program product or the computer program comprising computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and when the processor executes the computer instructions, a method for processing a RapidIO protocol during fault injection is implemented, the method comprising: Obtain performance indicator data of each node in the RapidIO network, including real-time monitoring of node signal strength indicators, error rate statistics, and packet loss records; Analyze performance indicator data and output fault prediction results; According to the fault prediction results, multi-level fault thresholds are set, and different levels are divided according to different types of faults, and corresponding multi-level initial fault thresholds are set; Adaptively adjust the fault threshold of each node based on fault prediction results and real-time monitoring performance indicator data; According to the fault level instruction system implements corresponding treatment measures.

[0116] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory.

[0117] By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0118] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0119] The above-described embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the attached claims.

Claims

1. A method for processing a RapidIO protocol during fault injection, characterized in that: The method comprises: Obtaining performance indicator data of each node in the RapidIO network, the performance indicator data including real-time monitored node signal strength indicators, error rate statistics, and data packet loss records; Analyze performance indicator data and output fault prediction results; According to the fault prediction results, multi-level fault thresholds are set, and different levels are divided according to different types of faults, and corresponding multi-level initial fault thresholds are set; Adaptively adjust the fault threshold of each node based on fault prediction results and real-time monitoring performance indicator data; According to the fault level instruction system implements corresponding treatment measures.

2. The method for processing a RapidIO protocol during fault injection according to claim 1, characterized in that: The analysis performance indicator data includes: Clean and format the acquired performance indicator data to obtain pre-processed data; Features useful for fault prediction are extracted from the preprocessed data, wherein the features include a change trend of signal strength, a change trend of error rate, and system load.

3. The method for processing the RapidIO protocol during fault injection according to claim 2, characterized in that: The output fault prediction result includes: Obtain preprocessed performance indicator data; The performance indicator data is input into a machine learning model, and a fault prediction result is output. The machine learning model is trained using the performance indicator data of each node in the RapidIO network and a fault probability sample as training data.

4. The method for processing the RapidIO protocol during fault injection according to claim 3, characterized in that: The machine learning model is a random forest model, a support vector machine model or a recurrent neural network model.

5. The method for processing the RapidIO protocol during fault injection according to claim 4, characterized in that: The machine learning model is optimized using a grid search method or a Bayesian optimization method, and the prediction performance is evaluated using a cross-validation technique.

6. The method for processing a RapidIO protocol during fault injection according to any one of claims 1 to 5, characterized in that: The multi-level fault threshold is set by analyzing historical fault data and combining industry standards. The fault levels include minor faults, moderate faults and severe faults, and the threshold corresponding to each fault level is divided according to the error rate of the node.

7. The method for processing the RapidIO protocol during fault injection according to claim 6, characterized in that: The corresponding processing measures are implemented according to the fault level instruction system, including: In response to the fault level being a minor fault, triggering a data packet retransmission mechanism; In response to the fault level being a medium fault, a load balancing mechanism is started and the data transmission task of the faulty node is allocated to a healthy node; In response to the fault level being a serious fault, an alarm message is sent to an administrator and the fault event is recorded.

8. A RapidIO protocol processing device during fault injection, characterized in that: include: An acquisition module is used to acquire performance indicator data of each node in the RapidIO network, wherein the performance indicator data includes a real-time monitored node signal strength indicator, error rate statistics, and data packet loss records; Analysis module, used to analyze performance indicator data and output fault prediction results; A setting module, used to set multi-level fault thresholds according to the fault prediction results, divide different faults into different levels according to different types of faults, and set corresponding multi-level initial fault thresholds; An adjustment module is used to adaptively adjust the fault threshold of each node based on the fault prediction results and real-time monitoring performance indicator data; The instruction module is used to instruct the system to implement corresponding processing measures according to the fault level.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for processing the RapidIO protocol during fault injection according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for processing a RapidIO protocol during fault injection according to any one of claims 1 to 7 are implemented.

Images