Fault diagnosis device and processing method thereof, management controller and electronic equipment

Abstract

The invention relates to the technical field of fault diagnosis, and discloses a fault diagnosis device and a processing method thereof, a management controller and electronic equipment, and the method comprises the steps that a log data storage module obtains and stores equipment log data; the priority determination module determines the hardware fault diagnosis priority according to the application scene of the fault diagnosis device; the log data processing module performs format conversion on the equipment log data to obtain vectorized log data, performs classification processing on the vectorized log data, and transmits corresponding classification processing results to the fault diagnosis module according to hardware fault diagnosis priorities; and the fault diagnosis module performs corresponding operation on the received classification processing result and outputs a fault diagnosis result. In this way, the device can diagnose hardware faults specifically and quickly according to different application scenes, resource waste is avoided, it is ensured that key data can be processed in time, the operation and maintenance cost is reduced, and the accuracy and reliability of fault diagnosis are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis, and in particular to a fault diagnosis device and a processing method thereof, a management controller, and an electronic device. Background Art

[0002] In equipment operation and maintenance, the fault conditions of equipment are mainly judged based on manual analysis and the experience of the operation and maintenance personnel. Equipment failure conditions are complex and diverse, and the operation and maintenance personnel need to spend a lot of time collecting equipment operation information and check and analyze each component of the equipment one by one. The process is extremely cumbersome, requires a lot of effort, and increases costs. In addition, the experience levels of different operation and maintenance personnel vary, resulting in inaccurate fault diagnosis results. Summary of the invention

[0003] In view of this, the purpose of the present invention is to provide a fault diagnosis device and its processing method, a management controller, and an electronic device, which can quickly complete equipment fault diagnosis, improve operation and maintenance efficiency, reduce operation and maintenance costs, and improve equipment availability and production efficiency.

[0004] In order to solve the above technical problems, the present invention provides a fault diagnosis device, comprising: Log data storage module, used to obtain and store device log data; A priority determination module, used to determine the hardware fault diagnosis priority according to the application scenario of the fault diagnosis device; A log data processing module, used for performing format conversion on the device log data to obtain vectorized log data, classifying the vectorized log data, and transmitting the corresponding classification processing results to the fault diagnosis module according to the hardware fault diagnosis priority; The fault diagnosis module is used to perform corresponding operations on the received classification processing results and output fault diagnosis results.

[0005] In order to solve the above technical problems, the present invention also provides a processing method of a fault diagnosis device, comprising: Using the log data storage module to store device log data; According to the application scenario of the fault diagnosis device, the priority of hardware fault diagnosis is determined by using the priority determination module; Using a log data processing module to convert the format of the device log data to obtain vectorized log data, classifying the vectorized log data, and transmitting the corresponding classification processing results to the fault diagnosis module according to the hardware fault diagnosis priority; The fault diagnosis module is used to perform corresponding operations on the received classification processing results and output fault diagnosis results.

[0006] In order to solve the above technical problem, the present invention further provides a management controller, comprising the above fault diagnosis device provided by the present invention.

[0007] In order to solve the above technical problem, the present invention further provides an electronic device, comprising the above management controller provided by the present invention.

[0008] It can be seen from the above technical scheme that a fault diagnosis device provided by the present invention includes: a log data storage module, which is used to obtain and store equipment log data; a priority determination module, which is used to determine the hardware fault diagnosis priority according to the application scenario of the fault diagnosis device; a log data processing module, which is used to convert the format of the equipment log data to obtain vectorized log data, classify the vectorized log data, and transmit the corresponding classification processing results to the fault diagnosis module according to the hardware fault diagnosis priority; the fault diagnosis module is used to perform corresponding operations on the received classification processing results and output the fault diagnosis results.

[0009] The beneficial effect of the present invention is that in the above-mentioned fault diagnosis device provided by the present invention, the log data storage module can be used to obtain and store the equipment log data, which provides a rich and accurate information source for fault diagnosis and helps to comprehensively record the operation status of the equipment; the priority determination module can be used to determine the hardware fault diagnosis priority according to the application scenario of the fault diagnosis device, so that the device can be targeted according to different use environments and needs. Diagnose hardware faults and avoid waste of resources; the log data processing module can be used to convert the format of the equipment log data to obtain vectorized log data, and the vectorized log data is classified and processed, so that complex log data can be presented in an easy-to-process form, which is convenient for subsequent operations. At the same time, according to the hardware fault diagnosis priority, the corresponding classification processing results are transmitted to the fault diagnosis module, which optimizes the data processing process and ensures that key data can be processed in time; finally, the fault diagnosis module is used to perform corresponding operations on the received classification processing results, and the fault diagnosis results are output. By performing precise operations on the processed and classified data, the accuracy and reliability of fault diagnosis can be improved, and maintenance personnel can be helped to quickly locate faults. The entire device can quickly complete equipment fault diagnosis, improve operation and maintenance efficiency, reduce operation and maintenance costs, and improve equipment availability and production efficiency.

[0010] In addition, the present invention also provides a corresponding processing method, a management controller, and an electronic device for the fault diagnosis device, which have the same or corresponding technical features as the above-mentioned fault diagnosis device, further making the above-mentioned fault diagnosis device more practical, and the processing method, management controller, and electronic device have corresponding advantages. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0012] Figure 1 A schematic diagram of the structure of a fault diagnosis device provided by an embodiment of the present invention; Figure 2 A schematic diagram of the structure of a log data processing module provided in an embodiment of the present invention; Figure 3 A schematic diagram of the structure of a fault diagnosis module provided by an embodiment of the present invention; Figure 4 A schematic diagram of an application of a fault diagnosis device provided by an embodiment of the present invention in a management controller; Figure 5 A flow chart of a processing method of a fault diagnosis device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0013] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0014] In order to enable those skilled in the art to better understand the solution of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods. Figure 1 Schematic diagram of a fault diagnosis device provided by an embodiment of the present invention. Figure 1 As shown, the fault diagnosis device provided by the embodiment of the present invention may include the following modules: Log data storage module 1, used to obtain and store device log data; A priority determination module 2, used to determine the hardware fault diagnosis priority according to the application scenario of the fault diagnosis device; The log data processing module 3 is used to convert the format of the device log data to obtain vectorized log data, classify the vectorized log data, and transmit the corresponding classification processing results to the fault diagnosis module 4 according to the hardware fault diagnosis priority; The fault diagnosis module 4 is used to perform corresponding operations on the received classification processing results and output the fault diagnosis results.

[0015] In the above-mentioned fault diagnosis device provided by the embodiment of the present invention, the log data storage module can be used to obtain and store equipment log data, which provides a rich and accurate information source for fault diagnosis and helps to comprehensively record the operation status of the equipment; the priority determination module can be used to determine the hardware fault diagnosis priority according to the application scenario of the fault diagnosis device, so that the device can diagnose hardware faults in a targeted manner according to different use environments and needs, avoiding waste of resources; the log data processing module can convert the format of the equipment log data to obtain vectorized log data, and classify the vectorized log data, so that complex log data can be presented in an easy-to-process form, which is convenient for subsequent operations, and at the same time, the corresponding classification processing results are transmitted to the fault diagnosis module according to the hardware fault diagnosis priority, which optimizes the data processing process and ensures that key data can be processed in time; finally, the fault diagnosis module is used to perform corresponding operations on the received classification processing results, and the fault diagnosis results are output. By performing precise operations on the processed and classified data, the accuracy and reliability of fault diagnosis can be improved, and maintenance personnel can be helped to quickly locate faults. The entire device can quickly complete equipment fault diagnosis, improve operation and maintenance efficiency, reduce operation and maintenance costs, and improve equipment availability and production efficiency.

[0016] It should be noted that the above-mentioned fault diagnosis device provided by the present invention can be applied to a management controller. The management controller can be a baseboard management controller (Baseboard Management Controller, BMC). The BMC chip can be used to be responsible for the hardware monitoring, fault management and remote control functions of the device, thereby ensuring the safety and stability of the device. However, since the computing resources and storage resources of the BMC chip are relatively limited, by adding the log data storage module 1, priority determination module 2, log data processing module 3 and fault diagnosis module 4 of the above-mentioned fault diagnosis device to the BMC chip, the device fault analysis delay can be effectively reduced, while enhancing the privacy and security of the device data.

[0017] The log data of the device (such as a server) acquired and stored by the log data storage module 1 is an important file for recording the operation status of the device, which is automatically generated by the device during operation. The log data contains key contents such as user requests, system events, and error information. It is recorded in chronological order and stored in a structured or unstructured form, and is mainly divided into types such as access logs, error logs, security logs, application logs, and performance logs. Among them, the access log is used to record detailed information of user requests, including Internet Protocol Address (IP), request path, Hypertext Transfer Protocol (HTTP) method, and status code. The error log is used to capture abnormal events in the operation of the device (such as a server). The security log is used to record security-related operations and warnings. The application log is used to reflect the operating status of the business logic. The performance log is used to monitor the resource usage of the device (such as a server). Log data is widely used in system monitoring, problem troubleshooting, user behavior analysis, etc. For example, the error log can quickly locate system problems and improve operation and maintenance efficiency; the security log can detect abnormal behavior and ensure system safety. The fault diagnosis device provided by the present invention can realize efficient monitoring and analysis of device log data, and then perform device fault diagnosis. In practical applications, the log data storage module 1 may be a storage module in a management controller.

[0018] Equipment (such as server) fault diagnosis refers to quickly locating the cause of the fault and taking measures to restore the normal operation of the system by analyzing the equipment's operating status and log information. Its core goal is to ensure the stability and continuity of the equipment. The main types of faults may include: processor failure (such as central processing unit failure), memory failure, hard disk failure, and power failure. Processor failures may include overheating, performance degradation, etc.; memory failures may include memory stick damage, etc.; hard disk failures may include disk aging, bad sectors, file system damage, etc.; power supply failures may include power module failure, unstable power supply, etc. The present invention can collect relatively complete fault information during the fault diagnosis process, and through real-time monitoring and combined with operation and maintenance experience, it can quickly locate the problem and formulate a solution.

[0019] The above-mentioned priority determination module 2 can determine the hardware fault diagnosis priority according to the application scenario of the fault diagnosis device. Of course, this module can determine the hardware fault diagnosis priority according to the application scenario of the management controller. Hardware fault diagnosis priority refers to the ranking of the diagnostic importance and urgency of different hardware faults. For example: for general servers, the diagnostic priority of central processing unit (CPU) fault, memory fault, and power supply fault is usually higher, because these faults will directly affect the basic functions and stability of the server, and the diagnostic priority of hard disk fault, fan fault, etc. is relatively low; and for artificial intelligence (AI) servers, due to the high dependence of AI training and reasoning on graphics processing unit (GPU) and memory, GPU fault and memory fault are diagnosed first. By determining the hardware fault diagnosis priority, the fault diagnosis device can give priority to those hardware faults that have a greater impact on the operation of the equipment and are more urgent, so as to more efficiently ensure the normal operation of the equipment and reduce the business interruption time and loss caused by hardware failure.

[0020] The above-mentioned log data processing module 3 can first convert the format of the device log data to obtain vectorized log data, so that the unstructured or semi-structured log data can be converted into a digital vector form that is easy for the computer to process, which makes the subsequent classification processing and analysis more efficient and improves the speed and accuracy of data processing; then the vectorized log data is classified and processed, and different types of log data are separated, and the corresponding classification processing results are transmitted to the fault diagnosis module 4 according to the hardware fault diagnosis priority determined by the priority determination module 2, for example: the processing results corresponding to the relevant log data of the server hardware with a higher diagnosis priority are transmitted to the fault diagnosis module 4. In this way, the fault diagnosis module 4 can give priority to high-priority fault information and avoid searching for key information in a large amount of irrelevant data, thereby speeding up the speed of fault diagnosis, timely discovering and solving serious hardware faults, and reducing equipment downtime.

[0021] The above-mentioned fault diagnosis module 4 can perform corresponding operations on the received classification processing results and output the fault diagnosis results. The fault diagnosis module 4 can be an AI fault diagnosis module to realize AI algorithm calculation acceleration and perform hardware fault diagnosis.

[0022] It should be pointed out that the present invention presents the fault diagnosis device in a modular manner, which allows the system to have better scalability and adaptability when facing equipment log data of different types and scales and different hardware fault diagnosis requirements. It can easily adjust the classification algorithm, priority rules or add new processing modules to adapt to the ever-changing equipment and business needs.

[0023] Furthermore, in a specific implementation, in the above-mentioned fault diagnosis device provided in an embodiment of the present invention, the log data processing module 3 may include a log parsing unit, a log vectorization unit and a log classification unit; the log parsing unit is used to parse the device log data and convert the parsed data into structured log data; the log vectorization unit is used to convert the structured log data converted by the log parsing unit into vectorized log data; the log classification unit is used to classify the vectorized log data according to the different characteristics of the device hardware, and transmit the corresponding one or more types of data to the fault diagnosis module 4 according to the hardware fault diagnosis priority.

[0024] Figure 2 The structure diagram of the log data processing module provided by the embodiment of the present invention is as follows. Figure 2 As shown, the log data processing module 3 includes a log parsing unit 31, a log vectorization unit 32 and a log classification unit 33. The log parsing unit 31 can parse the device log data and convert the parsed data (such as unstructured log data or semi-structured log data) into structured log data, so that the data has a unified format and specification, which is convenient for subsequent processing. The log vectorization unit 32 can convert the structured log data converted by the log parsing unit 31 into vectorized log data, and convert the text information into a digital vector form, which is more suitable for computer storage, calculation and analysis, for example: for the analysis process of machine learning algorithms, etc., to improve the efficiency and accuracy of data processing. The log classification unit 33 can classify and process the vectorized log data according to the different characteristics of the device hardware (such as processor, memory, hard disk, etc.), and transmit the corresponding one or more types of data to the fault diagnosis module 4 according to the hardware fault diagnosis priority. This enables targeted analysis and processing of the failure modes and characteristics of different hardware, better capture of features, improved accuracy of fault diagnosis, and ensures that the fault diagnosis module 4 can give priority to high-priority fault information to avoid delays in fault processing due to processing of low-priority information.

[0025] Furthermore, in a specific implementation, in the above-mentioned fault diagnosis device provided in an embodiment of the present invention, the log parsing unit 31 can be specifically used to read device log data, parse the read device log data according to predefined rules, extract feature fields, and convert the feature fields into structured log data.

[0026] In implementation, the log parsing unit 31 can first read the original device log data from the log data storage module 1, and then in order to extract valuable information from the original device log data, the log parsing unit 31 parses the read log data according to pre-set rules. These rules can be formulated according to the format characteristics of the log data and the type of information to be extracted. For example, the rules can specify how to identify different types of time, how to distinguish timestamps, message content and related parameters, etc. The log parsing unit can also extract key feature fields from the parsed log data, organize the extracted feature fields according to a certain structure and format, and form structured log data. This structured data format can make each feature field have a clear position and meaning, which is convenient for subsequent storage and analysis, can greatly reduce the amount of data, remove redundant information, and make subsequent data processing operations more efficient.

[0027] The above-mentioned characteristic fields may specifically include one or a combination of a timestamp, an Internet Protocol (IP) address, a Uniform Resource Locator (URL), and a status code. The structured format of the above-mentioned structured log data may be JSON (a lightweight data exchange format) or CSV (Comma-Separated Values).

[0028] Furthermore, in a specific implementation, in the above-mentioned fault diagnosis device provided by an embodiment of the present invention, the log vectorization unit 32 can be specifically used to normalize or standardize the numerical fields of the structured log data converted by the log parsing unit 31, convert the classification fields into numerical values ​​through a one-hot encoding method, and encode the text fields using an embedded method, so that the structured log data converted by the log parsing unit 31 is converted into vectorized log data.

[0029] In implementation, for the structured log data converted by the log parsing unit 31, involving feature fields such as timestamp, IP address, uniform resource locator, status code, etc., the log vectorization unit 32 can normalize or standardize the numerical field (such as the response time in the timestamp), which can eliminate the influence of the numerical differences of different features on the model and improve the accuracy and stability of the model. For example, the response time in the timestamp may vary greatly due to different devices or time periods. After normalization or standardization, the model can better learn its relationship with other features. The log vectorization unit 32 can convert the classification field (such as the status code) into a numerical value through the unique hot encoding method, so that the model can process the originally non-numerical classification data. The unique hot encoding maps each category to a binary vector, where only the corresponding category position is 1 and the other positions are 0. This can clearly express the category information, making it easier for the model to understand and distinguish different categories, thereby better mining the potential patterns in the data. Taking the status code as an example, through unique hot encoding, the model can directly use it as an input feature for learning without complex category comparison and conversion. The log vectorization unit 32 can encode text fields (such as uniform resource locators) in an embedded manner, which can capture potential information in the text fields and enrich the feature representation of the data. For example, for uniform resource locators, embedded coding can convert the domain name, path and other information it contains into vectors, which helps the model discover the association between the uniform resource locator and other log features, so as to better perform tasks such as log analysis and anomaly detection. In this way, the structured log data converted by the log parsing unit 31 is converted into vectorized log data. The vectorized data can be stored in a matrix form for subsequent log processing. The vectorized log data can be easily applied to various machine learning, deep learning and data analysis algorithms, such as cluster analysis, classification algorithms, neural networks, etc., to achieve log fault diagnosis.

[0030] Furthermore, in a specific implementation, in the above-mentioned fault diagnosis device provided in an embodiment of the present invention, the fault diagnosis module 4 may include an intelligent control unit, a cache unit and an accelerated computing unit; the intelligent control unit is used to configure the cache unit and the accelerated computing unit according to the hardware fault diagnosis type, and load the target algorithm model parameters; the cache unit is used to store the processing results, target algorithm model parameters, and fault diagnosis results transmitted by the log data processing module 3; the accelerated computing unit is used to perform corresponding accelerated calculations on the received classification processing results according to the target algorithm, and output the fault diagnosis results.

[0031] Figure 3 FIG. 1 is a schematic diagram of the structure of a fault diagnosis module provided in an embodiment of the present invention. Figure 3As shown, the fault diagnosis module 4 includes an intelligent control unit 41, a cache unit 42 and an accelerated calculation unit 43. The intelligent control unit 41 can dynamically configure the cache unit 42 and the accelerated calculation unit 43 according to the type of hardware fault diagnosis required, and load the target algorithm model parameters, so that the fault diagnosis module 4 can adapt to a variety of fault diagnosis scenarios, improve the flexibility and versatility of fault diagnosis, and ensure that a suitable algorithm model is used when performing fault diagnosis, which helps to improve the accuracy and efficiency of fault diagnosis; the cache unit 42 can store the processing results transmitted by the log data processing module 3, the target algorithm model parameters, and the fault diagnosis results, providing data support and temporary storage space for intermediate results for the fault diagnosis process, facilitating subsequent query, analysis and use, and helping to improve the data processing efficiency and stability of the entire system; the accelerated calculation unit 43 can perform corresponding accelerated calculations on the received classification processing results according to the target algorithm, output the fault diagnosis results, can quickly process a large amount of data, shorten the time of fault diagnosis, enable the system to detect and respond to hardware faults in a timely manner, and reduce the impact of faults on system operation.

[0032] The above-mentioned target algorithm can be an AI algorithm, which can include convolutional neural networks, recurrent neural networks, and Transformer (deep learning model architecture based on attention mechanism) neural networks. Applying these AI algorithms to the target algorithm can give full play to their respective advantages, select appropriate algorithms or combine multiple algorithms according to specific task requirements, thereby improving the accuracy, efficiency and generalization ability of the target algorithm.

[0033] Furthermore, in a specific implementation, in the above-mentioned fault diagnosis device provided in an embodiment of the present invention, the intelligent control unit 41 can be specifically used to dynamically set and adjust the cache unit 42 and the accelerated computing unit 43 according to the hardware fault type in the current fault diagnosis task, so that the cache unit 42 and the accelerated computing unit 43 are in a state most suitable for the current fault diagnosis task, and the target algorithm model parameters corresponding to the hardware fault diagnosis type are loaded into the corresponding accelerated computing unit 43.

[0034] In implementation, the intelligent control unit 41 dynamically sets and adjusts the cache unit 42 and the accelerated computing unit 43 according to the type of hardware fault, which allows the system to quickly call relevant data and algorithms, reduce unnecessary calculations and data transmission, and make the fault diagnosis process faster. The cache unit 42 and the accelerated computing unit 43 are in the most suitable state, which can ensure that the data acquired and processed by the system is accurate and effective, and provide high-quality data support for the diagnostic algorithm. Loading the target algorithm model parameters into the accelerated computing unit can give full play to its acceleration effect, speed up the model operation speed, and thus quickly obtain diagnostic results and shorten the fault diagnosis time. The system can automatically adjust the configuration according to different hardware fault types without manual intervention, and can adapt to a variety of complex fault scenarios.

[0035] Furthermore, in a specific implementation, in the above-mentioned fault diagnosis device provided in the embodiment of the present invention, if Figure 3 As shown, the cache unit 42 may specifically include an input cache, a parameter cache and an output cache; the input cache is used to store the processing results transmitted by the log data processing module 3; the parameter cache is used to store the target algorithm model parameters; the target algorithm model parameters may include weight parameters and bias parameters of the neural network model; the output cache is used to store the fault diagnosis results.

[0036] In implementation, the input cache can store the processing results transmitted by the log data processing module 3, so that the accelerated computing unit 43 can quickly obtain the required data, avoiding the time waste caused by waiting for data transmission, avoiding repeated processing and transmission of data, reducing the waste of system resources, and improving the data processing efficiency of the entire system. The parameter cache can store the target algorithm model parameters, such as the weight parameters and bias parameters of the neural network model, so that these parameters can be quickly accessed during the algorithm operation process, reducing the time to read parameters from the external storage device, and speeding up the training and reasoning speed of the model. The output cache can store fault diagnosis results, which is convenient for other modules or external systems to quickly obtain results when needed without having to calculate again, thereby improving the response speed of the system.

[0037] Furthermore, in a specific implementation, in the above-mentioned fault diagnosis device provided in the embodiment of the present invention, if Figure 3 As shown, the accelerated calculation unit 43 may specifically include multiple processing units (Processing Element, PE), a nonlinear calculation unit; the processing unit is used to perform corresponding linear calculations in the target algorithm on the received classification processing results through parallel computing; the nonlinear calculation unit is used to accelerate the calculation of the activation function in the target algorithm.

[0038] In implementation, the processing unit can use parallel computing to perform corresponding linear calculations (such as matrix multiplication) in the target algorithm on the received classification processing results, and can process multiple log data at the same time, speeding up the execution speed of the entire target algorithm. The design of multiple processing units allows the system to flexibly adjust the scale of parallel computing according to actual task requirements. The number of processing units can be dynamically allocated according to the complexity of the fault diagnosis task and the amount of data to achieve the best computing performance. The nonlinear calculation unit can accelerate the calculation of activation functions in the target algorithm, such as Rectified Linear Unit (ReLU), Gaussian Error Linear Unit (GELU), Sigmoid and other activation functions, and can quickly complete nonlinear transformations, significantly reduce the energy consumption when calculating activation functions, further improve the running speed of the algorithm, and enable the system to obtain fault diagnosis results faster; and the calculation of the activation function is separated and specially accelerated and optimized, which is convenient for separate optimization and expansion of the nonlinear calculation part when the system is upgraded or improved.

[0039] Furthermore, in a specific implementation, the above-mentioned fault diagnosis device provided in the embodiment of the present invention may further include: a clock module, which is used to provide the required time reference for the fault diagnosis device.

[0040] Figure 4 Schematic diagram of the application of the fault diagnosis device provided by the embodiment of the present invention in the management controller. Figure 4 As shown, the fault diagnosis device may include a clock module. The clock module can provide the required time reference for the fault diagnosis device, and specifically can provide a time reference for data sampling, data transmission and data processing in the fault diagnosis device, ensure the accuracy of data sampling, data transmission and data processing, avoid data deviation or error caused by time asynchrony, and thus provide reliable data support for subsequent fault diagnosis. The clock module can also record the time when the fault occurs, and can provide an accurate time mark for subsequent fault analysis. In addition, the clock module can also set a scheduled fault diagnosis task to help realize the automatic operation of the system, and perform fault diagnosis regularly at preset time intervals, so as to timely discover potential fault hazards without frequent manual intervention, thereby improving the intelligence level and operation efficiency of the system.

[0041] Furthermore, in a specific implementation, in the above-mentioned fault diagnosis device provided in the embodiment of the present invention, if Figure 4 As shown, it may also include: a power supply module, used to provide the required excitation power supply for the fault diagnosis device.

[0042] In implementation, the power module can ensure that each electronic component and module in the device obtains the appropriate voltage and current, avoids equipment failure or misjudgment due to power supply fluctuations, and ensures that the fault diagnosis device operates continuously and reliably, thereby improving the stability and reliability of the entire system. The power module of the present invention can have multiple safety protection functions such as overvoltage protection, overcurrent protection, and short circuit protection. When an abnormal situation occurs, such as excessive input voltage, excessive output current, or short circuit, the power module can quickly cut off the power supply or perform corresponding protection actions to prevent the fault from further expanding, protect the electronic components in the fault diagnosis device from damage, extend the service life of the equipment, and ensure the safe operation of the system.

[0043] Furthermore, in a specific implementation, in the above-mentioned fault diagnosis device provided by the embodiment of the present invention, the log data storage module 1, the log data processing module 3, the priority determination module 2 and the fault diagnosis module 4 are connected via a bus.

[0044] by Figure 4 For example, the management controller may include a processor, a storage module, a log data processing module, a fault diagnosis module, a clock module, and a power module. The processor here is the core module of the management controller, which is used for control and calculation. The processor may be an ARM (Advanced RISC Machines) processor or a RISC-V (open source instruction set architecture based on the principle of reduced instruction set) processor. The processor may be a single-core processor or a multi-core processor. The storage module here may include a log data storage module, which may be used to store data such as logs, including static random access memory (Static Random-Access Memory, SRAM), double data rate synchronous dynamic random access memory (Double Data Rate Synchronous Dynamic Random-Access Memory, DDR SDRAM), read-only memory (Read-Only Memory, ROM), flash memory (Flash Memory, FLASH), etc. Different modules are connected through buses, including Advanced eXtensible Interface (AXI) bus, Advanced High-performance Bus (AHB), and Advanced Peripheral Bus (APB).

[0045] It should be pointed out that the above-mentioned management controller (such as BMC chip) can be applied to various types of servers, including: servers built based on x86 architecture processors, ARM servers and RISC-V servers, etc., without limitation here.

[0046] In the above embodiments, the fault diagnosis device is described in detail. Based on the same inventive concept, the embodiments of the present invention also provide embodiments corresponding to the processing method of the fault diagnosis device.

[0047] The processing method of the fault diagnosis device provided in this embodiment is as follows: Figure 5 As shown, the specific steps include: S501, using a log data storage module to store device log data; S502, determining the hardware fault diagnosis priority using a priority determination module according to an application scenario of the fault diagnosis device; S503, using the log data processing module to convert the format of the device log data to obtain vectorized log data, classifying the vectorized log data, and transmitting the corresponding classification processing results to the fault diagnosis module according to the hardware fault diagnosis priority; S504: Use the fault diagnosis module to perform corresponding operations on the received classification processing results and output the fault diagnosis results.

[0048] In the processing method of the above-mentioned fault diagnosis device provided in an embodiment of the present invention, by executing the above-mentioned steps S501 to S504, the log data storage module can be used to obtain and store device log data, which provides a rich and accurate information source for fault diagnosis and helps to comprehensively record the operating status of the equipment; the priority determination module can be used to determine the hardware fault diagnosis priority according to the application scenario of the fault diagnosis device, so that the device can diagnose hardware faults in a targeted manner according to different usage environments and requirements, thereby avoiding waste of resources; the log data processing module can be used to convert the format of the device log data to obtain vectorized log data, and the vectorized log data is classified and processed, so that complex log data can be presented in an easy-to-process form, which is convenient for subsequent operations. At the same time, the corresponding classification processing results are transmitted to the fault diagnosis module according to the hardware fault diagnosis priority, thereby optimizing the data processing process and ensuring that key data can be processed in a timely manner; finally, the fault diagnosis module is used to perform corresponding operations on the received classification processing results and output the fault diagnosis results. By performing precise operations on the processed and classified data, the accuracy and reliability of fault diagnosis can be improved, helping maintenance personnel to quickly locate faults. The overall approach can quickly complete equipment fault diagnosis, improve operation and maintenance efficiency, reduce operation and maintenance costs, and improve equipment availability and production efficiency.

[0049] Since the embodiments of the processing method part correspond to the embodiments of the fault diagnosis device part, the embodiments of the processing method part refer to the description of the embodiments of the fault diagnosis device part, which will not be repeated here. And it has the same beneficial effects as the above-mentioned fault diagnosis device.

[0050] Furthermore, in a specific implementation, in the processing method of the above-mentioned fault diagnosis device provided in an embodiment of the present invention, step S503 uses a log data processing module to convert the format of the device log data to obtain vectorized log data, classifies the vectorized log data, and transmits the corresponding classification processing results to the fault diagnosis module according to the hardware fault diagnosis priority. Specifically, it may include: using a log parsing unit to parse the device log data, and converting the parsed data into structured log data; using a log vectorization unit to convert the structured log data converted by the log parsing unit into vectorized log data; finally, using a log classification unit to classify the vectorized log data according to the different characteristics of the device hardware, and transmitting one or more corresponding types of data to the fault diagnosis module according to the hardware fault diagnosis priority.

[0051] In implementation, in the above steps, the device log data is parsed using a log parsing unit, and the parsed data is converted into structured log data, which may specifically include: reading the device log data using a log parsing unit, parsing the read device log data according to predefined rules, extracting feature fields, and converting the feature fields into structured log data.

[0052] In the above steps, the structured log data converted by the log parsing unit is converted into vectorized log data using a log vectorization unit. Specifically, it may include: for the structured log data converted by the log parsing unit, the numeric fields are normalized or standardized using a log vectorization unit, the categorical fields are converted into numeric values ​​using a one-hot encoding method, and the text fields are encoded using an embedded method, so that the structured log data converted by the log parsing unit 31 is converted into vectorized log data.

[0053] Further, in a specific implementation, in the processing method of the above-mentioned fault diagnosis device provided in an embodiment of the present invention, step S504 utilizes the fault diagnosis module to perform corresponding operations on the received classification processing results, and outputs the fault diagnosis results, which may specifically include: utilizing the intelligent control unit to configure the cache unit and the accelerated computing unit according to the hardware fault diagnosis type, and loading the target algorithm model parameters; utilizing the cache unit to store the processing results, target algorithm model parameters, and fault diagnosis results transmitted by the log data processing module; utilizing the accelerated computing unit to perform corresponding accelerated calculations on the received classification processing results according to the target algorithm, and outputting the fault diagnosis results.

[0054] During implementation, in the above steps, the intelligent control unit is used to configure the cache unit and the accelerated computing unit according to the hardware fault diagnosis type, and the target algorithm model parameters are loaded. Specifically, it can include: according to the hardware fault type in the current fault diagnosis task, the intelligent control unit is used to dynamically set and adjust the cache unit and the accelerated computing unit, so that the cache unit and the accelerated computing unit are in the state most suitable for the current fault diagnosis task, and the target algorithm model parameters corresponding to the hardware fault diagnosis type are loaded into the corresponding accelerated computing unit.

[0055] In the above steps, the cache unit is used to store the processing results transmitted by the log data processing module, the target algorithm model parameters, and the fault diagnosis results, which can specifically include: using the input cache to store the processing results transmitted by the log data processing module; using the parameter cache to store the target algorithm model parameters; the target algorithm model parameters include the weight parameters and bias parameters of the neural network model; and finally using the output cache to store the fault diagnosis results.

[0056] In the above steps, the accelerated computing unit is used to perform corresponding accelerated computing on the received classification processing results according to the target algorithm, and the fault diagnosis results are output. Specifically, it can include: using multiple processing units to perform corresponding linear computing in the target algorithm on the received classification processing results through parallel computing; using a nonlinear computing unit to accelerate the computing of the activation function in the target algorithm.

[0057] Furthermore, in a specific implementation, the processing method of the above-mentioned fault diagnosis device provided in the embodiment of the present invention may also include: using a clock module fault diagnosis device to provide the required time reference; and / or, recording the time when the fault occurs; and / or, using the clock module to set a timed fault diagnosis task.

[0058] Furthermore, in a specific implementation, the processing method of the above-mentioned fault diagnosis device provided in the embodiment of the present invention may also include: using a power supply module to provide the required excitation power supply for the fault diagnosis device.

[0059] For more specific working processes of the above steps, please refer to the corresponding contents disclosed in the above embodiments, which will not be repeated here.

[0060] Based on the same inventive concept, an embodiment of the present invention also provides a management controller, including the above-mentioned fault diagnosis device. That is, the management controller may include a log data storage module, which is used to obtain and store device log data; a priority determination module, which is used to determine the hardware fault diagnosis priority according to the application scenario of the fault diagnosis device; a log data processing module, which is used to convert the format of the device log data to obtain vectorized log data, classify the vectorized log data, and transmit the corresponding classification processing results to the fault diagnosis module according to the hardware fault diagnosis priority; a fault diagnosis module, which is used to perform corresponding operations on the received classification processing results and output the fault diagnosis results. Since the principle of solving the problem by the management controller is similar to that of the aforementioned fault diagnosis device, the implementation of the management controller can refer to the implementation of the fault diagnosis device, and the repeated parts will not be repeated.

[0061] Based on the same inventive concept, an embodiment of the present invention further provides an electronic device, including the above-mentioned management controller. Since the principle of solving the problem by the electronic device is similar to that of the above-mentioned management controller, the implementation of the electronic device can refer to the implementation of the management controller, and the repeated parts will not be repeated.

[0062] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0063] Finally, it should be noted that, unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of the present invention; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention; the terms "include", "comprise" and "have" and any other variations thereof in the present invention are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the existence of other identical elements in the process, method, article or device including the above elements. In the present invention, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0064] For the above-mentioned embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the order of the actions described, because according to the present invention, some steps may be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0065] In the several embodiments provided by the present invention, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the above units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be in the form of telecommunication or other forms.

[0066] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0067] Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on these embodiments, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in this field can still combine, add, delete or make other adjustments to the features in the embodiments of the present invention according to the circumstances without conflict and without making creative work, so as to obtain different other technical solutions that do not deviate from the concept of the present invention in essence, and these technical solutions also belong to the scope of protection of the present invention.

[0068] The above is a detailed introduction to the fault diagnosis device and its processing method, management controller, and electronic device provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method and core ideas of the present invention, rather than to limit the scope of protection of the invention; at the same time, for those of ordinary skill in the art, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A fault diagnosis device, characterized in that: include: Log data storage module, used to obtain and store device log data; A priority determination module, used to determine the hardware fault diagnosis priority according to the application scenario of the fault diagnosis device; A log data processing module, used for performing format conversion on the device log data to obtain vectorized log data, classifying the vectorized log data, and transmitting the corresponding classification processing results to the fault diagnosis module according to the hardware fault diagnosis priority; The fault diagnosis module is used to perform corresponding operations on the received classification processing results and output fault diagnosis results.

2. The fault diagnosis device according to claim 1, characterized in that: The log data processing module includes a log parsing unit, a log vectorization unit and a log classification unit; The log parsing unit is used to parse the device log data and convert the parsed data into structured log data; The log vectorization unit is used to convert the structured log data converted by the log parsing unit into vectorized log data; The log classification unit is used to classify the vectorized log data according to different characteristics of the device hardware, and transmit the corresponding one or more types of data to the fault diagnosis module according to the hardware fault diagnosis priority.

3. The fault diagnosis device according to claim 2, characterized in that: The log parsing unit is used to read the device log data, parse the read device log data according to predefined rules, extract feature fields, and convert the feature fields into structured log data.

4. The fault diagnosis device according to claim 3, characterized in that: The log vectorization unit is used to normalize or standardize the numerical fields of the structured log data converted by the log parsing unit, convert the categorical fields into numerical values ​​through a one-hot encoding method, and encode the text fields using an embedded method, so that the structured log data converted by the log parsing unit is converted into vectorized log data.

5. The fault diagnosis device according to claim 4, characterized in that: The characteristic field includes one or a combination of a timestamp, a uniform resource locator, and a status code; The log vectorization unit is used to normalize or standardize the field corresponding to the timestamp, convert the field corresponding to the status code into a numerical value through a one-hot encoding method, and encode the field corresponding to the uniform resource locator using an embedded method, so that the structured log data converted by the log parsing unit is converted into vectorized log data.

6. The fault diagnosis device according to claim 1, characterized in that: The fault diagnosis module includes an intelligent control unit, a cache unit and an accelerated computing unit; The intelligent control unit is used to configure the cache unit and the accelerated computing unit according to the hardware fault diagnosis type, and load the target algorithm model parameters; The cache unit is used to store the processing results transmitted by the log data processing module, the target algorithm model parameters, and the fault diagnosis results; The accelerated calculation unit is used to perform corresponding accelerated calculation on the received classification processing result according to the target algorithm and output the fault diagnosis result.

7. The fault diagnosis device according to claim 6, characterized in that: The intelligent control unit is used to dynamically set and adjust the cache unit and the accelerated computing unit according to the hardware fault type in the current fault diagnosis task, so that the cache unit and the accelerated computing unit are in the most suitable state for the current fault diagnosis task, and load the target algorithm model parameters corresponding to the hardware fault diagnosis type into the corresponding accelerated computing unit.

8. The fault diagnosis device according to claim 6, characterized in that: The cache unit includes an input cache, a parameter cache and an output cache; The input buffer is used to store the processing results transmitted by the log data processing module; The parameter cache is used to store the target algorithm model parameters; the target algorithm model parameters include weight parameters and bias parameters of the neural network model; The output buffer is used to store the fault diagnosis result.

9. The fault diagnosis device according to claim 6, characterized in that: The accelerated computing unit includes a plurality of processing units and a nonlinear computing unit; The processing unit is used to perform corresponding linear calculations in the target algorithm on the received classification processing results by means of parallel computing; The nonlinear calculation unit is used to accelerate the calculation of the activation function in the target algorithm.

10. The fault diagnosis device according to claim 1, characterized in that: Also includes: The clock module is used to provide the required time reference for the fault diagnosis device.

11. The fault diagnosis device according to claim 1, characterized in that: Also includes: The power supply module is used to provide the required excitation power supply for the fault diagnosis device.

12. The fault diagnosis device according to claim 1, characterized in that: The log data storage module, the log data processing module, the priority determination module and the fault diagnosis module are connected via a bus.

13. A processing method of a fault diagnosis device according to any one of claims 1 to 12, characterized in that: include: Using the log data storage module to store device log data; According to the application scenario of the fault diagnosis device, the priority of hardware fault diagnosis is determined by using the priority determination module; Using a log data processing module to convert the format of the device log data to obtain vectorized log data, classifying the vectorized log data, and transmitting the corresponding classification processing results to the fault diagnosis module according to the hardware fault diagnosis priority; The fault diagnosis module is used to perform corresponding operations on the received classification processing results and output fault diagnosis results.

14. A management controller, characterized in that: It comprises a fault diagnosis device as claimed in any one of claims 1 to 12.

15. An electronic device, characterized in that: Comprising a management controller as claimed in claim 14.

Images