Server hardware information processing method and device

Abstract

The invention discloses a server hardware information processing method and device, and relates to the technical field of server hardware information processing, and the method comprises the steps: obtaining a dynamic change mode in a fault event mapping table, and carrying out the real-time monitoring of the numerical fluctuation and state conversion of hardware configuration parameters, if it is detected that the parameter value exceeds the normal range or the state is abnormally skipped, a deep analysis mechanism is triggered to carry out fine-grained information extraction on related hardware components, and specific fault hardware positions and configuration abnormity details are determined; constructing a multi-dimensional hardware health state evaluation matrix according to the configuration exception details and the historical fault modes, and judging the overall operation state and potential risk points of the current hardware system by comprehensively analyzing the hardware category, the model parameter, the timestamp information and the configuration change trend; and a complete hardware information analysis result supporting accurate fault positioning is obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of server hardware information processing, and in particular to a server hardware information processing method and device. BACKGROUND

[0002] Server hardware information processing is a core field of modern data center management and operation, directly affecting the efficiency of device monitoring, fault diagnosis and performance optimization. With the increasing demand for real-time and accuracy of server operating status, efficient parsing and extraction of hardware information have become key technical support. However, existing methods have significant limitations in processing complex hardware information, especially when facing diverse hardware configurations and dynamic system environments, it is often difficult to achieve accurate data extraction and fine-grained information analysis. This leads to difficulties for administrators to quickly locate hardware faults or accurately grasp device status in actual operation.

[0003] In hardware information processing, the core challenge is how to efficiently parse and extract multi-level hardware parameters. Server hardware information is usually stored in complex DMI data tables, containing a variety of parameters from motherboard models to BIOS configurations, with deep data structure levels and complex fields. Existing parsing techniques often stop at the coarse-grained category recognition level, making it difficult to delve into specific model parameters or configuration details. For example, when processing BIOS information, the system may only recognize the hardware category (such as motherboard or CPU), but cannot further extract the specific version number or timestamp, resulting in a lack of sufficient time precision and parameter details when analyzing hardware events in system logs. This makes it difficult for administrators to accurately determine the specific occurrence time of a hardware event and related configuration changes when troubleshooting, affecting the efficiency of fault location. In addition, this coarse-grained parsing also leads to incomplete information extraction, especially in complex scenarios where multiple devices are running cooperatively, dynamic changes in hardware information are difficult to capture and refine in real time.

[0004] Therefore, how to build a multi-level parsing structure that gradually refines from coarse-grained category recognition to precise extraction of specific parameters, and supports timestamp-level parsing of hardware events in system logs, has become a key problem in server hardware information processing. SUMMARY

[0005] The technical problem to be solved by the present application is to provide a server hardware information processing method and device to solve the problems of the background art.

[0006] The present application adopts the following technical solutions to solve the above technical problems: A server hardware information processing method, comprising: The original hardware information of the server DMI data table is acquired, multi-level structured decomposition is performed on the hardware information through a decision tree classification algorithm, whether the current analysis level reaches a preset fine granularity threshold is judged according to the hierarchical depth and field complexity of the hardware category, if not, the lower layer is continuously analyzed, and a layered hardware data set containing a motherboard model, CPU specification and memory configuration is obtained; The layered hardware data set is subjected to parameter extraction processing by adopting a sequence pattern mining algorithm, the key parameter mode and version number characteristics in the hardware configuration are identified, if the BIOS version information is detected, the version number and release time thereof are extracted, if the device serial number is detected, the complete model identification is extracted, and a hardware configuration profile with complete parameter details is obtained; According to the time-related field in the hardware configuration profile, a time series analysis algorithm is adopted to construct a timestamp analysis model of hardware events, the time series characteristics of the BIOS timestamp, device startup time and configuration change record are analyzed, the accurate time point and duration of the occurrence of the hardware event are judged, and the state change timeline of each hardware component is determined; The timeline information is associated and matched with the hardware event record in the system log, a time window sliding mechanism is adopted to accurately locate the hardware failure event, if the difference between the hardware event timestamp and the failure log timestamp is less than a preset threshold, an association relationship is established, and a failure event mapping table containing accurate time information and detailed configuration parameters is obtained; The dynamic change mode in the failure event mapping table is acquired, the numerical fluctuation and state conversion of the hardware configuration parameters are monitored in real time, if it is detected that the parameter value exceeds the normal range or the state abnormally jumps, a deep analysis mechanism is triggered to extract fine-grained information of the related hardware component, and the specific fault hardware position and configuration abnormality details are determined; According to the configuration abnormality details and historical failure mode, a multi-dimensional hardware health state evaluation matrix is constructed, the overall running state and potential risk point of the current hardware system are judged by comprehensively analyzing the hardware category, model parameter, timestamp information and configuration change trend, and a complete hardware information analysis result supporting accurate fault positioning is obtained.

[0007] As a further preferred scheme of the server hardware information processing method, the original hardware information of the server DMI data table is acquired, multi-level structured decomposition is performed on the hardware information through a decision tree classification algorithm, whether the current analysis level reaches a preset fine granularity threshold is judged according to the hierarchical depth and field complexity of the hardware category, if not, the lower layer is continuously analyzed, and a layered hardware data set containing a motherboard model, CPU specification and memory configuration is obtained, including: The hardware information in the server DMI data table is acquired, original fields of motherboard model, CPU specification and memory configuration are extracted through data preprocessing to obtain an initial hardware dataset; The initial hardware dataset is classified by using a decision tree classification algorithm, a hierarchy is generated according to the hardware category to obtain a classified hardware hierarchical dataset; The hierarchical depth and field complexity of the hardware hierarchical dataset are calculated to determine whether a preset fine granularity threshold is reached to obtain a hierarchical decomposition judgment result; If the hierarchical decomposition judgment result does not reach the preset fine granularity threshold, the hardware hierarchical dataset is further decomposed by using the decision tree classification algorithm to obtain a more refined hardware sub-hierarchical dataset; Detailed fields of the motherboard model, CPU specification and memory configuration are extracted from the hardware sub-hierarchical dataset to generate a hardware dataset containing a multi-level hierarchy; The integrity of the motherboard model, CPU specification and memory configuration is judged through field verification on the multi-level hierarchy hardware dataset to obtain a verified hierarchical hardware dataset; The verified hierarchical hardware dataset is grouped by using a clustering algorithm to generate a final hardware classification dataset according to the hardware category and configuration similarity.

[0008] As a further preferred scheme of the server hardware information processing method, the sequence pattern mining algorithm is used to perform parameter extraction processing on the hierarchical hardware dataset, key parameter patterns and version number features in the hardware configuration are identified, if BIOS version information is detected, the version number and release time are extracted, if a device serial number is detected, the complete model identification is extracted, and a hardware configuration profile with complete parameter details is obtained, including: The sequence pattern mining algorithm is used to perform initial processing on the hierarchical hardware dataset to extract key parameter patterns to obtain a parameter pattern set; The BIOS version information is detected by analyzing the parameter pattern set; If the BIOS version information exists, the version number and release time are extracted to generate a BIOS parameter set; The format integrity of the version number and release time is verified by using a regular expression matching method for the BIOS parameter set to obtain a format-verified BIOS parameter set; The device serial number is detected by analyzing the parameter pattern set; If the device serial number exists, the complete model identification is extracted to generate a device identification set; The format-verified BIOS parameter set and the device identification set are grouped by using a clustering algorithm to generate a classified hardware parameter set according to the hardware configuration similarity; The parameter integrity is judged by comparing the classified hardware parameter set with the preset hardware configuration template, and a hardware configuration file after integrity verification is obtained. According to the hardware configuration file after integrity verification, a data mapping method is used to associate the key parameter mode with the fields in the hardware configuration file, and a final hardware configuration file is generated.

[0009] As a further preferred scheme of the server hardware information processing method, according to the time-related fields in the hardware configuration file, a time series analysis algorithm is used to construct a timestamp analysis model of hardware events, the accurate time point and duration period of the occurrence of a hardware event are judged by analyzing the time series characteristics of the BIOS timestamp, device startup time and configuration change record, and the state change timeline of each hardware component is determined, including: The time series analysis algorithm is used to preprocess the time fields in the hardware configuration file, the BIOS timestamp, device startup time and configuration change record are extracted from the time series data set, and an initial time series data set is generated; If the timestamps in the initial time series data set are consistent, the initial time series data set is feature-extracted by a sliding window method, and a time series feature set is obtained; According to the time series feature set, the K-means clustering algorithm is used to group the time series features, and a time point set of hardware events is generated; By analyzing the time point set, the time point of each hardware event and the time interval of adjacent events are calculated, and a duration period set of hardware events is obtained; If the time interval in the duration period set exceeds a preset threshold, an abnormal hardware event is identified by a time series anomaly detection algorithm, and an abnormal event time set is obtained; According to the abnormal event time set and the time point set, a data mapping method is used to associate the time point and duration period of the hardware event with the hardware component, and a state change timeline of the hardware component is generated; By comparing the state change timeline with the time fields in the hardware configuration file, the integrity of the timeline is verified, and a final hardware event timeline is obtained.

[0010] As a further preferred scheme of the server hardware information processing method, the time axis information is associated and matched with the hardware event record in the system log, a hardware fault event is accurately positioned by using a time window sliding mechanism, and if the difference between the hardware event timestamp and the fault log timestamp is less than a preset threshold, an association relationship is established, and a fault event mapping table containing accurate time information and detailed configuration parameters is obtained, including: The hardware event record and time axis information are obtained from the system log, the timestamp is preliminarily matched by using a time window sliding mechanism, and an initial fault event set is obtained; If the timestamp difference in the initial fault event set is less than the preset threshold value, the association between the hardware event and the fault log is established through a timestamp matching algorithm, and an associated event set is generated; According to the associated event set, the configuration parameters of each fault event are extracted, and a data integration method is used to generate a fault event mapping table containing time information and configuration parameters; Through the fault event mapping table, a time series clustering algorithm is used to group the fault events, and a fault event classification set is obtained; If there is a timestamp overlap in the fault event classification set, the time window is redivided through a time window adjustment mechanism, and an optimized fault event set is generated; According to the optimized fault event set, a data mapping method is used to associate the fault event with the hardware component, and a hardware component fault time axis is obtained; Through the hardware component fault time axis, the consistency of the timestamp and the configuration parameter is verified, and a final fault event time axis is obtained.

[0011] As a further preferred scheme of the server hardware information processing method, the dynamic change mode in the fault event mapping table is obtained by monitoring the numerical fluctuation and state transition of the hardware configuration parameter in real time. If it is detected that the parameter value exceeds the normal range or the state abnormally jumps, a deep analysis mechanism is triggered to extract the fine-grained information of the related hardware component, to determine the specific fault hardware location and configuration abnormality details, including: Through the real-time monitoring mechanism, time series data is obtained from the hardware configuration parameter to determine the dynamic mode of parameter value fluctuation and state change; If the parameter value fluctuation exceeds the preset threshold range or the state abnormally transitions, a deep analysis mechanism is triggered to extract fine-grained information of the related hardware component; According to the fine-grained information, a classification algorithm is used to locate the fault hardware, and the fault hardware location is obtained; Through the fault hardware location, the associated configuration parameter record is obtained, and the configuration abnormality details are extracted; If the configuration abnormality details overlap in time with the time series data, a time series analysis algorithm is used to adjust the division of abnormal events, and an optimized abnormal event set is obtained; According to the optimized abnormal event set, a data mapping method is used to integrate the fault hardware location and the configuration abnormality details, and a fault analysis mapping table is generated; Through the fault analysis mapping table, the consistency of the dynamic change mode and the time series data is verified, and the final fault location result is determined.

[0012] As a further preferred scheme of the server hardware information processing method, the multi-dimensional hardware health state evaluation matrix is constructed according to the configuration exception details and the historical failure mode, the overall operation state and the potential risk point of the current hardware system are judged by comprehensively analyzing the hardware category, the model parameter, the timestamp information and the configuration change trend, and the complete hardware information analysis result supporting accurate fault positioning is obtained, including: The time series data is obtained from the hardware configuration parameters, the configuration change trend is extracted by using the time series analysis algorithm combined with the timestamp information, and the hardware parameter dynamic characteristics are obtained; According to the configuration change trend, the multi-dimensional hardware health state evaluation matrix is constructed combined with the hardware category and the model parameter, and the health state score is generated; If the health state score is lower than the preset threshold, the failure characteristics matched with the hardware category and the model parameter are extracted from the historical failure mode, and the potential risk point is determined; For the potential risk point, the hardware health state evaluation matrix is analyzed by using the classification algorithm, and the preliminary fault positioning result is obtained; According to the preliminary fault positioning result, the associated fine-grained configuration data is extracted from the hardware configuration parameters, and the hardware information analysis result is generated; Through the hardware information analysis result, the clustering algorithm is used to group the configuration change trend combined with the timestamp information, and the time distribution characteristics of the abnormal event are judged; According to the time distribution characteristics of the abnormal event, the hardware health state evaluation matrix is updated, and the optimized fault positioning result is obtained.

[0013] A server hardware information processing device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor of the claim, wherein the processor executes the computer program to realize the steps of the server hardware information processing method.

[0014] Compared with the prior art, the above technical scheme has the following technical effects: This invention discloses a method and apparatus for acquiring server hardware logs. Addressing business scenarios where server hardware information is complex and fault location is difficult, it uses a decision tree classification algorithm to structurally decompose the original hardware information in the DMI data table, extracting key parameters such as motherboard, CPU, and memory layer by layer to form a hierarchical hardware dataset. A sequence pattern mining algorithm is employed to extract details such as BIOS version and device serial number, constructing a complete hardware configuration profile. A hardware event timeline is built through time series analysis, and combined with system logs, a time window sliding mechanism is used to achieve precise fault event location. Furthermore, dynamic monitoring and deep analysis are used to capture parameter anomalies in real time, constructing a multi-dimensional health status assessment matrix. The most important innovation of this invention lies in integrating multi-level analysis, time series analysis, and dynamic monitoring to achieve precise hardware fault location and risk prediction, significantly improving system stability and maintenance efficiency, and providing efficient fault diagnosis and management capabilities for complex hardware environments. Attached Figure Description

[0015] Figure 1 This is a flowchart of a server hardware information processing method according to the present invention.

[0016] Figure 2 This is a schematic diagram of a server hardware information processing method according to the present invention. Figure One .

[0017] Figure 3 This is a schematic diagram of a server hardware information processing method according to the present invention. Figure Two . Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and thoroughly described below with reference to the accompanying drawings. The described embodiments are merely some embodiments of the present invention.

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention. The present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments. The purpose and effects of the present invention will become clearer. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0020] like Figures 1-3 As shown, a server hardware information processing method in this embodiment may specifically include: S101, obtain original hardware information of a server DMI data table, perform multi-level structured decomposition on the hardware information through a decision tree classification algorithm, judge whether a current analysis level reaches a preset fine granularity threshold according to a level depth and a field complexity of a hardware category, if not, continue to analyze a lower level, and obtain a layered hardware data set containing a motherboard model, a CPU specification and a memory configuration.

[0021] Obtain hardware information in a server DMI data table, extract original fields of a motherboard model, a CPU specification and a memory configuration through data preprocessing, and obtain an initial hardware data set. The initial hardware data set is classified by using a decision tree classification algorithm, a hierarchical structure is generated according to a hardware category, and a classified hardware level data set is obtained. The level depth and the field complexity of the hardware level data set are calculated to judge whether a preset fine granularity threshold is reached, and a hierarchical decomposition judgment result is obtained. If the hierarchical decomposition judgment result does not reach the preset fine granularity threshold, the decision tree classification algorithm is continued to be used to perform lower-level decomposition on the hardware level data set, and a more refined hardware sub-level data set is obtained. Detailed fields of the motherboard model, the CPU specification and the memory configuration are extracted from the hardware sub-level data set, and a hardware data set containing a multi-level structure is generated. The integrity of the motherboard model, the CPU specification and the memory configuration is judged by performing field verification on the multi-level structure hardware data set, and a verified layered hardware data set is obtained. The verified layered hardware data set is grouped by using a clustering algorithm, and a final hardware classification data set is generated according to the hardware category and the configuration similarity.

[0022] It should be noted that obtaining hardware information in a server DMI data table is the basis of hardware classification.

[0023] For example, the original fields of the motherboard model, the CPU specification and the memory configuration are extracted from the DMI data table of the server. The motherboard model may contain, for example, “X11DPi-NT”, the CPU specification includes the model such as “Intel Xeon Gold 6230” and the core number such as 20 cores, and the memory configuration covers the capacity such as 64 GB and the frequency such as 3200 MHz. Data preprocessing needs to clean up invalid fields such as null values or inconsistent formats to ensure the accuracy of the generated initial hardware data set. This step ensures the reliability of subsequent classification.

[0024] For example, the initial hardware data set is classified by using a decision tree classification algorithm, and a hierarchical structure is generated according to the hardware category such as motherboard, CPU and memory. The decision tree is divided by features such as motherboard chipset type, CPU core number and memory frequency, and a hardware level data set is generated.

[0025] For example, motherboards are classified by chipset, such as C621 and C422, CPUs are classified by core count, such as high-performance and low-performance, and memories are classified by frequency, such as high-frequency and low-frequency. Such classification forms a clear hierarchical structure, facilitating subsequent analysis.

[0026] Specifically, the hierarchical depth and field complexity of the hardware hierarchical dataset are calculated to determine whether the preset fine-grained threshold is reached.

[0027] For example, the hierarchical depth is 3 layers and the field complexity is 10 fields, and the preset threshold requires a depth of at least 4 layers and a complexity of 12 fields. If not up to standard, continue to decompose the lower layer with a decision tree.

[0028] For example, the motherboard category is refined into subcategories that support different memory types, such as DDR4 or DDR5, and further hardware sub-hierarchical datasets are generated. This decomposition improves the fine-grained classification and the relevance of application scenarios.

[0029] In one possible implementation, detailed fields are extracted based on the hardware sub-hierarchical dataset.

[0030] For example, motherboard models are refined to manufacturers and version numbers, such as "Supermicro X11DPi-NT v2.0", CPU specifications include frequencies such as 2.1GHz, and memory configurations are refined to channel numbers such as 6 channels. The generated multi-level hardware dataset contains main category and subcategory information, with clear hierarchy. Field verification ensures data integrity, such as checking if the motherboard model is missing the version number, if the CPU specification contains the frequency, and if the memory configuration specifies the channel number. The verified hierarchical hardware dataset reduces the risk of incorrect classification.

[0031] For example, a clustering algorithm such as K-means is used to group the verified hierarchical hardware dataset, and a final hardware classification dataset is generated based on hardware category and configuration similarity.

[0032] For example, servers with similar configurations are divided into high-performance computing groups and storage optimization groups. The high-performance computing group may contain high-frequency CPUs and large-capacity memories, while the storage optimization group contains low-frequency CPUs and large-channel memories. The grouping result supports optimized allocation of hardware resources, improving server management efficiency.

[0033] It should be noted that the above process ensures the fine-grained and accurate hardware dataset through multi-level classification and verification.

[0034] For example, the refined hardware sub-hierarchical dataset can be used for fault diagnosis, integrity verification reduces error rate, and clustering and grouping optimize resource scheduling. These technical effects collectively improve the intelligent level of server hardware management.

[0035] S102, a sequence pattern mining algorithm is used to perform parameter extraction processing on the hierarchical hardware dataset. By identifying key parameter patterns and version number features in the hardware configuration, if BIOS version information is detected, the version number and release time are extracted, and if a device serial number is detected, the complete model identifier is extracted, obtaining a hardware configuration profile with complete parameter details.

[0036] The sequence pattern mining algorithm is used to perform initial processing on the hierarchical hardware dataset, extract key parameter patterns, and obtain a parameter pattern set. By analyzing the parameter pattern set, it is detected whether BIOS version information exists. If BIOS version information exists, the version number and release time are extracted, and a BIOS parameter set is generated. For the BIOS parameter set, a regular expression matching method is used to verify the format integrity of the version number and release time, and a format-verified BIOS parameter set is obtained. By analyzing the parameter pattern set, it is detected whether a device serial number exists. If a device serial number exists, the complete model identifier is extracted, and a device identifier set is generated. A clustering algorithm is used to group the format-verified BIOS parameter set and the device identifier set, and according to the similarity of hardware configurations, a classified hardware parameter set is generated. By comparing the classified hardware parameter set with the preset hardware configuration template, the parameter integrity is judged, and a hardware configuration profile after integrity verification is obtained. According to the hardware configuration profile after integrity verification, a data mapping method is used to associate the key parameter patterns with the fields in the hardware configuration profile, and a final hardware configuration profile is generated.

[0037] For example, when using a sequence pattern mining algorithm to perform initial processing on a hierarchical hardware dataset, the Apriori algorithm can be used to extract frequently occurring parameter patterns. Assuming that the hierarchical hardware dataset contains fields such as motherboard model, CPU specification, and memory configuration, sequence pattern mining will identify patterns such as "motherboard model X11DPi-NT is often associated with CPU model Intel Xeon Gold 6230".

[0038] Specifically, the algorithm analyzes the appearance order and relevance of fields in the dataset to generate a parameter pattern set, such as the combination of motherboard model and memory frequency 3200MHz. Such a pattern set provides a basis for subsequent analysis, ensuring that the regularity of key parameters is captured and improving the accuracy of hardware configuration analysis.

[0039] In one possible implementation, when detecting BIOS version information, the parameter pattern set can be traversed to find records containing BIOS-related fields.

[0040] For example, the dataset may contain an entry "BIOS version 2.1a, release time 2023-05-10". By extracting these fields, a BIOS parameter set is generated.

[0041] Preferably, the BIOS version number format is verified by using a regular expression matching method, such as requiring the version number to be in the form of "number.letter", such as 2.1a, and checking whether the release time is in the format of "YYYY-MM-DD". Assuming that the BIOS version of a certain record is "2.1", the regular expression detects that the format is incomplete, and is marked as invalid. The BIOS parameter set after format verification is generated to ensure data normalization and facilitate subsequent classification.

[0042] For example, when detecting the device serial number, the complete model identifier such as "SN12345678" can be extracted from the parameter mode set as the unique identifier of the device. When generating the device identifier set, it is necessary to ensure that the serial number is complete and unique.

[0043] For example, the serial number of a certain server is "SN12345678-X11DPi-NT", and the device identifier set is formed after extraction. Such an identifier set facilitates the tracking of hardware individuals and provides a basis for subsequent grouping.

[0044] In one possible implementation, the K-means clustering algorithm is used to group the BIOS parameter set and the device identifier set. Assuming that the servers are divided into a high-performance group and a low-power consumption group according to the similarity of hardware configurations.

[0045] For example, the high-performance group contains devices with BIOS version 2.1a and CPU frequency 2.8 GHz, and the low-power consumption group contains devices with BIOS version 1.9 and CPU frequency 1.8 GHz. The generated classified hardware parameter set reflects the configuration characteristics, facilitating resource allocation.

[0046] Specifically, when comparing the classified hardware parameter set with the preset hardware configuration template, it can be checked whether it contains necessary fields such as motherboard model, BIOS version, CPU frequency, etc. Assuming that the template requires at least 10 fields, and a certain classified hardware parameter set contains only 8 fields, it is marked as incomplete. Through integrity verification, the hardware configuration file is generated to ensure data reliability.

[0047] For example, when using a data mapping method to associate key parameter modes with the hardware configuration file, "motherboard model X11DPi-NT and memory frequency 3200 MHz" in the parameter mode set can be mapped to a specific field in the file, such as the memory type supported by the motherboard, DDR4. Such mapping generates the final hardware configuration file, clearly showing the relevance between hardware, which helps to optimize the efficiency of server management.

[0048] S103, according to the time-related field in the hardware configuration profile, a time series analysis algorithm is used to construct a timestamp resolution model of hardware events, by analyzing the time series characteristics of BIOS timestamps, device startup times, and configuration change records, the accurate time point and duration period of hardware event occurrence are determined, and the state change timeline of each hardware component is determined.

[0049] The time series analysis algorithm is used to preprocess the time field in the hardware configuration profile, BIOS timestamps, device startup times, and configuration change records are extracted from the time series data set, and an initial time series data set is generated. If the timestamp format in the initial time series data set is consistent, the initial time series data set is extracted by the sliding window method, and the time series feature set is obtained. According to the time series feature set, the K-means clustering algorithm is used to group the time series features, and the time point set of the hardware event is generated. By analyzing the time point set, the time interval between each hardware event and the adjacent event is calculated, and the duration period set of the hardware event is obtained. If the time interval in the duration period set exceeds the preset threshold, the abnormal hardware event is identified by the time series anomaly detection algorithm, and the abnormal event time set is obtained. According to the abnormal event time set and the time point set, the data mapping method is used to associate the time point and duration period of the hardware event with the hardware component, and the state change timeline of the hardware component is generated. By comparing the state change timeline with the time field in the hardware configuration profile, the integrity of the timeline is verified, and the final hardware event timeline is obtained.

[0050] For example, when preprocessing the time field in the hardware configuration profile using the time series analysis algorithm, BIOS timestamps, device startup times, and configuration change records can be identified and extracted first. Suppose the hardware configuration profile contains the running log of a server, which records the BIOS timestamp of a certain server as 2023-06-15 10:00:00, the device startup time as 2023-06-15 10:05:00, and the configuration change record as 2023-06-20 14:30:00. By analyzing these time fields, an initial time series data set is generated, ensuring that the time format is uniform as YYYY-MM-DD HH:MM:SS, which facilitates subsequent analysis. This uniform format processing helps to reduce errors caused by inconsistent formats.

[0051] In one possible implementation, when the initial time series data set is extracted by the sliding window method, the window size can be set to 24 hours and the step size to 1 hour to extract the trend in the time series. Suppose the BIOS timestamp sequence of a certain server shows that it is updated once a day, the sliding window can capture the regularity of the update frequency, generate a time series feature set, and include update intervals, timestamp density, and other features. These features reflect the occurrence mode of hardware events and provide a basis for subsequent grouping.

[0052] For example, when grouping the time series feature set using the K-means clustering algorithm, the hardware events can be divided into high-frequency event groups and low-frequency event groups according to the timestamp density and event frequency. Assuming that the high-frequency event group contains daily updated BIOS timestamps, and the low-frequency event group contains weekly configuration change records. Through clustering, the time point set of hardware events is generated, such as the time points of the high-frequency group are 2023-06-15 10:00:00, 2023-06-16 10:00:00, etc. This grouping method facilitates the identification of the time distribution characteristics of hardware events.

[0053] In one possible implementation, when calculating the time interval in the time point set of the hardware event, the time point set can be traversed to calculate the time difference of adjacent events. Assuming that the BIOS update time points of a certain server are 2023-06-15 10:00:00 and 2023-06-16 10:00:00, and the time interval is 24 hours. If the preset threshold is 48 hours, the interval is normal. After generating the duration cycle set, if a certain configuration change interval is 72 hours, which exceeds the threshold, it is marked as a potential anomaly. Through a time series anomaly detection algorithm, such as a method based on mean and standard deviation, abnormal hardware events are identified, and an abnormal event time set is generated, such as a configuration change record at 2023-06-20 14:30:00. This anomaly detection helps to discover irregular hardware behavior.

[0054] For example, when using the data mapping method to associate the time points and duration cycles of hardware events with hardware components, the BIOS update time in the time point set can be mapped to a specific motherboard model, such as X11DPi-NT. Assuming that the BIOS update cycle of a certain motherboard is 24 hours, the generated state change timeline shows that it is running stably. Finally, by comparing the state change timeline with the time field in the hardware configuration profile, it is verified whether it contains all necessary time points. Assuming that the profile requires to contain the startup time and change records, if the timeline is missing the change records, it is marked as incomplete. This verification ensures the integrity of the hardware event timeline, facilitating the tracking of the regularity of hardware state changes.

[0055] S104, by associating and matching the timeline information with the hardware event records in the system log, using a time window sliding mechanism to accurately locate the hardware failure events, if the difference between the hardware event timestamp and the failure log timestamp is less than a preset threshold, an association relationship is established, and a failure event mapping table containing accurate time information and detailed configuration parameters is obtained.

[0056] Hardware event records and timeline information are obtained through system logs, a time window sliding mechanism is adopted to preliminarily match timestamps, and an initial fault event set is obtained. If the timestamp difference in the initial fault event set is less than a preset threshold, the association between the hardware event and the fault log is established through a timestamp matching algorithm, and an associated event set is generated. According to the associated event set, the configuration parameters of each fault event are extracted, and a data integration method is used to generate a fault event mapping table containing time information and configuration parameters. Through the fault event mapping table, a time series clustering algorithm is used to group fault events, and a fault event classification set is obtained. If there is a timestamp overlap in the fault event classification set, the time window is re-divided through a time window adjustment mechanism, and an optimized fault event set is generated. According to the optimized fault event set, a data mapping method is used to associate fault events with hardware components, and a hardware component fault timeline is obtained. Through the hardware component fault timeline, the consistency of the timestamp and the configuration parameter is verified, and a final fault event timeline is obtained.

[0057] For example, system logs usually contain information such as hardware running status, fault alarms, and timestamps, which can be used as basic data for analyzing hardware events.

[0058] In one possible implementation, when obtaining system logs, hardware events such as CPU overheating or memory failure records can be extracted from server running records. Assuming that a server log shows that the timestamp of a CPU overheating event is 2023-07-01 09:00:00, and the timestamp of a memory failure is 2023-07-02 15:30:00. These timestamps provide a data basis for subsequent time window sliding. In the time window sliding mechanism, the window size can be set to 12 hours, the step size is 1 hour, and the timestamp is preliminarily matched.

[0059] For example, when scanning the log with a sliding window, a CPU overheating event at 2023-07-01 09:00:00 and another fault event at 2023-07-01 20:00:00 are detected, and the time difference is 11 hours, which is less than the preset threshold of 24 hours, indicating that there may be an association. Through the timestamp matching algorithm, these events can be associated with the fault description in the log to generate an initial fault event set.

[0060] For example, the CPU overheating event can be matched with the "fan speed abnormality" record in the log to generate an associated event set.

[0061] Specifically, when extracting configuration parameters from the associated event set, the hardware state at the time of failure can be obtained, such as a CPU frequency of 3.2 GHz and a fan speed of 2000 RPM. Through a data integration method, the timestamp 2023-07-01 09:00:00 is combined with these parameters to generate a fault event mapping table that records the correspondence between fault time and hardware configuration. This mapping table provides a clear data structure for subsequent analysis.

[0062] In one embodiment, when grouping the fault event mapping table using a time series clustering algorithm, events can be divided into high-frequency fault groups and low-frequency fault groups according to the timestamp density and fault frequency.

[0063] For example, CPU overheating events occur daily and are classified into high-frequency groups; memory faults occur once a week and are classified into low-frequency groups. If overlapping timestamps are found, such as the CPU overheating timestamp 2023-07-01 09:00:00 coinciding with the fan fault timestamp, the window size is reduced to 6 hours through a time window adjustment mechanism, reclassified, and an optimized fault event set is generated to ensure clear event differentiation.

[0064] For example, the optimized fault event set can be used to associate fault events with hardware components. Assuming that the CPU overheating event is mapped to the processor model XeonE5-2699, a hardware component fault timeline is generated to show the fault distribution of this model from 2023-07-01 to 2023-07-03. When verifying the timeline, check the consistency of timestamps and configuration parameters, such as confirming that the CPU frequency record in the log matches the fault time. If missing fan speed records are found, mark them as incomplete. This verification ensures that the timeline accurately reflects the hardware state, facilitating the tracking of fault patterns.

[0065] In one possible implementation, the fault timeline can be further used to analyze the hardware operation mode.

[0066] For example, the timeline shows that the overheating events of XeonE5-2699 are concentrated in high-load periods, which can be inferred to be related to the design of the heat dissipation system. This analysis helps to optimize hardware maintenance strategies and improve system stability.

[0067] S105, acquire the dynamic change mode in the fault event mapping table, through real-time monitoring of the numerical fluctuation and state transition of the hardware configuration parameters, if the parameter value is detected to be out of the normal range or the state jumps abnormally, the deep analysis mechanism is triggered to extract the fine-grained information of the related hardware components, and the specific fault hardware location and configuration abnormality details are determined.

[0068] The time series data is obtained from the hardware configuration parameters through a real-time monitoring mechanism to determine the dynamic pattern of parameter value fluctuations and state changes. If the parameter value fluctuations exceed the preset threshold range or the state changes abnormally, a deep analysis mechanism is triggered to extract fine-grained information of the relevant hardware components. Based on the fine-grained information, a classification algorithm is used to locate the fault of the hardware components to obtain the fault hardware location. Through the fault hardware location, the associated configuration parameter records are obtained to extract configuration abnormality details. If the configuration abnormality details overlap with the time series data in terms of timestamps, a time series analysis algorithm is used to adjust the division of abnormal events to obtain an optimized abnormal event set. Based on the optimized abnormal event set, a data mapping method is used to integrate the fault hardware location and the configuration abnormality details to generate a fault analysis mapping table. Through the fault analysis mapping table, the consistency of the dynamic change pattern and the time series data is verified to determine the final fault location result.

[0069] For example, the real-time monitoring mechanism can continuously collect time series data of hardware configuration parameters through sensors and log systems to build a dynamic change pattern. Suppose the monitoring system of a server records CPU temperature, memory usage, and disk IO rate every second. The time series data shows that the CPU temperature rises from 60°C to 85°C from 10:00:00 to 10:30:00 on August 1, 2023, exceeding the preset threshold of 80°C. After triggering the deep analysis mechanism, the system extracts CPU-related fine-grained information, such as core frequency 3.5GHz and load 90%. Through a classification algorithm such as decision tree, the relationship between temperature abnormality and high load is analyzed to locate the fault to the CPU module. After the fault hardware location is determined, the configuration parameter records are obtained to find that the cooling module fan speed is only 1500RPM, lower than the normal 2000RPM, confirming the configuration abnormality.

[0070] Specifically, the time series analysis algorithm can be used to handle the timestamp overlap situation. Suppose the CPU temperature abnormality timestamp 2023-08-01 10:15:00 coincides with the fan speed abnormality record, the algorithm divides the abnormal events into 10-minute granularity through a time window to generate an optimized abnormal event set. The optimized abnormal event set shows that the CPU temperature abnormality and the fan speed reduction are highly related in time. Using a data mapping method, the CPU module fault location and the fan speed abnormality are integrated to generate a fault analysis mapping table, recording the fault time, hardware location, and configuration parameters, such as "2023-08-01 10:15:00, CPU module, fan speed 1500RPM".

[0071] In one embodiment, when verifying the dynamic change mode, the consistency of the time series data with the fault analysis mapping table is checked. Assuming that the time series shows that the CPU temperature continues to rise during the high load period, the mapping table confirms that the abnormal fan speed causes insufficient heat dissipation, and the verification result is consistent. The final fault location result points to the CPU cooling system, prompting the need to check the fan hardware or optimize the load scheduling. This method integrates multi-dimensional data to ensure accurate fault location and reduce misjudgment.

[0072] For example, when the classification algorithm locates the fault, it can combine historical data and real-time data to support the positioning result from multiple perspectives. Assuming that historical data shows that CPU temperature abnormalities are often accompanied by fan speed decline, and real-time data further confirms that the timestamps of the two overlap, the classification algorithm prioritizes the investigation of the cooling system. Optimizing the set of abnormal events refines the time granularity, distinguishes between independent faults and cascading faults, and improves analysis efficiency. The fault analysis mapping table provides a clear data structure for quickly tracing the root cause of the problem and optimizing the maintenance process.

[0073] S106, according to the configuration exception details and historical fault mode, a multi-dimensional hardware health state evaluation matrix is constructed, by comprehensively analyzing the hardware category, model parameters, timestamp information and configuration change trend, the overall operation state and potential risk point of the current hardware system are judged, and the complete hardware information analysis result supporting accurate fault location is obtained.

[0074] From the hardware configuration parameters, time series data is obtained, combined with timestamp information, time series analysis algorithm is used to extract configuration change trend, and hardware parameter dynamic characteristics are obtained. According to the configuration change trend, combined with the hardware category and model parameters, a multi-dimensional hardware health state evaluation matrix is constructed, and a health state score is generated. If the health state score is lower than the preset threshold, the fault characteristics matching the hardware category and model parameters are extracted from the historical fault mode, and the potential risk point is determined. For the potential risk point, a classification algorithm is used to analyze the hardware health state evaluation matrix, and a preliminary fault location result is obtained. According to the preliminary fault location result, the associated fine-grained configuration data is extracted from the hardware configuration parameters, and the hardware information analysis result is generated. Through the hardware information analysis result, combined with the timestamp information, a clustering algorithm is used to group the configuration change trend, and the time distribution characteristics of the abnormal event are judged. According to the time distribution characteristics of the abnormal event, the hardware health state evaluation matrix is updated, and the optimized fault location result is obtained.

[0075] For example, when acquiring time series data of hardware configuration parameters, the server's memory usage, disk read / write speed, and network bandwidth can be collected in real time through sensors, and combined with timestamps to generate continuous data streams. Assuming that a server records memory usage every minute, and during the period from 2023-08-0209:00:00 to 09:30:00, the memory usage rises from 50% to 95%, the time series analysis algorithm can identify this rapid upward trend and extract the dynamic feature as "memory usage running at a high level continuously". When building a multi-dimensional hardware health status evaluation matrix, combine hardware categories such as "memory module" and model parameters such as "DDR4 32GB", and integrate multiple dimensions such as memory usage, response delay, and error rate in time series data to form an evaluation matrix.

[0076] For example, the matrix records memory usage of 95%, response delay of 200ms, and error rate of 0.5%, and generates a health status score of 75 through weighted calculation, which is lower than the preset threshold of 80, indicating a potential risk.

[0077] Specifically, from the historical failure mode, extract the failure features related to DDR4 memory, such as "response delay increases under high usage". Through comparison, it is determined that the potential risk point is that the memory module may be overloaded and cause performance degradation. Using classification algorithms such as support vector machines to analyze the evaluation matrix, combined with memory usage and delay data, the fault is preliminarily located as a performance bottleneck of the memory module.

[0078] In one embodiment, fine-grained data such as the running frequency of the memory module 2400MHz and the voltage 1.2V are extracted from the configuration parameters to generate hardware information analysis results. Combined with timestamps, clustering algorithms divide the time distribution of memory usage anomalies into two groups: sudden high load from 09:15:00 to 09:20:00 and continuous high load from 09:20:00 to 09:30:00. The time distribution characteristics of abnormal events indicate that the fault may be triggered by a sudden task and continue to affect.

[0079] For example, when updating the health status evaluation matrix, add the clustering analysis results, and optimize the fault location to "the memory module fails to recover after sudden high load". Through multi-dimensional data integration, such as matching historical failure patterns with real-time data, it is confirmed that the root cause of the fault is that the memory module has insufficient resource allocation under high load, prompting the need to optimize task scheduling or upgrade memory capacity. This method ensures the comprehensiveness and accuracy of fault location through multi-level analysis.

[0080] It should be noted that when dealing with complex data, the time series analysis algorithm can extract trends through sliding time windows to reduce noise interference.

[0081] For example, a 5-minute window is used to analyze the fluctuation trend of memory usage, and more accurate dynamic features are generated. The classification algorithm verifies the fault location result from multiple aspects by combining historical and real-time data to enhance reliability. The clustering algorithm for grouping abnormal events helps to distinguish transient abnormalities from persistent abnormalities, improving analysis efficiency.

[0082] A server hardware information processing apparatus includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor of the claim, wherein the processor executes the computer program to implement the steps of the server hardware information processing method.

[0083] The above embodiments are only used to illustrate the technical solutions of the present application and are not limited. The present application has been described in detail only with reference to the preferred embodiments. Those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and all should be covered in the scope of the claims of the present application.

Claims

1. A method for processing server hardware information, characterized in that, include: The original hardware information of the server DMI data table is obtained, and the hardware information is decomposed into a multi-level structure using a decision tree classification algorithm. Based on the hierarchical depth and field complexity of the hardware category, it is determined whether the current parsing level has reached the preset fine-grained threshold. If it has not reached the threshold, the parsing continues to the next level to obtain a hierarchical hardware dataset containing motherboard model, CPU specifications, and memory configuration. The hierarchical hardware dataset is processed by a sequence pattern mining algorithm to extract parameters. By identifying key parameter patterns and version number features in the hardware configuration, if BIOS version information is detected, its version number and release time are extracted. If device serial number is detected, the complete model identifier is extracted, thus obtaining a hardware configuration file with complete parameter details. Based on the time-related fields in the hardware configuration file, a time series analysis algorithm is used to construct a timestamp parsing model for hardware events. By analyzing the time series characteristics of BIOS timestamps, device startup times, and configuration change records, the precise time point and duration of hardware events are determined, and the time axis of state changes for each hardware component is identified. The timeline information is correlated and matched with hardware event records in the system log. A time window sliding mechanism is used to accurately locate hardware fault events. If the difference between the hardware event timestamp and the fault log timestamp is less than a preset threshold, an association is established to obtain a fault event mapping table containing accurate time information and detailed configuration parameters. The dynamic change pattern in the fault event mapping table is obtained. By monitoring the numerical fluctuations and state transitions of hardware configuration parameters in real time, if the parameter value is detected to be outside the normal range or the state changes abnormally, a deep analysis mechanism is triggered to extract fine-grained information from the relevant hardware components to determine the specific location of the faulty hardware and the details of the abnormal configuration. Based on the configuration anomaly details and historical failure modes, a multi-dimensional hardware health status assessment matrix is ​​constructed. By comprehensively analyzing hardware category, model parameters, timestamp information, and configuration change trends, the overall operating status and potential risk points of the current hardware system are determined, resulting in complete hardware information analysis results that support accurate fault location.

2. The server hardware information processing method according to claim 1, characterized in that, The process involves obtaining the raw hardware information from the server's DMI data table, and then performing a multi-level structured decomposition of the hardware information using a decision tree classification algorithm. Based on the hierarchical depth and field complexity of the hardware categories, it is determined whether the current parsing level has reached a preset fine-grained threshold. If not, parsing continues to the next lower level to obtain a hierarchical hardware dataset containing motherboard model, CPU specifications, and memory configuration, including: Obtain hardware information from the server DMI data table, and extract the original fields of motherboard model, CPU specifications and memory configuration through data preprocessing to obtain the initial hardware dataset; The initial hardware dataset is classified using a decision tree classification algorithm. A hierarchical structure is generated based on the hardware categories to obtain the classified hardware hierarchical dataset. By calculating the hierarchical depth and field complexity of the hardware-level dataset, it is determined whether the preset fine-grained threshold has been reached, and the hierarchical decomposition judgment result is obtained. If the hierarchical decomposition result does not reach the preset fine-grained threshold, the decision tree classification algorithm will continue to be used to decompose the hardware hierarchical dataset into a lower level to obtain a more refined hardware sub-hierarchical dataset. Detailed fields of motherboard model, CPU specifications and memory configuration are extracted from the hardware sub-level dataset to generate a hardware dataset with a multi-level structure. By performing field validation on the multi-level hardware dataset, the integrity of the motherboard model, CPU specifications, and memory configuration is determined, and the validated hierarchical hardware dataset is obtained. Clustering algorithms are used to group the validated hierarchical hardware dataset, and the final hardware classification dataset is generated based on hardware category and configuration similarity.

3. The server hardware information processing method according to claim 1, characterized in that, The sequence pattern mining algorithm is used to extract parameters from the hierarchical hardware dataset. By identifying key parameter patterns and version number features in the hardware configuration, if BIOS version information is detected, its version number and release time are extracted; if a device serial number is detected, the complete model identifier is extracted, thus obtaining a hardware configuration file with complete parameter details, including: A sequence pattern mining algorithm is used to perform initial processing on the hierarchical hardware dataset to extract key parameter patterns and obtain a parameter pattern set. The presence of BIOS version information is detected by analyzing the parameter mode set. If BIOS version information exists, extract the version number and release time to generate a BIOS parameter set; For the BIOS parameter set, a regular expression matching method is used to verify the format integrity of the version number and release time, and obtain the format-verified BIOS parameter set; The presence of the device serial number is detected by analyzing the parameter pattern set. If the device serial number exists, extract the complete model identifier and generate a device identifier set; Clustering algorithms are used to group the format-verified BIOS parameter set and device identifier set, and a classified hardware parameter set is generated based on hardware configuration similarity. By comparing the categorized hardware parameter set with the preset hardware configuration template, the integrity of the parameters is judged, and the hardware configuration file after integrity verification is obtained. Based on the hardware configuration file after integrity verification, a data mapping method is used to associate key parameter patterns with fields in the hardware configuration file to generate the final hardware configuration file.

4. The server hardware information processing method according to claim 1, characterized in that, The process involves constructing a timestamp parsing model for hardware events using time-related fields in the hardware configuration file, employing a time series analysis algorithm. By analyzing the time series characteristics of BIOS timestamps, device boot times, and configuration change records, the precise time point and duration of hardware events are determined, and the timeline of state changes for each hardware component is identified. This includes: A time series analysis algorithm is used to preprocess the time field in the hardware configuration file, and BIOS timestamps, device boot time and configuration change records are extracted from the time series dataset to generate an initial time series dataset; If the timestamp formats in the initial time series dataset are consistent, then the sliding window method is used to extract features from the initial time series dataset to obtain the time series feature set; Based on the time series feature set, the K-means clustering algorithm is used to group the time series features to generate a time point set of hardware events; By analyzing the time point set, the time interval between the time point of each hardware event and the adjacent events is calculated, and the duration period set of the hardware events is obtained. If the time interval of the continuous periodic set exceeds the preset threshold, the abnormal hardware event is identified by the time series anomaly detection algorithm to obtain the abnormal event time set; Based on the abnormal event time set and time point set, a data mapping method is used to associate the time point and duration of hardware events with hardware components, generating a timeline of state changes of hardware components; By comparing the state change timeline with the time fields in the hardware configuration file, the integrity of the timeline is verified, and the final hardware event timeline is obtained.

5. A server hardware information processing method according to claim 1, characterized in that, The process involves associating and matching the timeline information with hardware event records in the system log, employing a time window sliding mechanism to accurately locate hardware fault events. If the difference between the hardware event timestamp and the fault log timestamp is less than a preset threshold, an association is established, resulting in a fault event mapping table containing precise time information and detailed configuration parameters, including: Hardware event records and timeline information are obtained from system logs, and a time window sliding mechanism is used to initially match timestamps to obtain an initial set of fault events; If the timestamp difference in the initial fault event set is less than a preset threshold, the association between hardware events and fault logs is established through a timestamp matching algorithm to generate an associated event set. Based on the associated event set, the configuration parameters of each fault event are extracted, and a fault event mapping table containing time information and configuration parameters is generated using a data integration method. Using a fault event mapping table, a time series clustering algorithm is employed to group fault events, resulting in a fault event classification set. If timestamps overlap in the fault event classification set, the time window is re-divided through the time window adjustment mechanism to generate an optimized fault event set; Based on the optimized fault event set, a data mapping method is used to associate the fault events with hardware components to obtain the hardware component fault timeline. By verifying the consistency between timestamps and configuration parameters through the hardware component failure timeline, the final failure event timeline is obtained.

6. The server hardware information processing method according to claim 1, characterized in that, The process of obtaining dynamic change patterns in the fault event mapping table involves real-time monitoring of hardware configuration parameter fluctuations and state transitions. If a parameter value exceeds the normal range or an abnormal state transition is detected, a deep analysis mechanism is triggered to extract fine-grained information from the relevant hardware components, determining the specific location of the faulty hardware and the details of the configuration anomaly, including: By using a real-time monitoring mechanism, time-series data is obtained from hardware configuration parameters to determine the dynamic patterns of parameter value fluctuations and state changes. If the parameter value fluctuates beyond the preset threshold range or the state changes abnormally, the deep analysis mechanism is triggered to extract fine-grained information about the relevant hardware components. Based on fine-grained information, a classification algorithm is used to locate the faulty hardware component and obtain the location of the faulty hardware. By identifying the location of the faulty hardware, obtain the associated configuration parameter records and extract the details of the configuration anomaly. If the configuration exception details overlap with the time series data in terms of timestamps, a time series analysis algorithm is used to adjust the division of exception events and obtain an optimized set of exception events. Based on the optimized abnormal event set, a data mapping method is used to integrate the location of the faulty hardware with the details of the configuration anomaly to generate a fault analysis mapping table; By using a fault analysis mapping table, the consistency between the dynamic change pattern and the time series data is verified, and the final fault location result is determined.

7. A server hardware information processing method according to claim 1, characterized in that, Based on the configuration anomaly details and historical failure modes, a multi-dimensional hardware health status assessment matrix is ​​constructed. By comprehensively analyzing hardware category, model parameters, timestamp information, and configuration change trends, the overall operating status and potential risk points of the current hardware system are determined, resulting in complete hardware information parsing results that support accurate fault location, including: Time series data is obtained from hardware configuration parameters, combined with timestamp information, and time series analysis algorithms are used to extract configuration change trends and obtain dynamic characteristics of hardware parameters; Based on configuration change trends, and combined with hardware category and model parameters, a multi-dimensional hardware health status assessment matrix is ​​constructed to generate a health status score. If the health status score is lower than the preset threshold, then extract the fault features that match the hardware category and model parameters from the historical fault modes to identify potential risk points. For potential risk points, a classification algorithm is used to analyze the hardware health status assessment matrix to obtain preliminary fault location results; Based on the preliminary fault location results, fine-grained configuration data related to the hardware configuration parameters are extracted to generate hardware information parsing results; By analyzing the hardware information and combining it with timestamp information, a clustering algorithm is used to group the configuration change trends and determine the time distribution characteristics of abnormal events. Based on the temporal distribution characteristics of abnormal events, the hardware health status assessment matrix is ​​updated to obtain optimized fault location results.

8. A server hardware information processing apparatus, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor as described in the claims, characterized in that, When the processor executes the computer program, it implements the steps of the server hardware information processing method as described in any one of claims 1 to 7.

Images