Server configuration data processing method, electronic device, storage medium and program product

Abstract

The present application discloses a server configuration data processing method, electronic device, storage medium, and program product, relating to the field of computer technology, and more particularly, to a method for performing targeted parsing of raw data streams of different communication protocols through a preset protocol adapter, converting them into configuration data packets corresponding to the communication protocols, thereby achieving effective integration of multi-source heterogeneous data and providing unified basic data for subsequent processing. A configuration information recognition model can extract unstructured text data from standardized data packets and convert it into structured configuration information. This conversion process transforms unstructured data, which was originally difficult to directly analyze and utilize, into queryable, statistically tractable, and correlated structured configuration information, thereby enabling rapid and accurate identification of configuration data.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a server configuration data processing method, electronic equipment, storage medium, and program product. Background Art

[0002] With the rapid development of information technology, server clusters have become the core infrastructure supporting various businesses, cloud services, and large-scale data processing. To meet diverse business needs, modern server clusters are often composed of different models of equipment and must be compatible with multiple communication protocols and data formats. This results in the configuration data within the cluster being multi-sourced and heterogeneous, with a large amount of information existing in the form of unstructured text.

[0003] In related technologies, a series of clear logical rules are usually preset manually to filter data that meets the rules from the configuration data, thereby identifying the target configuration data.

[0004] However, when faced with such multi-source heterogeneous data, related technologies have the problem of insufficient semantic recognition of unstructured text, and are prone to recognition errors due to format differences between different communication protocols, ultimately leading to inefficient configuration data recognition and low accuracy. Summary of the Invention

[0005] The present application provides a server configuration data processing method, electronic device, storage medium and program product to at least solve the problems of low efficiency and low accuracy in configuration data recognition in related technologies.

[0006] This application provides a method for processing server configuration data, including:

[0007] Get the original data stream of the server cluster;

[0008] Parsing the original data stream through a preset protocol adapter to obtain a configuration data packet corresponding to each protocol, wherein the configuration data packet includes unstructured text data;

[0009] Clean the configuration data packet to obtain a standardized configuration data table;

[0010] Cache and store the standardized configuration data table to obtain a stored standardized data packet;

[0011] Inputting the standardized data packet into the configuration information recognition model, so that the configuration information recognition model extracts unstructured text data from the standardized data packet and converts the unstructured text data into structured configuration information;

[0012] Output structured configuration information.

[0013] The present application also provides a server configuration data processing device, comprising:

[0014] The acquisition module is used to obtain the original data stream of the server cluster;

[0015] A processing module, configured to parse the original data stream through a preset protocol adapter to obtain a configuration data packet corresponding to each protocol, wherein the configuration data packet includes unstructured text data;

[0016] The processing module is also used to clean the configuration data packet to obtain a standardized configuration data table;

[0017] A storage module, configured to cache and store the standardized configuration data table to obtain a stored standardized data packet;

[0018] an identification module, configured to input the standardized data packet into the configuration information identification model, so that the configuration information identification model extracts unstructured text data from the standardized data packet and converts the unstructured text data into structured configuration information;

[0019] Output module, used to output structured configuration information.

[0020] The present application also provides an electronic device, comprising: a memory for storing a computer program; and a processor for implementing the steps of any of the above methods when executing the computer program.

[0021] The present application also provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned server configuration data processing methods are implemented.

[0022] The present application also provides a computer program product, including a computer program, which implements the steps of any of the above-mentioned server configuration data processing methods when executed by a processor.

[0023] This application uses a pre-set protocol adapter to perform targeted parsing of raw data streams from different communication protocols and convert them into configuration data packets for the corresponding communication protocols. This effectively integrates multi-source heterogeneous data and provides unified basic data for subsequent processing. The configuration information recognition model can extract unstructured text data from standardized data packets and convert it into structured configuration information. This conversion process transforms unstructured data, which was originally difficult to directly analyze and utilize, into queryable, statistically tractable, and correlated structured configuration information, thereby enabling rapid and accurate identification of configuration data. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0025] Figure 1 Schematic diagram of the process of the server configuration data processing method provided in the embodiment of the application Figure 1 ;

[0026] Figure 2 Schematic diagram of the process of the server configuration data processing method provided in the embodiment of the application Figure 2 ;

[0027] Figure 3 A schematic diagram of the structure of a server configuration data processing device provided in an embodiment of the present application;

[0028] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0029] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0030] It should be noted that, in the description of this application, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. The terms "first," "second," etc., in this application are used to distinguish similar objects, and are not used to describe a particular order or sequence.

[0031] With the rapid development of information technology, server clusters have become the core infrastructure supporting various businesses, cloud services, and large-scale data processing. To meet diverse business needs, modern server clusters are often composed of different models of equipment and must be compatible with multiple communication protocols and data formats. This results in the configuration data within the cluster being multi-sourced and heterogeneous, with a large amount of information existing in the form of unstructured text.

[0032] In related technologies, a series of clear logical rules are usually preset manually to filter data that meets the rules from the configuration data, thereby identifying the target configuration data.

[0033] However, when faced with such multi-source heterogeneous data, related technologies have the problem of insufficient semantic recognition of unstructured text, and are prone to recognition errors due to format differences between different communication protocols, ultimately leading to inefficient configuration data recognition and low accuracy.

[0034] To address the aforementioned technical issues, the inventors propose the following technical concept: Given the varying communication protocols used by server cluster configuration data, the inventors propose using a pre-set protocol adapter compatible with various communication protocols to accurately parse configuration data for each protocol. To address the insufficient semantic recognition of unstructured text, a configuration information recognition model can be employed to transform and recognize unstructured text data, enabling rapid and accurate recognition of configuration data.

[0035] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0036] Figure 1 Schematic diagram of the process of the server configuration data processing method provided in the embodiment of the application Figure 1 ,like Figure 1 As shown, the server configuration data processing method provided in the embodiment of the present application is applied to any electronic device, and the method includes:

[0037] S101. Obtain the original data stream of the server cluster.

[0038] Specifically, the original data stream includes but is not limited to command line output data stream, Application Programming Interface (API) response data stream, and log data stream.

[0039] S102: Parse the original data stream through a preset protocol adapter to obtain configuration data packets corresponding to each protocol, wherein the configuration data packets include unstructured text data.

[0040] The preset protocol adapter is adapted to multiple communication protocols, including but not limited to Secure Shell (SSH), Simple Network Management Protocol (SNMP), and Hypertext Transfer Protocol (HTTP).

[0041] For example, for a raw data stream using SSH as the communication protocol, the preset protocol adapter executes a preset program in the preset instruction library to parse the raw data stream to obtain the hardware parameters of the server cluster. For a raw data stream using SNMP as the communication protocol, the preset protocol adapter polls and parses the port traffic of the server cluster's switches using object identifiers (e.g., 1.3.6.1.2.1.31.1.1.1.6 corresponds to the number of bytes received on the interface). For a raw data stream using HTTP as the communication protocol, the preset protocol adapter parses and obtains the server cluster's container orchestration information by calling the corresponding API.

[0042] S103: Clean the configuration data packet to obtain a standardized configuration data table.

[0043] Specifically, missing fields in the configuration data packet are identified and filled or deleted accordingly to obtain a standardized configuration data table.

[0044] S104: Cache and store the standardized configuration data table to obtain a stored standardized data packet.

[0045] Specifically, a memory-disk hybrid cache method is adopted to cache and store the standardized configuration data table to obtain a stored standardized data packet.

[0046] S105 : Input the standardized data packet into the pre-trained configuration information recognition model, so that the pre-trained configuration information recognition model extracts unstructured text data from the standardized data packet and converts the unstructured text data into structured configuration information.

[0047] Pre-training is a deep learning model training strategy that uses large-scale datasets to initially train the model, enabling it to learn common feature representations. This process is similar to the basic learning stage humans go through before acquiring new knowledge, accumulating experience through extensive reading and observation.

[0048] Specifically, a standardized data packet is input into a pre-trained configuration information recognition model so that the pre-trained configuration information recognition model extracts unstructured text data from the standardized data packet (for example, "software service service:"Nginx", configuration parameter param: "worker_connections", parameter value value: 2048"). When the type of the standardized data packet is a configuration change, the unstructured text data is converted into description text data (for example, "Nginxworker_connections=2048"), and the entity type in the description text data (for example, software service type, configuration parameter type, or parameter value type) is identified. Dependency analysis is performed, and after identification, the description text data is converted into structured configuration information (for example, {"software": {"name":"MySQL","version":"8.0.26"}, "parameter": {"max_connections":500}}).

[0049] Among them, dependency analysis is used to establish the relationship between entities and clarify which software the parameters belong to.

[0050] S106: Output structured configuration information.

[0051] If possible, the structured configuration information is output to the automated operation and maintenance system so that the automated operation and maintenance system optimizes and manages the configuration of the server cluster according to the structured configuration information.

[0052] The server configuration data processing method provided in the embodiments of the present application, through a preset protocol adapter, can perform targeted parsing of raw data streams of different communication protocols and convert them into configuration data packets corresponding to the corresponding communication protocols. This effectively integrates multi-source heterogeneous data and provides unified basic data for subsequent processing. The configuration information recognition model can extract unstructured text data from standardized data packets and convert it into structured configuration information. This conversion process transforms unstructured data, which was originally difficult to directly analyze and utilize, into structured configuration information that can be searched, counted, and correlated, thereby enabling rapid and accurate identification of configuration data.

[0053] Figure 2 Schematic diagram of the process of the server configuration data processing method provided in the embodiment of the application Figure 1 ,like Figure 2 As shown, the method includes:

[0054] S201: Obtain the original data stream of the server cluster.

[0055] S202: Parse the original data stream through a preset protocol adapter to obtain configuration data packets corresponding to each protocol, wherein the configuration data packets include unstructured text data.

[0056] S203: Clean the configuration data packet to obtain a standardized configuration data table.

[0057] In a possible implementation, the target field in the configuration data packet is extracted to obtain target configuration data, and duplicate data in the target configuration data is removed to obtain a standardized configuration data table.

[0058] Specifically, use a regular expression (for example, re.findall(r'(\d+\.\d+\.\d+\.\d+). ?HTTP\ / \d\.\d\"\s(\d{3})', log), used to match and extract Internet addresses and status codes, matches and extracts the target field in the configuration data packet to obtain the target configuration data. Furthermore, a context-aware algorithm is used to associate contextual data in the target configuration data (for example, log entries with the same X-Request-ID are associated, and only the first entry is retained when the time window is less than 100ms). Duplicate data is identified and removed to obtain a standardized configuration data table.

[0059] S204: Cache and store the standardized configuration data table to obtain a stored standardized data packet.

[0060] In one possible implementation, the data in the standardized configuration data table is divided into n priority levels, where n is an integer greater than 1. The data in the standardized configuration data table is cached and stored in a priority queue according to the priority level to obtain stored standardized data packets.

[0061] Specifically, based on data urgency, business impact, and processing timeliness requirements, the data in the standardized configuration data table is divided into n priority levels, where n is an integer greater than 1. Based on the priority level, the data in the standardized configuration data table is cached and stored in a priority queue to obtain stored standardized data packets.

[0062] For example, based on the data urgency, business impact scope and processing timeliness requirements, the data in the standardized configuration data table is divided into three priority levels, namely high priority level, medium priority level and low priority level. The high priority level corresponds to processor temperature information, etc., the medium priority level corresponds to configuration file modification information, and the low priority level corresponds to memory usage information, etc.

[0063] Possibly, when the queue depth is greater than a preset queue depth threshold, the data in the standardized configuration data table is reduced to the lowest priority level.

[0064] S205 : Input the standardized data packet into the pre-trained configuration information recognition model, so that the pre-trained configuration information recognition model extracts unstructured text data from the standardized data packet and converts the unstructured text data into structured configuration information.

[0065] S206: Output structured configuration information.

[0066] S207: Extract a keyword set from the structured configuration information.

[0067] For example, keywords include but are not limited to parameter names and software names.

[0068] S208. Construct a high-dimensional semantic similarity matrix based on the keyword set.

[0069] Specifically, based on the keyword set, a high-dimensional semantic similarity matrix is ​​constructed through the word vector model (Word to Vector, Word2Vec) , as shown in the following formula (1):

[0070] (1)

[0071] Among them, i and j are constants, k is the keyword set, and v is the word vector. is a high-dimensional semantic similarity matrix.

[0072] S209 , performing low-dimensional compression processing on the high-dimensional semantic similarity matrix to obtain configuration vector information.

[0073] Specifically, the principal component analysis (PCA) dimensionality reduction technology is used to perform low-dimensional compression processing on the high-dimensional semantic similarity matrix to obtain configuration vector information, which is used to characterize server features.

[0074] S210: Generate configuration deviation information, configuration alarm level information, and repair solution information based on the configuration vector information.

[0075] In a possible implementation, the configuration vector information and the preset reference template are input into a preset cosine similarity model, and the obtained output is configuration deviation information; and configuration alarm level information is determined based on the configuration deviation information.

[0076] Specifically, the configuration vector information and the preset reference template are input into the cosine similarity model, and the output obtained is the configuration deviation information, as shown in the following formula (2):

[0077] (2)

[0078] Among them, D is the configuration deviation information, V is the configuration vector information, V base It is the default benchmark template.

[0079] The configuration alarm level information (eg, warning / critical / very critical) is determined based on the difference between the configuration deviation information and a preset deviation threshold, where the preset deviation threshold is obtained based on analysis of historical configuration deviation information.

[0080] Possibly, a confidence label is generated based on the configuration deviation information, and the confidence label is used to indicate the success rate of the server cluster repair.

[0081] In a possible implementation, the configuration vector information is compared with a preset reference template to obtain configuration difference vector information; the configuration difference vector information is matched with a preset rule base to generate repair solution information.

[0082] Specifically, according to the difference between the configuration vector information and the preset reference template, the configuration difference vector information is determined, and the configuration difference vector information is matched with the preset rule base to generate the repair solution information.

[0083] Exemplarily, the preset rule base includes but is not limited to: when the configuration difference vector information is less than or equal to -0.7, the generated repair solution information is to increase the memory parameter; when the configuration difference vector information is greater than or equal to 0.6, the generated repair solution information is to tighten the security policy.

[0084] If possible, identify the maximum offset dimension of the configuration difference vector information, and adjust kernel-level memory management parameters (for example, vm.max_map_count, which is used to control the number of memory mapping areas that a single process can occupy and is often used to optimize memory-intensive applications) based on the maximum offset dimension.

[0085] S211. Obtain performance prediction information of the server cluster.

[0086] In a possible implementation, historical performance data is extracted from stored standardized data packets; the historical performance data is input into a preset performance prediction model to obtain performance prediction information of the server cluster.

[0087] Specifically, the historical performance data is input into the preset performance prediction model, and the following formulas (3), (4), (5) and (6) are used to fit The performance is predicted, and the output is performance prediction information, including:

[0088] (3)

[0089] (4)

[0090] (5)

[0091] (6)

[0092] in, For performance prediction information, is the 1-step prediction result, is the k-step prediction result, is the sequence after the d-order difference of historical performance data, c is a constant term, and are the autoregressive and moving average coefficients, p and q are the autoregressive / moving average orders, is a white noise error term, a random disturbance at time point t.

[0093] S212: Integrate the performance prediction information, the repair solution information, the configuration deviation information, and the configuration alarm level information to generate an optimization recommendation report for the server cluster.

[0094] Where possible, the optimization suggestion report also includes a confidence level label, for example - Confidence: 92% .

[0095] For example, the optimization suggestion report is as follows:

[0096] ## Risk Audit Results

[0097] - Configuration deviation information: D=0.38 (Configuration alarm level information: Warning)

[0098] ## Dynamic Prediction

[0099] 

[0100] ## Optimization Suggestions

[0101] - Fix: `sysctl -w vm.max_map_count=262144`

[0102] - Expected effect: Reduce memory overflow risk by 30%

[0103] In a possible implementation, before generating the optimization recommendation report for the server cluster, the method further includes:

[0104] Based on the configuration vector information, the aggregation information and performance score of the server cluster are determined; based on the aggregation information, performance score, configuration deviation information and configuration alarm level information, a visual dynamic topology map is constructed; and the visual dynamic topology map is added to the optimization recommendation report.

[0105] Specifically, the aggregation information is used as the initial node position, the charge force coefficient and elastic force coefficient are defined by the configuration deviation information and the configuration alarm level information, the performance score is used to drive the topological node coloring, and based on the preset force-guided graph layout, the node position is controlled by the following formula (7):

[0106] (7)

[0107] Among them, F 电荷 is the charge force, F 弹簧 is the spring force, k e is the charge force coefficient, k s is the elastic coefficient, l is the natural length of the spring side, q i is the charge at node i, q j is the charge of node j, and r is the Euclidean distance between nodes i and j.

[0108] The natural length of the spring side is defined by the aggregate information, and the charge is defined by the performance score.

[0109] Possibly, the dynamic topology graph also includes a visual encoding of node sizes based on the number of processor cores to intuitively reflect resource scale. Edge weights in the link thickness graph database are used to represent performance bottlenecks.

[0110] Possibly, the dynamic topology graph is a WebGL Dynamic Topology Graph (WebGL).

[0111] In one possible implementation, a preset protocol adapter is used to collect physical connection topology data of a server cluster; a dependency list is determined based on the physical connection topology data and configuration vector information; based on the dependency list, graph structure data is constructed with servers as nodes and dependencies as edges; the graph structure data is input into a preset community discovery algorithm, and the output obtained is a community division result; the community division result is aggregated to obtain aggregated information of the server cluster.

[0112] Specifically, a pre-defined protocol adapter is used to collect the physical connection topology data of the server cluster. Key fields are extracted from this data and parsed for port binding information and service call chain information. This information, combined with the version compatibility features in the configuration vector, is then combined to determine a dependency list.

[0113] For example, the physical connection topology data is raw = "tcp6 0 0 :::3306 ::: LISTEN 1234 / mysqld", extract the key fields: {"port":3306, "pid":1234, "service":"mysqld"}, parse them into port binding information: PortBindings = {3306: {"pid":1234, "service":"mysqld"}}, parse the HTTP header X-Request-ID in the physical link topology data log to associate cross-node calls: Request: ServiceA →ServiceB (ID:7a3e) → ServiceC (ID:7a3e), parse them into service call chain information: ServiceA→ServiceB→ServiceC, combine the version compatibility feature in the configuration vector: for example, if ServiceB.version=2.1, ServiceC.version>=2.0 is required, determine the dependency list: EdgeList = [(ServiceA, ServiceB),(ServiceB, ServiceC), ...].

[0114] Furthermore, based on the dependency list, we construct a graph structure data with servers, services, and ports as nodes and dependencies (such as server A's process P1 calling server B's service S2) as edges, and build a hierarchical database. The formula (8) for dynamically updating edge weights is as follows:

[0115] (8)

[0116] Among them, W e is the edge weight, λ∈[0,1] is the weight adjustment factor, and the default value is 0.7.

[0117] Furthermore, the graph structure data is input into the preset community discovery algorithm to divide the logical resource groups (communities). The highly interconnected nodes are aggregated based on the principle of maximizing the modularity Q. The output is the community division result. The modularity Q is shown in the following formula (9):

[0118] (9)

[0119] Among them, m is the total edge weight, Wij comes from the graph structure data, ki is the degree of node i, It is 1 if node i and node j belong to the same community, and 0 otherwise.

[0120] The pre-set community discovery algorithm specifically executes the following steps: Nodes and weighted edges are read from graph data, with each node independently forming a community. Nodes are traversed, and the modularity (Q) change required to move them into neighboring communities is calculated. Communities with the largest modularity (Q) change are selected for merging. Traversal stops when the modularity (Q) exceeds a pre-set strong community threshold. The resulting community divisions are then output.

[0121] It is possible to periodically recalculate edge weights and reselect communities in response to system changes. An example is shown below:

[0122] MERGE (Web:Group {name:"Frontend", type:"Community"})

[0123] MERGE (ServiceA:Service {name:"ServiceA"})

[0124] SET ServiceA.community = "WebGroup"

[0125] CREATE (Web)-[:DEPENDS {weight:0.85}]->(DB:Group {name:"Database"})

[0126] Subsequently, the community division results are aggregated, and the nodes in the same community are encapsulated as super nodes. The super node attributes are equal to the weighted fusion of the node attributes in the community, and the aggregated information of the server cluster is obtained.

[0127] In one possible implementation, task queue length information is extracted from a stored standardized data packet; configuration vector information is normalized to obtain a normalized value; the normalized value and task queue length information are input into a preset performance scoring model, and the output obtained is a performance score of the server cluster.

[0128] Specifically, the preset performance scoring model is shown in the following formula (10):

[0129] (10)

[0130] Where P(t) is the performance score, C i (t) is the normalized value, w i is the weight coefficient, Q(t) is the task queue length information, Q max To preset the maximum task queue length information, and is the normalization factor.

[0131] Among them, the weight coefficient is determined by historical regression analysis, and the normalization factor is used to keep the performance score within the preset performance score range.

[0132] In a possible implementation, based on the configuration deviation information and the configuration alarm level information, a preset color space mapping model is used to construct a visual dynamic heat map; and the visual dynamic heat map is added to the optimization recommendation report.

[0133] Specifically, a preset color space mapping model is used to map the configuration deviation information and the configuration alarm level information to obtain hue, saturation (e.g., 100%), and brightness (e.g., 80%), and a dynamic heat map is constructed based on the hue, saturation, and brightness.

[0134] For example, a preset color space mapping model is used to map the configuration deviation information to obtain the hue, as shown in the following formula (11):

[0135] (11)

[0136] Among them, H is the hue and D is the configuration deviation information.

[0137] S213. Output an optimization suggestion report.

[0138] In one possible implementation, a user's configuration simulation modification instruction is obtained; the user's configuration simulation modification instruction is input into a server cluster configuration simulation model, and the output obtained is a simulation performance score; and the simulation performance score is sent to the user's terminal device.

[0139] It is possible to construct an objective function that minimizes delay and energy consumption, as shown in the following formula (12):

[0140] (12)

[0141] in, is the delay weight, and μ is the energy consumption penalty coefficient.

[0142] The performance prediction information and configuration vector information are input into the above objective function, the optimal parameter combination is solved by the particle swarm algorithm, and the parameters of the server cluster configuration simulation model are set to the optimal parameter combination.

[0143] Specifically, obtain the user's configuration simulation modification instruction (such as sysctl -w vm.max_map_count=262144), input the user's configuration simulation modification instruction into the server cluster configuration simulation model, and the output obtained is the simulation performance score, as shown in the following formula (13):

[0144] (13)

[0145] Where ΔP represents the change in performance score, and Represent the weight coefficients of the i-th and j-th parameters respectively, and are the changes of the i-th and j-th configuration parameters (normalized changes), α is the normalization factor, and i and j are constants.

[0146] The server configuration data processing method provided in the embodiment of the present application can perform targeted analysis on the original data streams of different communication protocols through a preset protocol adapter, and convert them into configuration data packets of the corresponding communication protocols, thereby realizing the effective integration of multi-source heterogeneous data and providing unified basic data for subsequent processing. The configuration information recognition model can extract unstructured text data from standardized data packets and convert it into structured configuration information. This conversion process allows unstructured data that was originally difficult to directly analyze and utilize to become queryable, statistically and associable structured configuration information, thereby realizing rapid and accurate identification of configuration data. At the same time, extracting a set of keywords from the structured configuration information can focus on the core configuration elements and lay an accurate feature foundation for subsequent analysis. By constructing a high-dimensional semantic similarity matrix, the intrinsic connection between configuration information can be deeply excavated from the semantic association level, and then the configuration vector information can be obtained through low-dimensional compression, and the complex structured information can be converted into a quantifiable and computable vector form, thereby realizing efficient dimensionality reduction and feature concentration of the configuration information. The system integrates performance prediction information, repair plan information, configuration deviation information, and configuration alarm level information to generate optimization recommendation reports, forming a complete closed loop from configuration information analysis to anomaly identification, performance prediction, and repair suggestions. This simplifies the configuration management process, improves efficiency, and ultimately helps to continuously optimize server configurations and ensure stable operation.

[0147] Furthermore, extracting target fields from the configuration data package to obtain target configuration data allows for precise screening of key information relevant to analysis needs while eliminating interference from irrelevant, redundant fields. Deduplication within the target configuration data creates a standardized configuration data table. This standardized table ensures data uniqueness and accuracy, avoiding analytical biases caused by duplicate data.

[0148] Furthermore, data in the standardized configuration data table is divided into multiple priority levels, enabling differentiated classification based on data importance. Combined with priority queues for cache storage, this ensures that high-priority data is processed and accessed first, preventing critical data from being delayed due to a backlog of lower-priority data, thereby improving the timeliness of data responses.

[0149] Furthermore, building a visual dynamic topology map transforms abstract aggregate information, performance scores, configuration deviation information, and configuration alarm level information into an intuitive graphical structure. This visual dynamic topology map clearly displays the relationships between various configuration elements of the server cluster and the mapping between performance and anomalies, making complex configuration status and potential problems clear at a glance.

[0150] A preset color space mapping model transforms abstract configuration deviation and configuration warning level information into a visual dynamic heat map, intuitively reflecting the degree of anomaly and risk level of configuration vector information through color depth. This visualization method breaks through the abstractness of numerical and textual descriptions and improves the perception of abnormal configurations.

[0151] The preset cosine similarity model calculates the cosine of the angle between the configuration vector information and the preset baseline template, accurately quantifying the similarity between the two from the perspective of vector space. The resulting configuration deviation information objectively reflects the difference between the current configuration and the baseline configuration. This avoids the problem of normal local parameters masking overall configuration anomalies, thereby improving the accuracy of server cluster configuration assessment.

[0152] Determining configuration difference vectors based on configuration vector information and a pre-set baseline template accurately captures the specific differences between the two from a vector perspective, avoiding ambiguous comparisons of the overall configuration. A pre-set rule base provides a standardized basis for remediation measures corresponding to configuration difference vectors. Regardless of the scenario, consistent remediation recommendations are generated based on unified rules.

[0153] By analyzing historical performance data through a preset performance prediction model, we can mine performance change patterns from historical data and output future performance prediction information, allowing us to adjust resource allocation or optimize configuration in advance to avoid actual performance failures.

[0154] A preset protocol adapter collects the physical connection topology data of the server cluster, combines it with configuration vector information to determine a dependency list, and associates the physical connections at the hardware level with the configuration features at the software level. A preset community discovery algorithm analyzes graph structure data and clusters servers with close dependencies into the same community. The community division results can quickly locate functional units in the system, avoiding the individual management of dispersed nodes and improving the efficiency of batch management of similar server clusters. The aggregated information obtained by aggregating the community division results can be used to perform unified operations on servers within the same community, avoiding repeated operations on individual nodes and significantly reducing operational costs and time.

[0155] The pre-set performance scoring model simultaneously inputs normalized configuration vector information and task queue length information, achieving a multi-dimensional fusion of "static configuration" and "dynamic load." This fusion more comprehensively depicts the actual operating status of the server, making the output performance score more accurate to real-world scenarios.

[0156] Users can manipulate the server cluster configuration simulation model using configuration simulation commands, generating simulated performance scores without having to directly modify the actual server configuration. The simulated performance scores intuitively reflect the performance trends associated with the configuration modification commands, allowing users to predict the impact of the changes before actually executing them.

[0157] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method.

[0158] Figure 3 This is a schematic diagram of the structure of the server configuration data processing device provided in the embodiment of the present application. Figure 3 As shown, an embodiment of the present application further provides a server configuration data processing device 30, comprising:

[0159] An acquisition module 301 is used to acquire the original data stream of the server cluster;

[0160] The processing module 302 is configured to parse the original data stream through a preset protocol adapter to obtain a configuration data packet corresponding to each protocol, wherein the configuration data packet includes unstructured text data;

[0161] The processing module 302 is further used to clean the configuration data packet to obtain a standardized configuration data table;

[0162] The storage module 303 is used to cache and store the standardized configuration data table to obtain a stored standardized data packet;

[0163] an identification module 304 for inputting the standardized data packet into the configuration information identification model so that the configuration information identification model extracts unstructured text data from the standardized data packet and converts the unstructured text data into structured configuration information;

[0164] The output module 305 is used to output structured configuration information.

[0165] In a possible implementation, the processing module 302 is specifically configured to:

[0166] Extract the target field in the configuration data packet to obtain target configuration data;

[0167] Remove duplicate data from the target configuration data to obtain a standardized configuration data table.

[0168] In a possible implementation, the storage module 303 is specifically configured to:

[0169] Divide the data in the standardized configuration data table into n priority levels, where n is an integer greater than 1;

[0170] According to the priority level, the data in the standardized configuration data table is cached and stored in the priority queue to obtain a stored standardized data packet.

[0171] In a possible implementation, the processing module 302 is further configured to:

[0172] Extracting a set of keywords from structured configuration information;

[0173] Based on the keyword set, a high-dimensional semantic similarity matrix is ​​constructed;

[0174] The high-dimensional semantic similarity matrix is ​​compressed into a low-dimensional form to obtain configuration vector information;

[0175] Generate configuration deviation information, configuration alarm level information, and repair solution information based on configuration vector information;

[0176] Obtain performance prediction information of server clusters;

[0177] Integrate performance prediction information, repair plan information, configuration deviation information, and configuration alarm level information to generate optimization recommendation reports for server clusters;

[0178] Output optimization recommendation report.

[0179] In a possible implementation, the processing module 302 is further configured to:

[0180] Determine the aggregate information and performance score of the server cluster based on the configuration vector information;

[0181] Build a visual dynamic topology map based on aggregated information, performance scores, configuration deviation information, and configuration alarm level information;

[0182] Add a visual dynamic topology map to the optimization recommendation report.

[0183] In a possible implementation, the processing module 302 is further configured to:

[0184] Based on configuration deviation information and configuration alarm level information, a preset color space mapping model is used to build a visual dynamic heat map;

[0185] Add visual dynamic heat map to optimization suggestion report.

[0186] In a possible implementation, the processing module 302 is further configured to:

[0187] Input the configuration vector information and the preset reference template into the preset cosine similarity model, and the output obtained is the configuration deviation information;

[0188] Based on the configuration deviation information, configuration alarm level information is determined.

[0189] In a possible implementation, the processing module 302 is further configured to:

[0190] Comparing the configuration vector information with a preset reference template to obtain configuration difference vector information;

[0191] The configuration difference vector information is matched with the preset rule base to generate repair solution information.

[0192] In a possible implementation, the processing module 302 is further configured to:

[0193] Extract historical performance data from stored standardized data packages;

[0194] Input historical performance data into the preset performance prediction model to obtain performance prediction information of the server cluster.

[0195] In a possible implementation, the processing module 302 is further configured to:

[0196] Use preset protocol adapters to collect physical connection topology data of the server cluster;

[0197] Determine a dependency list based on the physical connection topology data and configuration vector information;

[0198] Based on the dependency list, a graph structure data is constructed with servers as nodes and dependencies as edges;

[0199] Input the graph structure data into the preset community discovery algorithm, and the output is the community division result;

[0200] The community division results are aggregated to obtain the aggregated information of the server cluster.

[0201] In a possible implementation, the processing module 302 is further configured to:

[0202] Extracting task queue length information from the stored standardized data packets;

[0203] Normalizing the configuration vector information to obtain a normalized value;

[0204] In a possible implementation, the processing module 302 is further configured to:

[0205] Get the user's configuration simulation modification instructions;

[0206] Input the user's configuration simulation modification instructions into the server cluster configuration simulation model, and the output obtained is the simulation performance score;

[0207] The simulated performance score is sent to the user's terminal device.

[0208] For the description of the features in the embodiment corresponding to the server configuration data processing device, please refer to the relevant description of the embodiment corresponding to the server configuration data processing method, which will not be repeated here.

[0209] Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present application. Figure 4 As shown, the electronic device 40 provided in this embodiment includes: at least one processor 401 and a memory 402. Optionally, the electronic device 40 further includes a communication component 403. The processor 401, the memory 402 and the communication component 403 are connected via a bus.

[0210] In a specific implementation process, at least one processor 401 executes the computer-executable instructions stored in the memory 402 , so that the at least one processor 401 executes the above-mentioned embodiment of the server configuration data processing method.

[0211] The specific implementation process of the processor 401 can be found in the above method embodiment. Its implementation principle and technical effects are similar and will not be repeated here in this embodiment.

[0212] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the application may be directly executed by a hardware processor or by a combination of hardware and software modules within the processor.

[0213] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage.

[0214] A bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be categorized as address buses, data buses, and control buses. For ease of illustration, the buses in the drawings of this application are not limited to just one bus or just one type of bus.

[0215] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any one of the above-mentioned XX method embodiments when run.

[0216] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disk.

[0217] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps of any of the above-mentioned server configuration data processing method embodiments are implemented.

[0218] An embodiment of the present application also provides another computer program product, including a non-volatile computer-readable storage medium, the non-volatile computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, implementing the steps of any of the above-mentioned server configuration data processing method embodiments.

[0219] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0220] The above is a detailed introduction to a server configuration data processing method, electronic device, storage medium and program product provided by the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core ideas of the present application. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the claims of the present application.

Claims

1. A server configuration data processing method, characterized in that: include: Get the original data stream of the server cluster; Parsing the original data stream through a preset protocol adapter to obtain a configuration data packet corresponding to each protocol, wherein the configuration data packet includes unstructured text data; Cleaning the configuration data packet to obtain a standardized configuration data table; caching and storing the standardized configuration data table to obtain a stored standardized data packet; Inputting the standardized data packet into a pre-trained configuration information recognition model, so that the pre-trained configuration information recognition model extracts unstructured text data from the standardized data packet and converts the unstructured text data into structured configuration information; Outputting the structured configuration information of the server cluster; Extracting a keyword set from the structured configuration information; Based on the keyword set, construct a high-dimensional semantic similarity matrix; Performing low-dimensional compression processing on the high-dimensional semantic similarity matrix to obtain configuration vector information; Based on the configuration vector information, generate configuration deviation information, configuration alarm level information, and repair solution information; Obtaining performance prediction information of the server cluster; Integrating the performance prediction information, the repair solution information, the configuration deviation information, and the configuration alarm level information to generate an optimization recommendation report for the server cluster; Output the optimization suggestion report.

2. The method according to claim 1, characterized in that The cleaning process of the configuration data packet to obtain a standardized configuration data table includes: Extracting the target field in the configuration data packet to obtain target configuration data; Duplicate data in the target configuration data is removed to obtain a standardized configuration data table.

3. The method according to claim 1, characterized in that The step of caching and storing the standardized configuration data table to obtain a stored standardized data packet includes: dividing the data in the standardized configuration data table into a plurality of priority levels; According to the priority level, the data in the standardized configuration data table is cached and stored in a priority queue to obtain a stored standardized data packet.

4. The method according to claim 1, wherein Before generating the optimization recommendation report for the server cluster, the method further includes: Determining aggregate information and a performance score of the server cluster based on the configuration vector information; Constructing a visual dynamic topology map based on the aggregated information, the performance score, the configuration deviation information, and the configuration alarm level information; The visual dynamic topology map is added to the optimization suggestion report.

5. The method according to claim 1, wherein Before generating the optimization recommendation report for the server cluster, the method further includes: Based on the configuration deviation information and the configuration alarm level information, a preset color space mapping model is used to construct a visual dynamic heat map; The visual dynamic heat map is added to the optimization suggestion report.

6. The method according to claim 1, wherein Configuration deviation information and configuration alarm level information are generated based on the configuration vector information, including: Inputting the configuration vector information and the preset reference template into a preset cosine similarity model, the output obtained is configuration deviation information; Based on the configuration deviation information, configuration alarm level information is determined.

7. The method according to claim 1, characterized in that The repair solution information is generated based on the configuration vector information, including: Comparing the configuration vector information with a preset reference template to obtain configuration difference vector information; The configuration difference vector information is matched with a preset rule base to generate repair solution information.

8. The method according to claim 1, characterized in that The obtaining of the performance prediction information of the server cluster includes: extracting historical performance data from the stored standardized data packets; The historical performance data is input into a preset performance prediction model to obtain performance prediction information of the server cluster.

9. The method according to claim 4, characterized in that The determining, based on the configuration vector information, aggregation information of the server cluster includes: Use preset protocol adapters to collect physical connection topology data of the server cluster; Determining a dependency list based on the physical connection topology data and the configuration vector information; Based on the dependency list, construct graph structure data with servers as nodes and dependency relationships as edges; Inputting the graph structure data into a preset community discovery algorithm, the output obtained is the community division result; Aggregation processing is performed on the community division results to obtain aggregated information of the server cluster.

10. The method according to claim 4, characterized in that Determining the performance score of the server cluster based on the configuration vector information includes: Extracting task queue length information from the stored standardized data packet; Normalizing the configuration vector information to obtain a normalized value; The normalized value and the task queue length information are input into a preset performance scoring model, and the obtained output is the performance score of the server cluster.

11. The method according to claim 1, characterized in that Also includes: Get the user's configuration simulation modification instructions; Inputting the user's configuration simulation modification instruction into the server cluster configuration simulation model, and obtaining the output as a simulation performance score; The simulation performance score is sent to the user's terminal device.

12. An electronic device, characterized in that: include: memory for storing computer programs; A processor, configured to implement the steps of the server configuration data processing method according to any one of claims 1 to 11 when executing the computer program.

13. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps of the server configuration data processing method according to any one of claims 1 to 11 are implemented.

14. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the server configuration data processing method according to any one of claims 1 to 11 are implemented.

Images