Server transaction elastic demand calculation method and application system

Abstract

The invention relates to the field of transaction management, in particular to an elastic demand calculation method for server transactions and an application system. The method comprises the following steps: acquiring a client protocol and server operation data; analyzing the server operation data, and determining operation resource information; analyzing the operation resource information, and determining a resource consumption trend and a resource distribution condition; determining a customer service type according to the customer protocol; determining a concurrent processing demand according to the customer service type and the resource allocation condition; and according to the resource consumption trend and the concurrent processing demand, generating an elastic demand parameter, and according to the elastic demand parameter, generating a resource configuration suggestion. According to the customer service type and the resource allocation condition, the concurrent processing requirement is determined, and insufficient capacity caused by capacity expansion delay is avoided. According to the resource consumption trend and the concurrent processing demand, the elastic demand parameter is generated, and the resource configuration suggestion is generated according to the elastic demand parameter, so that the realization of accurate triggering expansion, cost balance, stability and response speed is facilitated.

Description

Technical Field

[0001] This application relates to the field of transaction management, and in particular, to a method for calculating elastic demand for server transactions and an application system. Background Art

[0002] When configuring server resources, due to the significant differences in resource demand patterns for different services, targeted arrangements are required; otherwise, customer needs cannot be met.

[0003] Existing methods usually allocate resources based on the contracts signed by customers or historical resource usage data. This approach relies on fixed capacity planning or simple periodic forecasting, which is likely to result in delays or over-responses in elastic scaling. Summary of the Invention

[0004] This application provides a method for calculating elastic demand for server transactions and an application system to solve the above problems.

[0005] In a first aspect, this application provides a method for calculating elastic demand for server transactions, the method comprising: Obtain a customer agreement and server operation data; analyze the server operation data to determine operation resource information; Analyze the operation resource information to determine resource consumption trends and resource allocation situations; Determine the customer service type according to the customer agreement; Determine the concurrent processing requirements according to the customer service type and the resource allocation situation; Generate elastic demand parameters according to the resource consumption trends and the concurrent processing requirements, and generate a resource configuration recommendation according to the elastic demand parameters.

[0006] Through this solution, obtaining a customer agreement and server operation data can avoid violating data protection regulations such as GDPR. Analyzing the server operation data to determine operation resource information provides compliant and high-signal-to-noise data input for resource trend analysis. Analyzing the operation resource information to determine resource consumption trends and resource allocation situations helps to eliminate the problem of insufficient ability to identify complex demand patterns and reduce the noise misjudgment rate. Determining the customer service type according to the customer agreement can avoid misclassification based on a single indicator and improve the accuracy of scenario matching. Determining the concurrent processing requirements according to the customer service type and the resource allocation situation can avoid capacity shortages caused by expansion delays. Generating elastic demand parameters according to the resource consumption trends and the concurrent processing requirements, and generating a resource configuration recommendation according to the elastic demand parameters helps to achieve precise trigger for expansion and balance cost, stability, and response speed.

[0007] Optionally, the obtaining server operation data includes: Parse the customer agreement to determine the authorized whitelist, desensitization requirements, and service scenario identifiers; Determine the data types for which running data can be obtained according to the authorized whitelist; Perform compliance processing on the existing data according to the data types, the desensitization requirements, and the service scenario identifiers, and use the dataset after compliance processing as the server running data.

[0008] Through this solution, parse the customer agreement to determine the authorized whitelist, desensitization requirements, and service scenario identifiers, avoid the risk of privacy leakage caused by over-range collection, eliminate the noise interference introduced by the direct exposure of sensitive fields in the original data, ensure that the business feature extraction rules are strictly matched with the actual business types of customers, and eliminate the problem of misconfigured scaling strategies caused by the decoupling of business features and resource requirements. According to the authorized whitelist, determine the data types for which running data can be obtained, and avoid feature loss or redundant data accumulation caused by ambiguous data collection scopes. Perform compliance processing on the existing data according to the data types, the desensitization requirements, and the service scenario identifiers, and use the dataset after compliance processing as the server running data, which helps to eliminate the periodic feature extraction deviation caused by misaligned data collection timings.

[0009] Optionally, the determining the customer business type according to the customer agreement includes: Obtain the scenario type code based on the service scenario identifier; Analyze the server running data to determine the API call sequence during the running process; Determine the business operation mode features according to the API call sequence; Match the business operation mode features with the scenario type code; Determine the customer business type according to the matching degree between the scenario type code and the business operation mode features.

[0010] Through this solution, obtain the scenario type code based on the service scenario identifier, eliminate the ambiguity of the business type description in the protocol text, and provide a unified input for business feature matching. Analyze the server running data to determine the API call sequence during the running process, which helps to quantify the real trajectory of business operations and capture the ignored high-frequency continuous call combinations. Determine the business operation mode features according to the API call sequence, which helps to eliminate the problem of distorted feature extraction. Determine the customer business type according to the matching degree between the scenario type code and the business operation mode features, which helps to achieve the goal of a multi-dimensional decision matrix.

[0011] Optionally, the performing compliance processing on the existing data according to the data types, the desensitization requirements, and the service scenario identifiers includes: Determine the priority sorting rule of the authorized whitelist according to the service scenario identifier; Analyze the desensitization requirements to determine the sensitive field mapping table; Establish a field replacement strategy according to the sensitive field mapping table; Filter the existing data according to the data type to obtain available data; Retain the authorized fields of the available data according to the priority sorting rule, and apply the field replacement strategy to process the sensitive data in the available data to complete the compliance processing.

[0012] Through this solution, determining the priority sorting rule of the authorized whitelist according to the service scenario identifier helps to eliminate the problem of feature extraction distortion. Analyzing the desensitization requirements and determining the sensitive field mapping table helps to eliminate the risks of privacy leakage and data noise interference. Establishing a field replacement strategy according to the sensitive field mapping table achieves a balance between compliance processing and business feature retention. Filtering the existing data according to the data type to obtain available data solves the problems of data redundancy and compliance conflicts. Retaining the authorized fields of the available data according to the priority sorting rule and applying the field replacement strategy to process the sensitive data in the available data to complete the compliance processing achieves the dual goals of privacy protection and business analysis, eliminates the risk of privacy leakage, and improves data consistency through standardized processing.

[0013] Optionally, the generating of the elastic demand parameter according to the resource consumption trend and the concurrent processing requirement includes: Decompose the resource consumption trend based on the timestamp to determine the resource consumption period; Analyze the concurrent processing requirement to determine the resource usage pattern; Generate the elastic demand parameter according to the resource consumption period and the resource usage pattern.

[0014] Through this solution, decomposing the resource consumption trend based on the timestamp to determine the resource consumption period eliminates the baseline drift caused by time zone differences or inconsistent sampling intervals, ensures the global comparability of time series analysis, and provides a standardized input for elastic computing. Analyzing the concurrent processing requirement to determine the resource usage pattern avoids the defect of not establishing a dynamic association mechanism between business types and resource usage patterns, and quantifies the contribution of different scenarios to the elastic demand. Generating the elastic demand parameter according to the resource consumption period and the resource usage pattern helps to eliminate the lag problem and improve the prediction accuracy.

[0015] Optionally, the generating of the resource configuration suggestion according to the elastic demand parameter includes: Analyze the elastic demand parameter to determine the scaling trigger point and the scaling mode; Generate the resource configuration suggestion according to the scaling trigger point and the scaling mode.

[0016] Through this solution, analyzing the elastic demand parameters, determining the expansion trigger point and expansion mode helps to eliminate the problem of missing emergency detection in the rough expansion trigger point, achieve the goal of business-resource coupling analysis, and avoid misjudgment of business characteristics. According to the expansion trigger point and expansion mode, resource configuration suggestions are generated. Optionally, the analyzing the elastic demand parameters and determining the expansion mode includes: Based on the customer agreement, obtain SLA default cost data; analyze the SLA default cost data to quantify the service interruption tolerance. Obtain real-time resource information; analyze the real-time resource information to determine the availability index. Analyze the availability index and calculate the resource supply elasticity coefficient. Analyze the historical bill payment cycle to determine the cost sensitivity. Determine the expansion mode according to the service interruption tolerance, the resource supply elasticity coefficient and the cost sensitivity.

[0017] Through this solution, based on the customer agreement, obtaining SLA default cost data helps to eliminate the problem of missing decision-making basis caused by ignoring the quantification of SLA default costs. Analyzing the SLA default cost data to quantify the service interruption tolerance eliminates the problem of policy wavering caused by the ambiguity of multi-clause weights. Obtaining real-time resource information overcomes the blind area of expansion decision-making caused by the lack of resource supply elasticity coefficient. Analyzing the real-time resource information to determine the availability index helps to eliminate the decision-making delay caused by the unstructured evaluation of resource status. Analyzing the availability index and calculating the resource supply elasticity coefficient compresses the multi-dimensional resource status into a single decision-making parameter, eliminating the policy conflicts caused by the isolated analysis of multi-dimensional indicators. Analyzing the historical bill payment cycle to determine the cost sensitivity helps to overcome the problem of inefficient resource reuse strategies. Determining the expansion mode according to the service interruption tolerance, the resource supply elasticity coefficient and the cost sensitivity helps to achieve the design goal of the multi-dimensional decision matrix.

[0018] Optionally, the generating resource configuration suggestions according to the expansion trigger point and the expansion mode includes: Obtain the historical expansion records, and based on the expansion trigger point, analyze the historical expansion records to determine the expansion response delay time. Analyze the resource consumption trend to determine the resource exhaustion countdown. Generate a preventive expansion time window according to the time difference between the expansion response delay time and the resource exhaustion countdown. Generate resource configuration suggestions including a time window identifier according to the preventive expansion time window and the expansion mode.

[0019] Through this solution, historical expansion records are obtained. Based on the expansion trigger points, the historical expansion records are analyzed to determine the expansion response delay time, covering the delay fluctuations in extreme scenarios and ensuring the robustness of the time window calculation. The resource consumption trend is analyzed to determine the resource exhaustion countdown, avoiding one-sided decisions caused by a single metric. According to the time difference between the expansion response delay time and the resource exhaustion countdown, a preventive expansion time window is generated, which helps to eliminate the problem of a sharp increase in emergency expansion costs caused by not reserving the delay time. According to the preventive expansion time window and the expansion mode, a resource allocation recommendation including a time window identifier is generated to ensure the global consistency of the time window strategy with the cost sensitivity and service stability requirements.

[0020] Optionally, generating elastic demand parameters according to the resource consumption cycle and the resource usage pattern includes: Based on the resource usage pattern, analyze the server operation data to determine the cause of mode switching; Analyze the cause of mode switching to determine the cause attribute; According to the cause attribute, analyze the resource consumption cycle to determine the influence weight of each cause of mode switching on the resource usage pattern; Generate elastic demand parameters according to the influence weight.

[0021] Through this solution, based on the resource usage pattern, analyze the server operation data to determine the cause of mode switching, eliminating the problem of prediction distortion caused by not distinguishing business operation characteristics from noise events. Analyze the cause of mode switching to determine the cause attribute, which helps to eliminate the problem of missing the detection window of sudden events due to not analyzing the residual term and improve the prediction accuracy of the expansion trigger point. According to the cause attribute, analyze the resource consumption cycle to determine the influence weight of each cause of mode switching on the resource usage pattern, avoiding the imbalance of resource allocation caused by one-dimensional decision-making. Generate elastic demand parameters according to the influence weight, which helps to support the generation of a multi-dimensional decision matrix and achieve the balance of cost, stability and response speed.

[0022] In a second aspect, the present application provides an elastic demand calculation application system for server transactions, and the system includes: A data analysis module, configured to obtain a customer agreement and server operation data; analyze the server operation data to determine operation resource information; An information analysis module, configured to analyze the operation resource information to determine the resource consumption trend and the resource allocation situation; A type determination module, configured to determine the customer business type according to the customer agreement; A demand determination module, configured to determine the concurrent processing demand according to the customer business type and the resource allocation situation; A recommendation generation module, configured to generate elastic demand parameters according to the resource consumption trend and the concurrent processing requirement, and generate a resource configuration recommendation according to the elastic demand parameters.

[0023] Optionally, the elastic demand calculation application system for server transactions further includes a data determination module, configured to: parse the client protocol to determine an authorized whitelist, a desensitization requirement, and a service scenario identifier; determine the data types for which operational data can be obtained according to the authorized whitelist; perform compliance processing on existing data according to the data types, the desensitization requirement, and the service scenario identifier, and use the dataset after the compliance processing as the server operational data.

[0024] Optionally, when determining the customer business type according to the client protocol, the type determination module is configured to: obtain a scenario type code based on the service scenario identifier; analyze the server operational data to determine the API call sequence during operation; determine the business operation mode characteristics according to the API call sequence; match the business operation mode characteristics with the scenario type code; and determine the customer business type according to the matching degree between the scenario type code and the business operation mode characteristics.

[0025] Optionally, when performing compliance processing on existing data according to the data types, the desensitization requirement, and the service scenario identifier, the data determination module is configured to: determine the priority sorting rule of the authorized whitelist according to the service scenario identifier; analyze the desensitization requirement to determine a sensitive field mapping table; establish a field replacement policy according to the sensitive field mapping table; filter the existing data according to the data types to obtain available data; retain the authorized fields of the available data according to the priority sorting rule, and process the sensitive data in the available data using the field replacement policy to complete the compliance processing.

[0026] Optionally, when generating elastic demand parameters according to the resource consumption trend and the concurrent processing requirement, the recommendation generation module is configured to: decompose the resource consumption trend based on a timestamp to determine a resource consumption period; parse the concurrent processing requirement to determine a resource usage pattern; and generate elastic demand parameters according to the resource consumption period and the resource usage pattern.

[0027] Optionally, when generating a resource configuration recommendation according to the elastic demand parameters, the recommendation generation module is configured to: analyze the elastic demand parameters to determine an expansion trigger point and an expansion mode; and generate a resource configuration recommendation according to the expansion trigger point and the expansion mode.

[0028] Optionally, the elastic demand calculation application system for server transactions further includes a mode determination module, which is configured to: obtain SLA default cost data based on the customer agreement; analyze the SLA default cost data to quantify the service interruption tolerance; obtain real-time resource information; analyze the real-time resource information to determine the availability metric; analyze the availability metric to calculate the resource supply elasticity coefficient; analyze the historical bill payment cycle to determine the cost sensitivity; and determine the scaling mode according to the service interruption tolerance, the resource supply elasticity coefficient, and the cost sensitivity.

[0029] Optionally, when the recommendation generation module generates a resource configuration recommendation according to the scaling trigger point and the scaling mode, it is configured to: obtain the historical scaling records, analyze the historical scaling records based on the scaling trigger point to determine the scaling response delay time; analyze the resource consumption trend to determine the resource exhaustion countdown; generate a preventive scaling time window according to the time difference between the scaling response delay time and the resource exhaustion countdown; and generate a resource configuration recommendation including a time window identifier according to the preventive scaling time window and the scaling mode.

[0030] Optionally, when the recommendation generation module generates elastic demand parameters according to the resource consumption cycle and the resource usage pattern, it is configured to: analyze the server operation data based on the resource usage pattern to determine the mode switching incentive; analyze the mode switching incentive to determine the incentive attribute; analyze the resource consumption cycle according to the incentive attribute to determine the influence weight of each mode switching incentive on the resource usage pattern; and generate elastic demand parameters according to the influence weight. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0032] Figure 1 FIG. 1 is a schematic diagram of an application scenario provided by an embodiment of the present application; Figure 2 FIG. 2 is a flowchart of a method for calculating elastic demand for server transactions provided by an embodiment of the present application; Figure 3 FIG. 3 is a schematic structural diagram of an elastic demand calculation application system for server transactions provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0034] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0035] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0036] Existing methods usually allocate resources based on contracts signed by customers or historical resource usage data. This approach relies on fixed capacity planning or simple periodic forecasts, which is likely to lead to high resource idleness, delayed elastic expansion or over-response. For example, financial trading systems still maintain high-specification server clusters on non-trading days, resulting in energy and cost waste.

[0037] Based on this, the present application provides a method and application system for calculating elastic demand for server transactions to obtain customer agreements and server operation data to avoid violating data protection regulations such as GDPR. Analyze server operation data, determine operation resource information, and provide compliant and high signal-to-noise ratio data input for resource trend analysis. Analyze operation resource information, determine resource consumption trends and resource allocation, which helps to eliminate the problem of insufficient recognition of composite demand patterns and reduce the noise misjudgment rate. According to the customer agreement, determine the customer's business type, avoid misclassification based on a single indicator, and improve the accuracy of scene matching. According to the customer's business type and resource allocation, determine the concurrent processing requirements to avoid insufficient capacity due to expansion delays. Generate elastic demand parameters based on resource consumption trends and concurrent processing requirements, and generate resource configuration suggestions based on the elastic demand parameters, which helps to accurately trigger expansion and balance cost, stability and response speed.

[0038] Figure 1 A schematic diagram of an application scenario provided by the present application. When configuring server resources, the method provided by the present application is applied.

[0039] Specifically, the method provided in this application is applied to any server. The server interacts with the client device, obtains and analyzes the server operation data through the client device, and determines the operation resource information. Analyze the operation resource information, determine the resource consumption trend and resource allocation situation, retrieve the customer agreement from the internal database, and determine the customer business type according to the customer agreement, avoiding misclassification based on a single indicator and improving the scene matching accuracy. According to the customer business type and resource allocation situation, determine the concurrent processing requirements, avoiding capacity shortage caused by expansion delay. Generate elastic demand parameters according to the resource consumption trend and concurrent processing requirements, and generate resource configuration suggestions according to the elastic demand parameters, which helps to achieve accurate triggering of expansion and balance costs, stability and response speed. The specific implementation method can refer to the following embodiments.

[0040] Figure 2 FIG. is a flowchart of a method for calculating elastic demand for server transactions provided in an embodiment of this application. The method of this embodiment can be applied to the server in the above scenario. As Figure 2 shown, the method includes: S201. Obtain the customer agreement and server operation data; analyze the server operation data to determine the operation resource information; The customer agreement may be a contract document signed by the customer and the cloud service provider.

[0041] The server operation data may be the original logs and monitoring metrics generated during the server operation.

[0042] The operation resource information may be structured data of the resource usage status.

[0043] Specifically, automatically pull the customer agreement stored in the internal server and signed by the customer during service subscription through the cloud platform API; at the same time, capture the server operation data of the user device in real time. Parse the scene type identifier in the customer agreement and extract the sensitive field mapping table; then, based on the server operation data, perform dynamic desensitization according to the authorization whitelist priority rule to determine the operation resource information: First, perform encrypted hashing on the fields marked as personal identity information in the sensitive field mapping table; Second, retain the original values of the business feature fields authorized by the whitelist; Finally, generate compliant operation resource information including desensitized structured data and metadata tags.

[0044] S202. Analyze the operation resource information to determine the resource consumption trend and resource allocation situation; The resource consumption trend may be the specific manifestation of how the resource is consumed in the usage process.

[0045] The resource allocation situation may be the topological state description of the current resource pool.

[0046] Specifically, for the running resource information, STL algorithms are adopted, and the trend term, periodic term, and residual term are separated through iterative weighted least squares; for the residual term, the moving standard deviation curve is calculated, and the abnormal interval is marked as the candidate interval for burst traffic; then, the business event feature codes defined in the customer agreement are extracted to generate the business event density curve; the business event density curve and the residual term of resource consumption are aligned through dynamic time warping, and the Pearson correlation coefficient is calculated; thus, the resource consumption trend and resource allocation situation are output.

[0047] S203. Determine the customer business type according to the customer agreement; The customer business type can be a classification label jointly determined according to the business scenario identifier in the customer agreement and the API call characteristics in the server log.

[0048] Specifically, according to the customer agreement, the identifier keywords are extracted using the regular expressions predefined in the protocol template; then, the synonyms of the unstandardized expressions are converted; thus, the business scenario identifiers with confidence scores are extracted; then, the sliding window size is set as the SLA metric collection period defined in the customer agreement, and the call frequency and resource consumption ratio of each API endpoint within the window are counted to generate the heat map of operation types; the heat map of operation types is normalized to construct the API call characteristics defined in the dimension; subsequently, a feature fusion matrix is constructed, and when the matching degree between the business scenario identifier and the API call characteristics is too high, the customer business type is determined.

[0049] S204. Determine the concurrent processing requirements according to the customer business type and resource allocation situation; The concurrent processing requirements can be the target resource specifications calculated based on the business type and resource allocation situation.

[0050] Specifically, based on the resource allocation situation, sequence pattern mining is performed on the desensitized API logs to identify high-frequency operation chains; then, a business type-model mapping rule library is established according to the identifiers of the customer business type; finally, based on the linear superposition model, the influence values of the characteristics of unexpected events are fused to generate the final concurrent processing requirement parameters.

[0051] S205. Generate elastic demand parameters according to the resource consumption trend and concurrent processing requirements, and generate resource configuration suggestions according to the elastic demand parameters.

[0052] The elastic demand parameters can be multi-dimensional quantitative indicators used to generate the scaling strategy.

[0053] The resource configuration suggestions can be scaling decision instructions.

[0054] Specifically, calculate the burst traffic risk level based on the residual moving standard deviation in the resource consumption trend and the concurrent processing requirements; then, generate elastic demand parameters in combination with the SLA default cost in the customer business type; subsequently, based on the elastic calculation theory in the dynamic resource scheduling model, monitor the resource consumption fluctuations using the moving standard deviation in time series analysis, and generate a resource supply capacity report in combination with the resource reservation strategy in the capacity planning algorithm; furthermore, construct an expansion decision matrix based on the elastic demand parameters, the resource supply capacity report, and the customer business type; finally, output resource allocation suggestions according to the matrix weights.

[0055] Through this solution, obtain the customer agreement and server operation data to avoid violating data protection regulations such as GDPR. Analyze the server operation data to determine the operation resource information, providing compliant and high-signal-to-noise data input for resource trend analysis. Analyze the operation resource information to determine the resource consumption trend and resource allocation situation, which helps to eliminate the problem of insufficient ability to identify complex demand patterns and reduce the noise misjudgment rate at the same time. Determine the customer business type according to the customer agreement to avoid misclassification based on a single indicator and improve the scenario matching accuracy. Determine the concurrent processing requirements according to the customer business type and resource allocation situation to avoid capacity shortage caused by expansion delay. Generate elastic demand parameters based on the resource consumption trend and concurrent processing requirements, and generate resource allocation suggestions according to the elastic demand parameters, which helps to achieve precise trigger for expansion and balance costs, stability, and response speed.

[0056] In some embodiments, parse the customer agreement to determine the authorized whitelist, desensitization requirements, and service scenario identifier; according to the authorized whitelist, determine the data types of the operation data that can be obtained; according to the data types, desensitization requirements, and service scenario identifier, perform compliance processing on the existing data, and use the compliant processed data set as the server operation data.

[0057] The authorized whitelist can be a list of server operation data fields allowed to be collected. The desensitization requirements can be specified data privacy protection rules. The service scenario identifier can be a unique code identifying the customer business type. The operation data that can be obtained can be a subset of the server raw log data filtered according to the authorized whitelist. The data type can be a specific data field category. The existing data can be the initial data set in the server raw log that has not been processed. The data set can be a structured data set.

[0058] Specifically, based on the customer agreement, the authorization field markers in the predefined clause template defined by the mapping relationship between the customer business scenario and the compliance clause are matched using regular expressions. The authorization whitelist, desensitization requirements, and service scenario identifiers are extracted according to the authorization field markers. Furthermore, the authorization whitelist is compared with the field names of the server's original logs in full volume to filter out the data types allowed to be collected in the available operation data. Then, based on the data types, the predefined desensitization rule mapping table is loaded according to the service scenario identifiers. Subsequently, the desensitization operation is performed field by field according to the desensitization requirements. Furthermore, it is checked whether the desensitized data retains the business feature fields corresponding to the service scenario identifiers. Finally, the desensitized data is bound to the service scenario identifiers to generate the server operation data.

[0059] Through this solution, the customer agreement is parsed to determine the authorization whitelist, desensitization requirements, and service scenario identifiers, avoiding the risk of privacy leakage caused by out-of-scope collection, eliminating the noise interference introduced by the direct exposure of sensitive fields in the original data, ensuring that the business feature extraction rules are strictly matched with the actual business types of customers, and eliminating the problem of mismatched scaling strategies caused by the decoupling of business features and resource requirements. According to the authorization whitelist, the data types of the available operation data are determined, avoiding the loss of features or the accumulation of redundant data caused by ambiguous data collection scopes. According to the data types, desensitization requirements, and service scenario identifiers, the existing data is processed in a compliant manner, and the dataset after compliant processing is used as the server operation data, which helps to eliminate the periodic feature extraction deviation caused by the misalignment of data collection time series.

[0060] In some embodiments, based on the service scenario identifier, the scenario type code is obtained; the server operation data is analyzed to determine the API call sequence during the operation; according to the API call sequence, the business operation mode features are determined; the business operation mode features are matched with the scenario type code; according to the matching degree between the scenario type code and the business operation mode features, the customer business type is determined.

[0061] The scenario type code can be a predefined standardized business scenario label. The API call sequence can be a set of API endpoint call records. The business operation mode features can be a quantitative feature vector representing the resource demand mode of a specific business scenario. The matching degree can be the degree of similarity between the actual business operation mode features and the corresponding standard features of the scenario type code.

[0062] Specifically, a preset scenario type coding table is constructed by abstractly classifying the common operation modes of business scenarios according to the domain ontology, which converts the service scenario identifier into the corresponding scenario type code. Extract the original logs from the server operation data, and generate an API call sequence sorted by the timestamp. Generate a feature vector according to the business feature template associated with the scenario type code; then, calculate the feature matching degree of the API call sequence; thereby extract the business operation mode features. The cosine similarity algorithm is used to compare the business operation mode feature vector with the standard feature vector corresponding to the scenario type code. Finally, calculate the matching degree weight of the scenario type code and the business operation mode feature, and execute a hierarchical decision to determine the customer business type.

[0063] Through this solution, based on the service scenario identifier, obtain the scenario type code, eliminate the ambiguity of the business type description in the protocol text, and provide a unified input for business feature matching. Analyze the server operation data to determine the API call sequence during the operation process, which helps to quantify the real track of business operations and capture the ignored high-frequency continuous call combinations. Determine the business operation mode features according to the API call sequence, which helps to eliminate the problem of distorted feature extraction. Determine the customer business type according to the matching degree between the scenario type code and the business operation mode features, which helps to achieve the goal of a multi-dimensional decision matrix.

[0064] In some embodiments, according to the service scenario identifier, determine the priority sorting rule of the authorization whitelist; analyze the desensitization requirements to determine the sensitive field mapping table; establish a field replacement strategy according to the sensitive field mapping table; screen the existing data according to the data type to obtain the available data; retain the authorized fields of the available data according to the priority sorting rule, and apply the field replacement strategy to process the sensitive data in the available data to complete the compliance processing.

[0065] The priority sorting rule can be a field retention priority list. The sensitive field mapping table can be a mapping relationship table established based on the desensitization requirements. The field replacement strategy can be a specific data processing rule generated based on the sensitive field mapping table. The available data can be compliance candidate data. The authorized field can be the original data field. The sensitive data can be fields containing personal privacy, business secrets, or fields restricted by regulations.

[0066] Specifically, according to the service scenario identifier, load the predefined scenario-authorization whitelist association table to determine the priority sorting rule of authorization fields in the current scenario. Then, parse the desensitization requirements in the customer agreement and construct a sensitive field mapping table in combination with industry data compliance standards. Furthermore, based on the sensitive field mapping table, establish a field replacement strategy for different data types. Classify and process according to the data storage format based on the data type to determine the available data: First, extract the target fields according to the authorization whitelist and hard-filter the unauthorized columns; then, scan the text based on the regular expression engine to identify the sensitive field patterns and mark the available data. Finally, retain the authorized fields of the available data according to the priority sorting rule, standardize the format of the retained fields; and apply the field replacement strategy to eliminate sensitive data; generate an anonymized dataset with a version identifier.

[0067] Through this solution, determining the priority sorting rule of the authorization whitelist according to the service scenario identifier helps to eliminate the problem of feature extraction distortion. Analyzing the desensitization requirements and determining the sensitive field mapping table helps to eliminate the problems of privacy leakage risk and data noise interference. Establishing a field replacement strategy based on the sensitive field mapping table achieves a balance between compliance processing and business feature retention. Filtering the existing data according to the data type to obtain the available data solves the problems of data redundancy and compliance conflicts. Retaining the authorized fields of the available data according to the priority sorting rule and applying the field replacement strategy to process the sensitive data in the available data to complete the compliance processing, achieving the dual goals of privacy protection and business analysis, eliminating the privacy leakage risk, and at the same time improving data consistency through standardization processing.

[0068] In some embodiments, based on the timestamp, decompose the resource consumption trend to determine the resource consumption cycle; parse the concurrent processing requirements to determine the resource usage pattern; according to the resource consumption cycle and the resource usage pattern, generate elastic demand parameters.

[0069] The resource consumption cycle can be a regular fluctuation feature extracted from the resource consumption trend.

[0070] The resource usage pattern can be the corresponding relationship between business operation characteristics and resource consumption.

[0071] Specifically, align the resource consumption data based on a unified timestamp to eliminate time zone differences and sampling interval biases. Then, apply the STL decomposition algorithm to separate the periodic term, trend term, and residual term. Furthermore, detect the frequency component with the strongest energy in the periodic term through Fourier transform and mark it as the resource consumption period. Extract the number of concurrent requests, request type distribution, and single-request resource occupancy weight from the available data. Then, according to the service scenario identifier, load the predefined resource usage pattern template to generate the resource usage pattern. Furthermore, generate the periodic drive parameter based on the length and amplitude of the resource consumption period. Then, generate the concurrent mode parameter according to the resource usage pattern. Fuse the periodic drive parameter and the concurrent mode parameter according to the scenario weight to generate the elastic demand parameter.

[0072] Through this solution, based on the timestamp, decompose the resource consumption trend, determine the resource consumption period, eliminate the baseline drift caused by time zone differences or inconsistent sampling intervals, ensure the global comparability of time series analysis, and provide standardized input for elastic computing. Analyze the concurrent processing requirements, determine the resource usage pattern, avoid the defect of not establishing a dynamic association mechanism between the business type and the resource usage pattern, and quantify the contribution degree of different scenarios to the elastic demand. Generate the elastic demand parameter according to the resource consumption period and the resource usage pattern, which helps to eliminate the hysteresis problem and improve the prediction accuracy.

[0073] In some embodiments, analyze the elastic demand parameter to determine the expansion trigger point and the expansion mode. According to the expansion trigger point and the expansion mode, generate the resource configuration suggestion.

[0074] The expansion trigger point can be the resource expansion execution time point. The expansion mode can be the resource expansion strategy.

[0075] Specifically, calculate the dynamic trigger baseline and the preventive trigger time based on the periodic drive parameter and the concurrent mode parameter in the elastic demand parameter. Then, dynamically correct the trigger baseline according to the standard deviation of the residual term of the resource consumption trend and the resource pool elasticity coefficient. Subsequently, analyze the SLA default cost data and convert the default cost into the expansion urgency weight. Finally, combine the dynamic trigger baseline and the expansion urgency weight to generate the expansion trigger point.

[0076] Load predefined decision rules according to the resource usage pattern type and cost sensitivity classification in the elastic demand parameters to match the mode selection rules; then, verify the instance startup delay and network bandwidth margin constrained by the elasticity coefficient; subsequently, calculate the cost sensitivity based on the customer's historical payment cycle and SLA default cost data; finally, generate the scaling-up mode by combining the results of the mode selection rules, the cost sensitivity weight, and the corrected value of the elasticity coefficient constraint. Decompose the scaling-up trigger point into a timestamp, a trigger condition, and a scenario code; according to the scaling-up mode, inject a horizontal scaling parameter set and a vertical scaling parameter set into the configuration suggestion; furthermore, combine the timestamp, the trigger condition, the scenario code, and the scaling parameter set into a structured JSON object; subsequently, automatically append data desensitization rules according to the scenario code; add a two-factor authentication mark to sensitive operations; thus, generate a resource configuration suggestion.

[0077] Through this solution, analyzing the elastic demand parameters, determining the scaling-up trigger point and the scaling-up mode helps to eliminate the problem of missing detection of unexpected events in the rough scaling-up trigger point, achieve the goal of business-resource coupling analysis, and avoid misjudgment of business characteristics. Generate a resource configuration suggestion according to the scaling-up trigger point and the scaling-up mode. In some embodiments, based on the customer agreement, obtain the SLA default cost data; parse the SLA default cost data to quantify the service interruption tolerance; obtain the real-time resource information; analyze the real-time resource information to determine the availability metric; analyze the availability metric to calculate the resource supply elasticity coefficient; parse the historical bill payment cycle to determine the cost sensitivity; determine the scaling-up mode according to the service interruption tolerance, the resource supply elasticity coefficient, and the cost sensitivity.

[0078] The SLA default cost data may be the compensation clause data clearly stipulated in the customer agreement due to service interruption. The service interruption tolerance may be an indicator used to characterize the customer's tolerance of service interruption risk. The real-time resource information may be the resource pool status data collected in real time. The availability metric may be a quantified metric generated after analyzing the real-time resource information. The resource supply elasticity coefficient may be a dynamic value reflecting the current scaling-up ability of the resource pool. The historical bill payment cycle may be the payment mode characteristics reflected in the customer's historical payment records. The cost sensitivity may be the customer's sensitivity to the resource redundancy cost.

[0079] Specifically, extract SLA default cost data from the customer agreement. Based on the SLA default cost data, calculate the compensation weight factor according to the proportion of the maximum compensation amount for a single service interruption to the total contract amount; then, classify by the maximum allowed interruption duration to determine the time sensitivity level; subsequently, generate the service interruption tolerance based on the compensation weight factor and the time sensitivity level. Obtain real-time resource information through the cloud platform API. Then, normalize the real-time resource information to determine the availability metric. According to the availability metric, generate the resource supply elasticity coefficient using the elasticity coefficient calculation formula. Extract the payment cycle type, historical payment punctuality rate, and resource release delay tolerance from the billing database; then, determine the cost sensitivity according to the payment mode. Generate the scaling mode according to the service interruption tolerance, resource supply elasticity coefficient, and cost sensitivity based on the decision matrix result.

[0080] Through this solution, based on the customer agreement, obtaining SLA default cost data helps to eliminate the problem of missing decision-making basis caused by ignoring the quantification of SLA default costs. Analyze the SLA default cost data, quantify the service interruption tolerance, and eliminate the problem of policy wavering caused by the ambiguity of multi-clause weights. Obtain real-time resource information and overcome the blind spot of scaling decision-making caused by the lack of resource supply elasticity coefficient. Analyze the real-time resource information, determine the availability metric, which helps to eliminate the decision-making delay caused by the unstructured evaluation of resource status. Analyze the availability metric, calculate the resource supply elasticity coefficient, compress the multi-dimensional resource status into a single decision-making parameter, and eliminate the policy conflict caused by the isolated analysis of multi-dimensional metrics. Analyze the historical bill payment cycle, determine the cost sensitivity, which helps to overcome the problem of inefficient resource reuse strategy. Determine the scaling mode according to the service interruption tolerance, resource supply elasticity coefficient, and cost sensitivity, which helps to achieve the design goal of the multi-dimensional decision matrix.

[0081] In some embodiments, obtain the historical scaling records, analyze the historical scaling records based on the scaling trigger point to determine the scaling response delay time; analyze the resource consumption trend to determine the resource depletion countdown; generate a preventive scaling time window according to the time difference between the scaling response delay time and the resource depletion countdown; generate a resource configuration recommendation including a time window identifier according to the preventive scaling time window and the scaling mode.

[0082] The historical scaling records can be a structured data set stored in the scaling operation log database, recording the scaling trigger time, scaling effective time, and scaling mode of each scaling operation.

[0083] The scaling response delay time can be a statistic of the time difference from the trigger of the scaling operation to the actual effectiveness of the new resource instance.

[0084] The resource depletion countdown can be the remaining time predicted for the complete depletion of the resource pool.

[0085] The time difference can be the difference between the countdown for resource exhaustion and the delay time of the capacity expansion response.

[0086] The preventive capacity expansion time window can be the execution time period of the capacity expansion operation generated according to the time difference.

[0087] The time window identifier can be a metadata tag attached to the resource configuration suggestion.

[0088] Specifically, extract the capacity expansion trigger time, capacity expansion effective time, and capacity expansion mode from the capacity expansion operation log database; then, for historical records of the same capacity expansion mode, calculate the difference between the capacity expansion effective time and the capacity expansion trigger time to determine the capacity expansion response delay time. Based on the resource consumption trend monitored in real time, according to the linear regression model, with the time series as the independent variable and the resource consumption as the dependent variable, fit the linear equation by the least squares method, and calculate the slope as the resource consumption rate; then, combine the current remaining resource amount to predict the countdown for resource exhaustion. According to the capacity expansion response delay time and the countdown for resource exhaustion, calculate the time difference to generate the preventive capacity expansion time window. According to the capacity expansion mode and the preventive capacity expansion time window, select the operation instruction from the preset policy library; then, attach the time window metadata to the resource configuration suggestion to generate a resource configuration suggestion containing the time window identifier.

[0089] Through this solution, obtain historical capacity expansion records, analyze the historical capacity expansion records based on the capacity expansion trigger point, determine the capacity expansion response delay time, cover the delay fluctuations in extreme scenarios, and ensure the robustness of the time window calculation. Analyze the resource consumption trend, determine the countdown for resource exhaustion, and avoid one-sided decisions caused by a single indicator. Generate the preventive capacity expansion time window according to the time difference between the capacity expansion response delay time and the countdown for resource exhaustion, which helps to eliminate the problem of a sharp increase in emergency capacity expansion costs caused by not reserving the delay time. Generate a resource configuration suggestion containing the time window identifier according to the preventive capacity expansion time window and the capacity expansion mode, and ensure the global consistency of the time window policy with the cost sensitivity and service stability requirements.

[0090] In some embodiments, based on the resource usage pattern, analyze the server operation data to determine the cause of mode switching; analyze the cause of mode switching to determine the cause attribute; according to the cause attribute, analyze the resource consumption cycle to determine the influence weight of each cause of mode switching on the resource usage pattern; according to the influence weight, generate the elastic demand parameter.

[0091] The cause of mode switching can be an event associated with the business operation characteristics and resource consumption characteristics.

[0092] The cause attribute can be an attribute category divided according to the influence persistence and predictability of the cause of mode switching on resource consumption.

[0093] The impact weight can be a normalized value that quantifies the impact degree of a single mode switch incentive on the resource usage pattern.

[0094] Specifically, extract the API call sequence strongly associated with the business scenario from the server operation data; then, based on the scenario type coding table, map the original operation log to a standardized business event identifier; subsequently, collect the resource consumption time series data in real time; furthermore, eliminate short-term noise through the moving average method and identify the sudden increase or decrease inflection points synchronized with the business event; thus, determine the mode switch incentive. Based on the mode switch incentive, according to the sensitive field mapping table, perform dynamic desensitization on the original data associated with the incentive; thus, determine the incentive attribute. Calculate the standard deviation change rate of the residual term when each mode switch incentive occurs; then, use the variance contribution analysis method to calculate the proportion of the explanatory power of each mode switch incentive in the resource consumption cycle; thus, determine the impact weight of each mode switch incentive on the resource usage pattern. Furthermore, calculate the expansion urgency coefficient according to the product of the incentive weight and the resource exhaustion countdown; then, match the preset template according to the incentive type to determine the resource demand increment; if the incentive attribute is external and sudden, superimpose the safety redundancy in the elastic parameter; at the same time, if there is a downward trend in the periodic term associated with the incentive, reduce the resource demand increment according to the trend slope; thus, generate the final elastic demand parameter.

[0095] Through this solution, based on the resource usage pattern, analyze the server operation data, determine the mode switch incentive, and eliminate the prediction distortion problem caused by failing to distinguish business operation characteristics from noise events. Analyze the mode switch incentive and determine the incentive attribute, which helps to eliminate the problem of missing the sudden event detection window due to failure to analyze the residual term and improve the prediction accuracy of the expansion trigger point. According to the incentive attribute, analyze the resource consumption cycle, determine the impact weight of each mode switch incentive on the resource usage pattern, and avoid the resource allocation imbalance caused by single-dimensional decision-making. According to the impact weight, generate the elastic demand parameter, which helps to support the generation of a multi-dimensional decision matrix and achieve the balance of cost, stability, and response speed.

[0096] Figure 3 The following is a schematic structural diagram of an elastic demand calculation application system for server transactions provided by an embodiment of the present application, as Figure 3 shown, the elastic demand calculation application system 300 for server transactions in this embodiment includes: a data analysis module 301, an information analysis module 302, a type determination module 303, a demand determination module 304, and a suggestion generation module 305.

[0097] The data analysis module 301 is used to obtain the customer agreement and server operation data; analyze the server operation data to determine the operation resource information; The information analysis module 302 is used to analyze the operation resource information to determine the resource consumption trend and resource allocation situation; A type determination module 303, configured to determine the customer service type according to the customer agreement; A requirement determination module 304, configured to determine the concurrent processing requirement according to the customer service type and the resource allocation situation; A recommendation generation module 305, configured to generate elastic requirement parameters according to the resource consumption trend and the concurrent processing requirement, and generate a resource configuration recommendation according to the elastic requirement parameters.

[0098] Optionally, the elastic demand calculation application system for server transactions further includes a data determination module 306, configured to: parse the customer agreement to determine an authorized whitelist, a desensitization requirement, and a service scenario identifier; determine the data types of the operational data that can be obtained according to the authorized whitelist; perform compliance processing on the existing data according to the data types, the desensitization requirement, and the service scenario identifier, and use the dataset after the compliance processing as the server operational data.

[0099] Optionally, when determining the customer service type according to the customer agreement, the type determination module 303 is configured to: obtain a scenario type code based on the service scenario identifier; analyze the server operational data to determine the API call sequence during the operation process; determine the business operation mode feature according to the API call sequence; match the business operation mode feature with the scenario type code; and determine the customer service type according to the matching degree between the scenario type code and the business operation mode feature.

[0100] Optionally, when performing compliance processing on the existing data according to the data types, the desensitization requirement, and the service scenario identifier, the data determination module 306 is configured to: determine the priority sorting rule of the authorized whitelist according to the service scenario identifier; analyze the desensitization requirement to determine a sensitive field mapping table; establish a field replacement policy according to the sensitive field mapping table; filter the existing data according to the data types to obtain available data; retain the authorized fields of the available data according to the priority sorting rule, and process the sensitive data in the available data using the field replacement policy to complete the compliance processing.

[0101] Optionally, when generating elastic requirement parameters according to the resource consumption trend and the concurrent processing requirement, the recommendation generation module 305 is configured to: decompose the resource consumption trend based on a time stamp to determine a resource consumption period; parse the concurrent processing requirement to determine a resource usage mode; and generate elastic requirement parameters according to the resource consumption period and the resource usage mode.

[0102] Optionally, when generating a resource configuration suggestion according to the elastic demand parameter, the suggestion generation module 305 is configured to: analyze the elastic demand parameter to determine an expansion trigger point and an expansion mode; and generate a resource configuration suggestion according to the expansion trigger point and the expansion mode.

[0103] Optionally, the elastic demand calculation application system for server transactions further includes a mode determination module 307, configured to: obtain SLA default cost data based on the customer agreement; parse the SLA default cost data to quantify the service interruption tolerance; obtain real-time resource information; analyze the real-time resource information to determine an availability metric; analyze the availability metric to calculate a resource supply elasticity coefficient; parse the historical bill payment cycle to determine the cost sensitivity; and determine an expansion mode according to the service interruption tolerance, the resource supply elasticity coefficient, and the cost sensitivity.

[0104] Optionally, when generating a resource configuration suggestion according to the expansion trigger point and the expansion mode, the suggestion generation module 305 is configured to: obtain a historical expansion record, analyze the historical expansion record based on the expansion trigger point to determine an expansion response delay time; analyze the resource consumption trend to determine a resource exhaustion countdown; generate a preventive expansion time window according to the time difference between the expansion response delay time and the resource exhaustion countdown; and generate a resource configuration suggestion including a time window identifier according to the preventive expansion time window and the expansion mode.

[0105] Optionally, when generating an elastic demand parameter according to the resource consumption cycle and the resource usage pattern, the suggestion generation module 305 is configured to: analyze the server operation data based on the resource usage pattern to determine an incentive for mode switching; analyze the incentive for mode switching to determine an incentive attribute; analyze the resource consumption cycle according to the incentive attribute to determine the influence weight of each incentive for mode switching on the resource usage pattern; and generate an elastic demand parameter according to the influence weight.

[0106] The system of this embodiment can be used to execute the method of any of the above embodiments, and its implementation principle and technical effects are similar, which will not be elaborated here.

Claims

1. A method for calculating elastic demand in server transactions, characterized in that, Including: Obtain the customer agreement and server operation data; Analyze the server operation data to determine the operation resource information; Analyze the operation resource information to determine the resource consumption trend and resource allocation situation; Determine the customer business type according to the customer agreement; Determine the concurrent processing requirements according to the customer business type and the resource allocation situation; Generate elastic demand parameters according to the resource consumption trend and the concurrent processing requirements, and generate resource configuration suggestions according to the elastic demand parameters.

2. The method according to claim 1, characterized in that, The obtaining of the server operation data includes: Parse the customer agreement to determine the authorized whitelist, desensitization requirements, and service scenario identifier; Determine the data types for which operation data can be obtained according to the authorized whitelist; Perform compliance processing on the existing data according to the data types, the desensitization requirements, and the service scenario identifier, and use the dataset after compliance processing as the server operation data.

3. The method according to claim 2, wherein The determining of the customer business type according to the customer agreement includes: Obtain the scenario type code based on the service scenario identifier; Analyze the server operation data to determine the API call sequence during operation; Determine the business operation mode characteristics according to the API call sequence; Match the business operation mode characteristics with the scenario type code; Determine the customer business type according to the matching degree between the scenario type code and the business operation mode characteristics.

4. The method according to claim 2, wherein The performing of compliance processing on the existing data according to the data types, the desensitization requirements, and the service scenario identifier includes: Determine the priority sorting rule of the authorized whitelist according to the service scenario identifier; Analyze the desensitization requirements to determine the sensitive field mapping table; Establish a field replacement strategy according to the sensitive field mapping table; Filter the existing data according to the data types to obtain available data; Retain the authorized fields of the available data according to the priority sorting rule, and apply the field replacement strategy to process the sensitive data in the available data to complete the compliance processing.

5. The method according to claim 1, wherein The generating of elastic demand parameters according to the resource consumption trend and the concurrent processing requirements includes: Decompose the resource consumption trend based on the timestamp to determine the resource consumption cycle; Parse the concurrent processing requirements to determine the resource usage mode; Generate elastic demand parameters according to the resource consumption cycle and the resource usage mode.

6. The method according to claim 1, wherein The generating of resource configuration suggestions according to the elastic demand parameters includes: Analyze the elastic demand parameters to determine the expansion trigger point and expansion mode; Generate resource configuration suggestions according to the expansion trigger point and the expansion mode.

7. The method according to claim 6, wherein The analyzing of the elastic demand parameters to determine the expansion mode includes: Obtain the SLA default cost data based on the customer agreement; Parse the SLA default cost data to quantify the service interruption tolerance; Obtain the real-time resource information; Analyze the real-time resource information to determine the availability index; Analyze the availability index to calculate the resource supply elasticity coefficient; Parse the historical bill payment cycle to determine the cost sensitivity; Determine the expansion mode according to the service interruption tolerance, the resource supply elasticity coefficient, and the cost sensitivity.

8. The method according to claim 6, characterized in that, Generating a resource configuration recommendation according to the capacity expansion trigger point and the capacity expansion mode, including: Obtaining historical capacity expansion records, analyzing the historical capacity expansion records based on the capacity expansion trigger point, and determining the capacity expansion response delay time; Analyzing the resource consumption trend and determining the resource exhaustion countdown; Generating a preventive capacity expansion time window according to the time difference between the capacity expansion response delay time and the resource exhaustion countdown; Generating a resource configuration recommendation including a time window identifier according to the preventive capacity expansion time window and the capacity expansion mode.

9. The method according to claim 5, wherein Generating elastic demand parameters according to the resource consumption cycle and the resource usage mode, including: Analyzing the server operation data based on the resource usage mode and determining the mode switching incentive; Analyzing the mode switching incentive and determining the incentive attribute; According to the incentive attribute, analyzing the resource consumption cycle and determining the influence weight of each mode switching incentive on the resource usage mode; Generating elastic demand parameters according to the influence weight.

10. An elastic demand calculation application system for server transactions, characterized in that, Applied to the method according to any one of claims 1-9, including: A data analysis module for obtaining a customer agreement and server operation data; analyzing the server operation data and determining operation resource information; An information analysis module for analyzing the operation resource information and determining the resource consumption trend and resource allocation situation; A type determination module for determining the customer service type according to the customer agreement; A demand determination module for determining the concurrent processing demand according to the customer service type and the resource allocation situation; A recommendation generation module for generating elastic demand parameters according to the resource consumption trend and the concurrent processing demand, and generating a resource configuration recommendation according to the elastic demand parameters.

Images