A method and system for configuring resources for multi-product service activation

Abstract

The application provides a multi-product service opening resource configuration method and system, belongs to the technical field of resource configuration, and comprises the following steps: constructing a service resource parameter library, forming a mapping relationship between resources and multi-product service demands, collecting and processing real-time resource operation data required by multi-product service opening to obtain effective resource data; analyzing multi-product information input by a user, extracting demand keywords and matching the service resource parameter library to generate a product-resource configuration file; predicting resource demand and generating a pre-scheduling instruction based on the effective resource data and historical service data; performing resource allocation and protocol binding according to the pre-scheduling instruction to complete product service resource configuration, monitoring the resource configuration process, identifying and processing exceptions, and updating exception information to an exception knowledge base. The application solves the problems that resource and service demand matching is not accurate in the multi-product service opening process, resource allocation and protocol binding efficiency is low and compatibility is difficult to guarantee, and configuration process exceptions are difficult to identify and process in time.

Description

Technical Field

[0001] This invention relates to the field of resource allocation technology, and in particular to a method and system for resource allocation for multi-product service activation. Background Technology

[0002] As businesses diversify and multiple products are launched simultaneously, the demand for shared access and terminal resources is becoming increasingly urgent. It is necessary to achieve precise matching of "multiple products and resources" through resource allocation to support the stable implementation of the business. At the same time, it is necessary to take into account both resource utilization efficiency and configuration efficiency to avoid resource idleness or business launch delays.

[0003] The current resource allocation process for multi-product service activation suffers from several technical deficiencies: First, there is a lack of a unified business resource parameter library. The mapping relationship between resources and multi-product business requirements relies on manual establishment, which is not only inefficient but also prone to mismatch between resources and business requirements due to human judgment bias. Second, multi-product information parsing relies on manual extraction of requirements, lacking automated keyword recognition and library matching mechanisms. When there are many product types and complex functions, requirement parsing is time-consuming and prone to missing key requirements. Third, resource requirement prediction is mostly based on historical experience values, without combining real-time resource operation data and scientific algorithms, making it difficult to accurately predict peak resource demand, resulting in insufficient resources during peak periods and idle resources during off-peak periods. Fourth, in the process of resource allocation and protocol binding, priorities are not divided according to business importance, and there is a lack of protocol and resource compatibility verification, which easily leads to high-priority business resources being occupied and configuration failures caused by protocol and resource incompatibility. Fifth, the monitoring and handling of anomalies in the configuration process relies on manual investigation, which cannot identify resource operation deviations in real time, resulting in delayed anomaly handling and further extending the service activation cycle.

[0004] Therefore, the aforementioned shortcomings in existing technologies directly lead to low efficiency in multi-product service activation, low resource utilization, and high configuration anomaly rates. These problems are particularly pronounced in cross-industry multi-product scenarios, where significant differences in business needs and complex resource types exacerbate the issues. These problems not only increase enterprise operating costs but may also negatively impact customer experience due to service activation delays or configuration anomalies, failing to meet the enterprise's need for large-scale, efficient multi-product activation. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, the purpose of this invention is to provide a method and system for configuring resources for multi-product business activation. This method integrates complex logic such as multi-product resource sharing relationships and configuration order control into the system, eliminating the need to pre-define fixed configuration processes. It returns to the essential nature of resource and business carrying, significantly reducing the number of processes, while improving the accuracy of resource demand prediction and the efficiency of anomaly handling. This makes multi-product business activation simpler and easier to interact with.

[0006] To achieve the above objectives, the present invention provides the following solution:

[0007] One objective of this invention is to provide a method for configuring resources for activating multiple product services, comprising the following steps:

[0008] S1. Construct a business resource parameter library to form a mapping relationship between resources and the business requirements of multiple products, collect and process real-time resource operation data required for the activation of multiple product services, and obtain effective resource data.

[0009] S2. Parse the multi-product information input by the user, extract the required keywords and match them with the business resource parameter library to generate a product-resource configuration file;

[0010] S3. Based on the effective resource data and historical business data, predict resource demand and generate pre-scheduling instructions;

[0011] S4. Perform resource allocation and protocol binding according to the pre-scheduling instructions to complete the product business resource configuration. During the resource configuration monitoring process, identify and handle anomalies and update the anomaly information to the anomaly knowledge base.

[0012] Preferably, in S1, the construction of the business resource parameter library includes:

[0013] Sub-libraries are divided according to resource function;

[0014] Establish a mapping relationship between sub-library resources and business requirements, specifically: map bandwidth parameters according to business data transmission volume, map encryption protocol parameters according to business security level, and map computing power parameters according to business concurrency.

[0015] The collection and processing of real-time resource operation data required for the activation of multiple product services includes:

[0016] Collect real-time resource status information, including port traffic, CPU utilization, voltage stability, and memory usage.

[0017] The collected real-time status information of the resources is denoised and standardized to obtain valid resource data. The format of the valid resource data matches the storage format of the business resource parameter card.

[0018] Preferably, in S2, parsing the multi-product information input by the user, extracting requirement keywords and matching them with the business resource parameter library to generate a product-resource configuration file includes:

[0019] Natural language processing algorithms are used to parse multi-product information input by users and extract no less than two types of requirement keywords: industry keywords representing the business field and functional requirement keywords representing the core functions of the business.

[0020] Based on the industry keywords, the corresponding type sub-library in the business resource parameter library is matched to obtain the range of suitable resource types;

[0021] Within the scope of the resource types, specific resource parameters are matched based on the functional requirement keywords to determine the appropriate resource model, corresponding protocol type, and resource requirement.

[0022] By integrating the resource models, protocol types, and resource requirements, a standardized product-resource configuration file is generated.

[0023] Preferably, in S3, predicting resource requirements based on the valid resource data and historical business data includes:

[0024] Summarize valid resource data with historical business data of the same period in the past, with the preset quantity determined by the historical periodic pattern of business activation.

[0025] The peak resource demand for the predicted duration is calculated using a pre-set algorithm to determine the backup resource data that needs to be activated. The formula is as follows:

[0026] ;

[0027] in, For the future Peak resource demand at any given moment; The weights used for time series model fusion have a value range of 0 to 1 and are adaptively adjusted by backpropagation of historical prediction errors. For Long Short-Term Memory Networks to assess the temporal characteristics of resources The predicted output, resource time series characteristics It is a time-series sequence containing resource operation data from the past 24 hours; For gated loop units to access business characteristics Predicted output, business characteristics This is a business feature vector that includes the number of services launched during the same period and the proportion of product types. The business load impact coefficient has a value range of 0.1 to 0.5 and is obtained by training the correlation between business type and resource consumption. This represents the duration of the current peak business period, in hours. This represents the current idle period in hours. The predicted duration is in hours.

[0028] Preferably, in S4, the step of performing resource allocation and protocol binding according to the pre-scheduling instruction to complete the multi-product service resource configuration includes:

[0029] The resource allocation priority and protocol binding rules in the pre-scheduling instruction are parsed. The priorities are sorted according to the importance level of the business, and the importance level is pre-stored in the business resource parameter library.

[0030] Resource allocation is performed sequentially according to the aforementioned priority: core resources are reserved for high-priority services, and edge resources are allocated for low-priority services. The resource allocation results are then recorded.

[0031] Based on the protocol type in the product-resource configuration file, the allocated resources are bound to the corresponding business protocol. During the binding process, the compatibility between the protocol and the resource is verified. If they are incompatible, the parameter library matching and correction is triggered, and the compatible protocol is re-determined.

[0032] Integrate resource allocation results with protocol binding results to generate a resource configuration completion report for the activation of multiple product services;

[0033] The process of identifying and handling anomalies during resource configuration monitoring includes:

[0034] Real-time comparison of the deviation between the real-time operating parameters of resources and the standard operating parameters of resources pre-stored in the business resource parameter library;

[0035] When the deviation exceeds the anomaly judgment threshold pre-stored in the business resource parameter library, an anomaly handling instruction is generated, and the operation of switching the backup resource to the preset backup port or power-off isolation of the faulty resource is performed. The anomaly type, handling instruction and adapted resource information are updated to the anomaly knowledge base.

[0036] The second objective of this invention is to provide a multi-product service activation resource configuration system, comprising:

[0037] The parameter library construction module is used to build a business resource parameter library and establish a mapping relationship between resources and business requirements of multiple products.

[0038] The data processing module is used to collect and process real-time resource operation data required for the activation of multiple product services, and output valid resource data.

[0039] The information parsing module is used to parse multi-product information input by the user, extract requirement keywords and match them with the business resource parameter library to generate product-resource configuration profiles;

[0040] The demand forecasting module is used to predict resource demand and generate pre-scheduling instructions based on available resource data and historical business data.

[0041] The configuration execution module is used to perform resource allocation and protocol binding according to pre-scheduled instructions, and complete the configuration of multi-product business resources.

[0042] The exception handling module is used to monitor the resource configuration process, identify and handle exceptions, and update the exception knowledge base.

[0043] Preferably, the information parsing module includes:

[0044] The keyword extraction unit is used to parse multi-product information using natural language processing algorithms and extract industry keywords and functional requirement keywords.

[0045] The library matching unit is used to match the corresponding type sub-library based on industry keywords, and then match specific resource parameters in the sub-library based on functional requirement keywords;

[0046] The file generation unit is used to integrate resource models, protocol types, and resource requirements to generate standardized product-resource configuration files.

[0047] Preferably, the demand forecasting module includes:

[0048] The data aggregation unit is used to integrate valid resource data with a preset quantity of historical business data from the same period in the past, the preset quantity being determined by the historical periodic patterns of business activation.

[0049] The peak calculation unit is used to calculate the peak resource demand for the predicted future duration using a preset algorithm, and to determine the number of standby resources that need to be activated.

[0050] Preferably, the configuration execution module includes:

[0051] The instruction parsing unit is used to parse the resource allocation priority and protocol binding rules in the pre-scheduling instruction, wherein the priority is sorted according to the level of business importance;

[0052] The resource allocation unit is used to perform resource allocation according to priority, reserving core resources for high-priority services and allocating peripheral resources for low-priority services;

[0053] The protocol binding unit is used to bind the allocated resources with the corresponding business protocol. During the binding process, compatibility is verified, and parameter library matching and correction are triggered when incompatibility occurs.

[0054] The results generation unit is used to integrate the resource allocation and protocol binding results to generate a resource configuration completion report.

[0055] Preferably, the exception handling module includes:

[0056] The deviation monitoring unit is used to compare the deviation between the real-time operating parameters of resources and the pre-stored standard operating parameters of resources in the business resource parameter library.

[0057] The instruction generation unit is used to generate a processing instruction for switching the backup resource to a preset backup port or for power-off isolation of the faulty resource when the deviation exceeds the anomaly judgment threshold pre-stored in the business resource parameter library, and to trigger an update of the anomaly knowledge base.

[0058] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0059] (1) This invention improves the accuracy of matching multiple products and resources and solves the problem of manual matching bias. By constructing a business resource parameter library containing resource-business mapping relationship, and combining natural language processing algorithm to extract industry and functional requirement keywords and match the parameter library to generate standardized product-resource configuration files, it replaces the manual mapping and parsing of requirements, avoids the problem of resource and business requirement mismatch caused by human judgment bias, and transforms the parsing of multiple product information from manual operation to automated processing, which greatly improves the efficiency of requirement parsing and resource matching.

[0060] (2) This invention improves the accuracy of resource demand prediction and configuration execution efficiency, optimizes resource utilization, and predicts the peak of resource demand based on effective resource data and historical business data by integrating LSTM, GRU models and business load influence factors through a preset algorithm. It can accurately predict the resource demand in the future within a certain period of time, avoiding resource shortage during peak periods and idle resources during off-peak periods. At the same time, resources are allocated according to the priority of business importance, and compatibility is verified and correction is triggered when binding protocols. This ensures the supply of high-priority business resources and reduces configuration failures caused by protocol incompatibility, thereby improving resource configuration execution efficiency and resource utilization.

[0061] (3) This invention ensures the stability of the multi-product resource configuration process and reduces the risk of service activation. By comparing the deviation of resource operation parameters with standard parameters in real time and automatically performing backup resource switching or fault isolation when the threshold is exceeded, and updating the abnormal knowledge base, it realizes real-time identification and rapid processing of abnormalities, reducing the interruption of service activation caused by abnormalities. Moreover, the various modules of the system work together to integrate complex resource sharing relationships and configuration control into the system. There is no need to preset fixed processes, returning to the essence of resource and service carrying, greatly reducing the number of configuration processes, and reducing the complexity of system operation and maintenance and service activation. Attached Figure Description

[0062] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0063] Figure 1 A flowchart illustrating the principle of a multi-product service activation resource configuration method provided in this embodiment of the invention;

[0064] Figure 2 This is a schematic diagram of a multi-product service activation resource configuration system provided in an embodiment of the present invention. Detailed Implementation

[0065] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0066] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0067] One embodiment of the present invention, such as Figure 1 As shown, this embodiment provides a method for configuring resources for enabling multi-product services, including the following steps:

[0068] S1. Construct a business resource parameter library to form a mapping relationship between resources and the business requirements of multiple products, collect and process real-time resource operation data required for the activation of multiple product services, and obtain effective resource data.

[0069] S2. Parse the multi-product information input by the user, extract the required keywords and match them with the business resource parameter library to generate a product-resource configuration file;

[0070] S3. Based on the effective resource data and historical business data, predict resource demand and generate pre-scheduling instructions;

[0071] S4. Perform resource allocation and protocol binding according to the pre-scheduling instructions to complete the product business resource configuration. During the resource configuration monitoring process, identify and handle anomalies and update the anomaly information to the anomaly knowledge base.

[0072] In step S1 above, constructing the business resource parameter library includes dividing the resource into sub-libraries according to resource functions and establishing a mapping relationship between sub-library resources and business requirements. Specifically, this involves establishing corresponding associations with resource bandwidth parameters, encryption protocol parameters, and computing power parameters through three dimensions: business data transmission volume, business security level, and business concurrency. The collection and processing of real-time resource operation data required for the activation of multiple product services includes collecting key information reflecting the resource operation status, denoising the collected information to eliminate interference data, and standardizing the data to unify the data format. The final valid resource data must be consistent with the storage format of the business resource parameter library to ensure the smoothness of subsequent data calls and matching.

[0073] Furthermore, in practical implementation, the construction of the business resource parameter library aims to achieve a precise quantitative correlation between business requirements and resource parameters: sub-libraries divided by resource function must cover access, computing power, and terminal resources required for business activation. Resource parameters within each sub-library must form clear correspondence rules with business requirement dimensions. For example, based on different ranges of business data transmission volume, access resources with corresponding bandwidth capabilities are matched; based on differences in business security levels, resources with corresponding encryption capabilities are matched; based on the scale of business concurrency, resources meeting computing power requirements are matched. This correlation must be determined based on the compatibility analysis of business characteristics and resource capabilities to avoid resource mismatch with requirements. In the real-time resource operation data processing stage, noise reduction requires the adoption of adaptation strategies for different types of resource status information to eliminate the impact of instantaneous fluctuations or abnormal interference on data validity; standardization processing requires unifying the data volume and format to ensure that valid resource data can be directly compared and correlated with resource parameters in the parameter library. If there are format differences, automatic adjustment and correction are required to ensure data compatibility.

[0074] In step S2 above, parsing the multi-product information input by the user, extracting requirement keywords and matching them with the business resource parameter library to generate a product-resource configuration file includes using natural language processing technology to parse the multi-product information input by the user and extracting two types of key information: one type is industry keywords representing the business's domain, and the other type is functional requirement keywords representing the core functions of the business; based on the extracted industry keywords, matching the corresponding type sub-library in the business resource parameter library narrows the resource selection range; within the determined resource type range, further matching specific resource parameters in combination with functional requirement keywords to clarify the suitable resource model, corresponding protocol type, and required resource quantity; finally, integrating the above information to form a standardized product-resource configuration file with a unified format and complete content.

[0075] Specifically, the key to this step lies in achieving precise matching of multi-product information and resources through automated parsing and hierarchical matching: the application of natural language processing technology needs to focus on targeted keyword identification to ensure accurate differentiation between industry attributes and functional requirements, and the two types of keywords need to be associated to avoid interpreting functional requirements in isolation from the industry context; the matching of industry keywords and type sub-libraries needs to be based on preset domain-resource correspondence rules to ensure that the selected sub-libraries can cover all resource types required by the industry's business; the matching of functional requirement keywords and specific resource parameters needs to delve into the details of resource capabilities, and determine the resource model, protocol type and quantity in combination with the actual needs of business functions to avoid insufficient or redundant resource capabilities; the generated product-resource configuration file needs to contain the core information of the business and resource association, with complete fields and standardized format, providing a clear basis for subsequent resource configuration.

[0076] In step S3 above, predicting resource demand based on the effective resource data and historical business data includes:

[0077] Summarize valid resource data with historical business data of the same period in the past, with the preset quantity determined by the historical periodic pattern of business activation.

[0078] The peak resource demand for the predicted duration is calculated using a pre-set algorithm to determine the backup resource data that needs to be activated. The formula is as follows:

[0079] ;

[0080] in, For the future Peak resource demand at any given moment; The weights used for time series model fusion have a value range of 0 to 1 and are adaptively adjusted by backpropagation of historical prediction errors. For Long Short-Term Memory networks to assess the temporal characteristics of resources Predicted output, resource time series characteristics It is a time-series sequence containing resource operation data from the past 24 hours; For gated loop units to access business characteristics Predicted output, business characteristics This is a business feature vector that includes the number of services launched during the same period and the proportion of product types. The business load impact coefficient has a value range of 0.1 to 0.5 and is obtained by training the correlation between business type and resource consumption. This represents the duration of the current peak business period, in hours. This represents the current idle period in hours. The predicted duration is in hours.

[0081] Specifically, this step involves achieving scientific prediction of resource demand through data fusion and algorithmic calculation: the selection of historical data should focus on the cyclical patterns of the same period, avoiding the use of data that differs significantly from the characteristics of the current business cycle, ensuring that historical data can reflect the resource demand characteristics of similar businesses within similar cycles; the design of the preset algorithm should integrate the real-time characteristics of resource operation with the historical demand characteristics of the business, while considering the impact of changes in business load on resource demand, and the key parameters in the algorithm should be able to adaptively adjust according to historical prediction errors to improve prediction accuracy; the calculation of peak resource demand should be combined with the duration and scale of future business operations to ensure that the peak can cover the maximum resource demand of the business during that period; the pre-scheduling instructions should clearly specify the resource type, the number of standby resources to be activated, the effective time, and the applicable businesses to ensure that the instructions are executable.

[0082] In step S4 above, the step of performing resource allocation and protocol binding according to the pre-scheduling instruction to complete the multi-product service resource configuration includes parsing the resource allocation priority and protocol binding rules contained in the pre-scheduling instruction. The resource allocation priority needs to be determined according to the importance level of the business, and the importance level of the business needs to be pre-stored in the business resource parameter library. The resource allocation is performed according to the determined priority order, reserving core resources with better performance and stronger stability for high-priority businesses, and allocating edge resources for low-priority businesses. At the same time, the detailed results of resource allocation are recorded. Based on the protocol type determined in the product-resource configuration file, the allocated resources are bound to the corresponding business protocol. During the binding process, the compatibility between the protocol and the resource needs to be verified. If there is an incompatibility problem, the matching correction mechanism of the business resource parameter library is triggered to re-determine the appropriate protocol. The resource allocation results and protocol binding results are integrated to generate a resource configuration completion report for the activation of multi-product services.

[0083] The process of monitoring resource configuration, identifying and handling anomalies, includes real-time collection of resource operation parameters during resource configuration, comparison with pre-stored standard resource operation parameters in the business resource parameter library, and calculation of the deviation between the two. When the deviation exceeds the anomaly judgment threshold pre-stored in the business resource parameter library, an anomaly handling instruction is automatically generated. Based on the anomaly type, operations such as switching the backup resource to a preset backup port or power-off isolation of the faulty resource are performed. At the same time, the anomaly type, handling instruction, and adapted resource information are updated to the anomaly knowledge base to provide a reference for subsequent handling of similar anomalies.

[0084] This step requires priority scheduling and real-time anomaly control to ensure the efficiency and stability of multi-product resource allocation: the priority mechanism for resource allocation must clearly define the criteria for distinguishing between core and edge resources, ensuring that high-priority services receive high-quality resource support while preventing low-priority services from consuming too many core resources; the compatibility verification of protocol binding must delve into the adaptation details of resources and protocols, and if incompatibility is found, a correction mechanism must be initiated in a timely manner to avoid configuration failure due to protocol issues; the resource configuration completion report must comprehensively reflect the configuration results, providing a basis for service activation and subsequent operation and maintenance; anomaly monitoring must be highly real-time, deviation calculation must accurately reflect the differences between resource operating status and standards, and the anomaly judgment threshold must be determined according to the service's requirements for resource stability; anomaly handling must respond quickly, the handling measures must effectively solve the anomaly problem, and the archiving of anomaly information must be standardized to ensure that the anomaly knowledge base can continuously accumulate handling experience.

[0085] Another embodiment of the present invention, such as Figure 2As shown in the figure, this embodiment provides a multi-product service activation resource configuration system, including a parameter library construction module, a data processing module, an information parsing module, a demand prediction module, a configuration execution module, and an exception handling module. The parameter library construction module is used to build a business resource parameter library and establish a mapping relationship between resources and multi-product service requirements. The data processing module is used to collect and process real-time resource operation data required for multi-product service activation and output valid resource data. The information parsing module is used to parse multi-product information input by the user, extract demand keywords, match them with the business resource parameter library, and generate product-resource configuration files. The demand prediction module is used to predict resource requirements and generate pre-scheduling instructions based on valid resource data and historical business data. The configuration execution module is used to execute resource allocation and protocol binding according to the pre-scheduling instructions to complete the multi-product service resource configuration. The exception handling module is used to monitor the resource configuration process, identify and handle exceptions, and update the exception knowledge base.

[0086] The information parsing module includes a keyword extraction unit, a library matching unit, and a file generation unit. The keyword extraction unit uses natural language processing algorithms to parse multi-product information and extract industry keywords and functional requirement keywords. The library matching unit matches the corresponding type sub-library based on industry keywords and then matches specific resource parameters in the sub-library based on functional requirement keywords. The file generation unit integrates resource models, protocol types, and resource requirements to generate standardized product-resource configuration files.

[0087] The information parsing module works collaboratively with three units to transform multi-product information into resource configuration files: the keyword extraction unit needs to accurately identify two types of keywords to ensure a clear distinction and close correlation between industry attributes and functional requirements; the library matching unit needs to first narrow down the resource range using industry keywords, and then lock in specific resources using functional requirement keywords to avoid inefficiency or insufficient accuracy caused by a matching range that is too large or too small; the file generation unit needs to integrate the matching results to ensure that the file information is complete, formatted correctly, and can directly support subsequent configuration operations. At the same time, the data flow between the three units needs to be smooth to avoid data stagnation or loss.

[0088] Furthermore, the demand forecasting module includes a data aggregation unit and a peak calculation unit. The data aggregation unit integrates valid resource data with historical business data from the same period of the past, with the preset quantity determined by the historical cyclical patterns of business activation. The peak calculation unit calculates the peak resource demand for the predicted duration using a preset algorithm, determining the number of standby resources to be activated. The demand forecasting module requires the collaboration of these two units to achieve scientific prediction of resource demand: the data aggregation unit must select comparable historical data and integrate it with real-time valid resource data to ensure that the data foundation reflects the actual characteristics of business and resources; the peak calculation unit must accurately calculate the peak demand using an algorithm, combine it with resource inventory to determine the number of standby resources to be activated, and the algorithm parameters must have adaptive adjustment capabilities to improve prediction accuracy. Simultaneously, both units must maintain real-time data transmission to avoid data lag affecting the prediction results.

[0089] The configuration execution module includes an instruction parsing unit, a resource allocation unit, a protocol binding unit, and a result generation unit. The instruction parsing unit parses the resource allocation priority and protocol binding rules in the pre-scheduled instructions, with priorities ordered according to business importance. The resource allocation unit performs resource allocation according to priority, reserving core resources for high-priority services and allocating edge resources for low-priority services. The protocol binding unit binds the allocated resources with the corresponding business protocols, verifying compatibility during the binding process and triggering parameter library matching and correction if incompatible. The result generation unit integrates the resource allocation and protocol binding results to generate a resource configuration completion report.

[0090] The configuration execution module requires the efficient implementation of resource configuration through the collaborative processes of four units: the instruction parsing unit must accurately interpret the key information in the pre-scheduled instructions to provide a clear basis for subsequent configuration; the resource allocation unit must strictly follow priority rules, reasonably divide core and edge resources, and ensure the rationality of resource allocation; the protocol binding unit must pay attention to compatibility verification, promptly handle incompatibility issues, and avoid configuration failures; and the result generation unit must comprehensively summarize the configuration results to provide a reference for business activation and operation and maintenance. At the same time, the four units must be closely connected to form a complete configuration process to avoid process interruption or repeated operations.

[0091] The anomaly handling module comprises a deviation monitoring unit and an instruction generation unit. The deviation monitoring unit compares the real-time operating parameters of resources with the pre-stored standard operating parameters in the business resource parameter library. The instruction generation unit generates instructions to switch backup resources to preset backup ports or to power-off isolate faulty resources when the deviation exceeds a pre-stored anomaly threshold in the business resource parameter library, and triggers an update to the anomaly knowledge base. The anomaly handling module relies on the collaboration of these two units to achieve real-time anomaly control and experience accumulation: the deviation monitoring unit collects and compares resource operating parameters in real-time, accurately calculates deviations, and ensures timely anomaly detection; the instruction generation unit generates effective processing instructions based on the deviation, quickly resolves anomalies, and updates the anomaly information to the knowledge base for future reference. The two units form a closed loop to ensure a complete process from anomaly detection to processing and archiving, improving system stability.

[0092] Based on the above, the system works as follows: First, the parameter library construction module builds a business resource parameter library, forming a mapping relationship between resources and the business requirements of multiple products. Then, the data processing module collects real-time resource operation data required for the activation of multiple product services, performs noise reduction and standardization on the collected data, and outputs valid resource data. Subsequently, the information parsing module uses a keyword extraction unit to parse the multi-product information input by the user using natural language processing algorithms, extracting industry keywords representing the business's domain and functional requirement keywords representing the core functions of the business. The library matching unit matches the corresponding type sub-library in the business resource parameter library based on the industry keywords, and then matches specific resource parameters in the sub-library based on the functional requirement keywords. The file generation unit integrates resource models, protocol types, and resource requirements to generate standardized product-resource configuration files. The demand prediction module integrates valid resource data with historical business data of the same period in the past (the preset quantity is determined by the historical cycle pattern of business activation) through the data aggregation unit. The peak calculation unit calculates the peak resource demand for the predicted duration using a preset algorithm, determines the number of standby resources to be activated, and generates pre-scheduling instructions. The configuration execution module parses the pre-scheduling instructions through the instruction parsing unit. The resource allocation priority and protocol binding rules in the instructions (priority is sorted according to the business importance level, which is pre-stored in the business resource parameter library) are implemented. The resource allocation unit executes resource allocation in order of priority, reserving core resources for high-priority businesses and allocating edge resources for low-priority businesses, and records the resource allocation results. The protocol binding unit binds the allocated resources with the corresponding business protocols based on the protocol types in the product-resource configuration file. During the binding process, the compatibility between the protocol and the resource is verified. If incompatibility occurs, parameter library matching and correction are triggered. The result generation unit integrates the resource allocation results and protocol binding results to generate a resource configuration completion report for the activation of multiple product services. Finally, the exception handling module compares the deviation of the real-time operating parameters of the resources with the standard operating parameters of the resources pre-stored in the business resource parameter library in real time through the deviation monitoring unit. When the deviation exceeds the exception judgment threshold pre-stored in the business resource parameter library, the instruction generation unit generates a processing instruction to switch the backup resource to the preset backup port or to disconnect and isolate the faulty resource. After executing the instruction, the exception type, processing instruction, and adapted resource information are updated to the exception knowledge base, forming a complete work closed loop of construction, processing, parsing, prediction, execution, and control, realizing accurate and efficient configuration of multi-product business resources.

[0093] Therefore, by adopting the above-mentioned multi-product business activation resource configuration method and system, the complex logic such as multi-product resource sharing relationship and configuration order control is integrated into the system. There is no need to predefine the fixed configuration process, which returns to the essence of resource and business carrying, greatly reduces the number of processes, and improves the accuracy of resource demand prediction and the efficiency of anomaly handling, making multi-product business activation simpler and easier to interact with.

[0094] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for allocating resources for activating multiple product services, characterized in that, Includes the following steps: S1. Construct a business resource parameter library to form a mapping relationship between resources and the business requirements of multiple products, collect and process real-time resource operation data required for the activation of multiple product services, and obtain effective resource data. S2. Parse the multi-product information input by the user, extract the required keywords and match them with the business resource parameter library to generate a product-resource configuration file; S3. Based on the effective resource data and historical service data, predict resource demand and generate pre-scheduling instructions; including: Summarize valid resource data with historical business data of the same period in the past, with the preset quantity determined by the historical periodic pattern of business activation. The peak resource demand for the predicted duration is calculated using a pre-set algorithm to determine the backup resource data that needs to be activated. The formula is as follows: ； in, For the future Peak resource demand at any given moment; The weights used for time series model fusion have a value range of 0 to 1 and are adaptively adjusted by backpropagation of historical prediction errors. For Long Short-Term Memory Networks to assess the temporal characteristics of resources The predicted output, resource time series characteristics It is a time-series sequence containing resource operation data from the past 24 hours; For gated loop units to access business characteristics Predicted output, business characteristics This is a business feature vector that includes the number of services launched during the same period and the proportion of product types. The business load impact coefficient has a value range of 0.1 to 0.5 and is obtained by training the correlation between business type and resource consumption. This represents the duration of the current peak business period, in hours. This represents the current idle period in hours. The predicted duration is in hours; S4. Execute resource allocation and protocol binding according to the pre-scheduled instructions to complete product business resource configuration. During the resource configuration monitoring process, identify and handle anomalies, and update the anomaly information to the anomaly knowledge base; including: The resource allocation priority and protocol binding rules in the pre-scheduling instruction are parsed. The priorities are sorted according to the importance level of the business, and the importance level is pre-stored in the business resource parameter library. Resource allocation is performed sequentially according to the aforementioned priority: core resources are reserved for high-priority services, and edge resources are allocated for low-priority services. The resource allocation results are then recorded. Based on the protocol type in the product-resource configuration file, the allocated resources are bound to the corresponding business protocol. During the binding process, the compatibility between the protocol and the resource is verified. If they are incompatible, the parameter library matching and correction is triggered, and the compatible protocol is re-determined. Integrate resource allocation results with protocol binding results to generate a resource configuration completion report for the activation of multiple product services; The process of identifying and handling anomalies during resource configuration monitoring includes: Real-time comparison of the deviation between the real-time operating parameters of resources and the standard operating parameters of resources pre-stored in the business resource parameter library; When the deviation exceeds the anomaly judgment threshold pre-stored in the business resource parameter library, an anomaly handling instruction is generated, and the operation of switching the backup resource to the preset backup port or power-off isolation of the faulty resource is performed. The anomaly type, handling instruction and adapted resource information are updated to the anomaly knowledge base.

2. The method for allocating resources for multi-product service activation according to claim 1, characterized in that, In S1, the construction of the business resource parameter library includes: Sub-libraries are divided according to resource function; Establish a mapping relationship between sub-library resources and business requirements, specifically: map bandwidth parameters according to business data transmission volume, map encryption protocol parameters according to business security level, and map computing power parameters according to business concurrency. The collection and processing of real-time resource operation data required for the activation of multiple product services includes: Collect real-time resource status information, including port traffic, CPU utilization, voltage stability, and memory usage. The collected real-time status information of the resources is denoised and standardized to obtain valid resource data. The format of the valid resource data matches the storage format of the business resource parameter card.

3. The method for allocating resources for multi-product service activation according to claim 1, characterized in that, In S2, parsing the multi-product information input by the user, extracting requirement keywords and matching them with the business resource parameter library to generate a product-resource configuration file includes: Natural language processing algorithms are used to parse multi-product information input by users and extract no less than two types of requirement keywords: industry keywords representing the business field and functional requirement keywords representing the core functions of the business. Based on the industry keywords, the corresponding type sub-library in the business resource parameter library is matched to obtain the range of suitable resource types; Within the scope of the resource types, specific resource parameters are matched based on the functional requirement keywords to determine the appropriate resource model, corresponding protocol type, and resource requirement. By integrating the resource models, protocol types, and resource requirements, a standardized product-resource configuration file is generated.

4. A resource allocation system for multi-product service activation, characterized in that, include: The parameter library construction module is used to build a business resource parameter library and establish a mapping relationship between resources and business requirements of multiple products. The data processing module is used to collect and process real-time resource operation data required for the activation of multiple product services, and output valid resource data. The information parsing module is used to parse multi-product information input by the user, extract requirement keywords and match them with the business resource parameter library to generate product-resource configuration profiles; The demand forecasting module is used to predict resource demand and generate pre-scheduling instructions based on available resource data and historical business data. The demand forecasting module includes: The data aggregation unit is used to integrate valid resource data with a preset quantity of historical business data from the same period in the past, the preset quantity being determined by the historical periodic patterns of business activation. The peak demand calculation unit is used to calculate the peak resource demand for the predicted duration using a preset algorithm, and to determine the number of standby resources to be activated. The formula is as follows: ； in, For the future Peak resource demand at any given moment; The weights used for time series model fusion have a value range of 0 to 1 and are adaptively adjusted by backpropagation of historical prediction errors. For Long Short-Term Memory Networks to assess the temporal characteristics of resources The predicted output, resource time series characteristics It is a time-series sequence containing resource operation data from the past 24 hours; For gated loop units to access business characteristics Predicted output, business characteristics This is a business feature vector that includes the number of services launched during the same period and the proportion of product types. The business load impact coefficient has a value range of 0.1 to 0.5 and is obtained by training the correlation between business type and resource consumption. This represents the duration of the current peak business period, in hours. This represents the current idle period in hours. The predicted duration is in hours; The configuration execution module is used to perform resource allocation and protocol binding according to pre-scheduled instructions, and to complete the configuration of multi-product service resources; the configuration execution module includes: The instruction parsing unit is used to parse the resource allocation priority and protocol binding rules in the pre-scheduling instruction, wherein the priority is sorted according to the level of business importance; The resource allocation unit is used to perform resource allocation according to priority, reserving core resources for high-priority services and allocating peripheral resources for low-priority services; The protocol binding unit is used to bind the allocated resources with the corresponding business protocol. During the binding process, compatibility is verified, and parameter library matching and correction are triggered when incompatibility occurs. The result generation unit is used to integrate resource allocation and protocol binding results to generate a resource configuration completion report. An exception handling module is used to monitor the resource configuration process, identify and handle exceptions, and update the exception knowledge base; the exception handling module includes: The deviation monitoring unit is used to compare the deviation between the real-time operating parameters of resources and the pre-stored standard operating parameters of resources in the business resource parameter library. The instruction generation unit is used to generate a processing instruction for switching the backup resource to a preset backup port or for power-off isolation of the faulty resource when the deviation exceeds the anomaly judgment threshold pre-stored in the service resource parameter library, and to trigger an update of the anomaly knowledge base.

5. A multi-product service activation resource allocation system according to claim 4, characterized in that, The information parsing module includes: The keyword extraction unit is used to parse multi-product information using natural language processing algorithms and extract industry keywords and functional requirement keywords. The library matching unit is used to match the corresponding type sub-library based on industry keywords, and then match specific resource parameters in the sub-library based on functional requirement keywords; The file generation unit is used to integrate resource models, protocol types, and resource requirements to generate standardized product-resource configuration files.

Images