Data center distribution method and device and electronic equipment

Abstract

The embodiment of the invention provides a data center distribution method, a data center distribution device and electronic equipment. The method comprises the following steps: receiving and analyzing an opening request of an Internet of Things tenant, and obtaining service product information and / or initial service scene classification information; based on the service product information and / or the initial service scene classification information, according to a preset service scene determination rule, obtaining target service scene classification information corresponding to the opening request; matching in a data center distribution rule set according to the target business scene classification information to obtain an initial data center code; according to the tenant information of the Internet of Things tenants and the initial data center code, retrieving a tenant distribution record table, and obtaining a retrieval result corresponding to the target business scene classification information; and determining a target data center code according to a retrieval result, and performing data center distribution based on the target data center code. According to the invention, the accuracy and stability of Internet of Things tenant data center distribution and the overall operation and maintenance efficiency are improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a data center allocation method, allocation device and electronic device. Background Technology

[0002] IoT platforms serving multiple enterprises and tenants typically employ a distributed deployment model across multiple locations and data centers. Different data centers vary in hardware performance, network conditions, redundancy, and security capabilities, resulting in different service levels and resource caps. For each new tenant, the appropriate selection of a data center to support their business not only determines whether they receive service quality assurance commensurate with the importance of their business, but also impacts load balancing, resource utilization, and the efficiency of subsequent operation, maintenance, and expansion across data centers.

[0003] In existing technologies, IoT platforms typically rely on manual methods for data center allocation when activating services for tenants. Specifically, front-end business personnel fill in information such as the tenant's name, industry, and main business or product description in an activation ticket. Then, back-end operators review the ticket content and, based on their experience or internally maintained rule tables, manually assess the tenant's industry type, application scenario, and business importance, manually selecting one or a few data centers as the location for that tenant's services. In practice, because the information entered in the ticket may be incomplete, incorrect, or non-standard, operators often need to communicate additionally with business personnel or customers via phone, email, etc., for confirmation. When information is incomplete, to improve processing speed, operators sometimes directly allocate a large number of tenants to a single fixed data center according to default rules.

[0004] It is evident that when work order information is incomplete or non-standard, relying solely on manual reading and subjective judgment can easily lead to mismatches between the allocation results and the tenant's business, making it difficult to guarantee the accuracy of the allocation. Summary of the Invention

[0005] This application provides a data center allocation method, allocation device, and electronic device to improve the accuracy of data center allocation.

[0006] In a first aspect, embodiments of this application provide a data center allocation method, including:

[0007] Receive and parse the activation request from the IoT tenant, and when it is determined that the activation request carries business product information and / or initial business scenario classification information, obtain the business product information and / or initial business scenario classification information;

[0008] Based on the business product information and / or initial business scenario classification information, and according to the preset business scenario determination rules, the target business scenario classification information corresponding to the activation request is obtained;

[0009] Based on the target business scenario classification information, a matching process is performed in the data center allocation rule set to obtain the initial data center code corresponding to the target business scenario classification information;

[0010] Obtain the tenant information of the IoT tenant, and based on the tenant information and the initial data center code, retrieve the tenant allocation record table to obtain the retrieval results corresponding to the target business scenario classification information;

[0011] Based on the search results, the target data center code corresponding to the target business scenario classification information is determined, and the activation request is allocated a data center based on the target data center code.

[0012] In one possible implementation, determining that the activation request carries service product information includes:

[0013] Obtain the product description information carried in the activation request, and preprocess the product description information to obtain the processed product description text;

[0014] Based on a preset vocabulary library, the processed product description text is subjected to synonym replacement processing to obtain standard description text, and the standard description text is subjected to word segmentation processing to obtain word segmentation results.

[0015] The vector similarity is calculated between the word segmentation results and each standard word in the preset standard product list, and a set of candidate product names matching the word segmentation results is determined based on the similarity results, which serves as the list of products to be activated.

[0016] If no candidate product name is found in the list of products to be activated, it is determined that the activation request does not contain business product information.

[0017] If there is at least one candidate product name in the list of products to be activated, then it is determined that the activation request carries business product information.

[0018] In one possible implementation, the step of obtaining the target business scenario classification information corresponding to the activation request based on the initial business scenario classification information and according to a preset business scenario determination rule includes:

[0019] If it is determined that the activation request does not carry business product information, then it is determined whether the activation request carries initial business scenario classification information.

[0020] If the activation request carries initial business scenario classification information, then the initial business scenario classification information is recorded as the target business scenario classification information corresponding to the activation request.

[0021] In one possible implementation, the step of obtaining the target business scenario classification information corresponding to the activation request based on the business product information and according to preset business scenario determination rules includes:

[0022] Based on the list of activated products, the business scenario category records corresponding to each candidate product name are searched in the preset product information table, and deduplication is performed to obtain a candidate set of scenario categories.

[0023] When the scenario classification candidate set contains multiple different business scenario classification information, each business scenario classification information is respectively confirmed as multiple target business scenario classification information corresponding to the activation request;

[0024] Alternatively, if the scenario classification candidate set contains single business scenario classification information, the business scenario classification information will be confirmed as the target business scenario classification information corresponding to the activation request.

[0025] In one possible implementation, it also includes:

[0026] When it is determined that the activation request does not carry business product information and does not carry initial business scenario classification information, the tenant information of the IoT tenant is obtained;

[0027] Based on the tenant information, the publicly available text information of the tenant corresponding to the tenant information is obtained by calling an external public information service;

[0028] The publicly available text information of the tenant is preprocessed and segmented to obtain multiple text units, and entity information corresponding to the business scenario is identified and obtained based on the multiple text units.

[0029] Based on the entity information, a matching process is performed in the business scenario classification knowledge graph to obtain multiple candidate business scenario classification information and the confidence level corresponding to each candidate business scenario classification information.

[0030] Based on the confidence level corresponding to each candidate business scenario classification information, filter and obtain candidate business scenario classification information with a confidence level greater than or equal to a preset threshold to obtain a scenario classification candidate set;

[0031] When the scenario classification candidate set contains multiple different business scenario classification information, each business scenario classification information is respectively confirmed as multiple target business scenario classification information corresponding to the activation request;

[0032] Alternatively, if the scenario classification candidate set contains single business scenario classification information, the business scenario classification information will be confirmed as the target business scenario classification information corresponding to the activation request.

[0033] In one possible implementation, the data center allocation rule set includes multiple data center allocation rules, which include business scenario classification conditions, data center codes, and priority information.

[0034] Accordingly, the step of matching the target business scenario classification information in the data center allocation rule set to obtain the initial data center code corresponding to the target business scenario classification information includes:

[0035] Read the target business scenario classification information to obtain the industry affiliation code, sub-scenario code, and product classification code from the target business scenario classification information;

[0036] Obtain the allocation rules for each data center, and obtain the business scenario classification conditions, data center codes, and priority information corresponding to the data center allocation rules;

[0037] Determine whether the industry affiliation, sub-scenario, and product classification in each business scenario classification condition match the industry affiliation code, sub-scenario code, and product classification code, respectively, and add the matching data center allocation rules to the candidate set;

[0038] From the candidate set, the data center allocation rule with the highest priority is selected according to the priority information, and the data center code corresponding to the data center allocation rule with the highest priority is determined as the initial data center code corresponding to the target business scenario classification information.

[0039] In one possible implementation, the step of retrieving the tenant allocation record table based on the tenant information and the initial data center code to obtain the retrieval results corresponding to the target business scenario classification information includes:

[0040] Based on the preset tenant allocation record table, according to the initial data center code, the tenant allocation records with the same initial data center code are filtered out to obtain a candidate tenant allocation record set;

[0041] The tenant information and the historical allocation records in the candidate tenant allocation record set are preprocessed to obtain the tenant information text and the historical tenant information text.

[0042] Based on preset semantic matching rules, the similarity between the tenant information text and the information text of each historical tenant is calculated;

[0043] If there is historical tenant information text with a similarity greater than or equal to a preset similarity threshold, the search result will be recorded as having a corresponding tenant allocation record.

[0044] If no historical tenant information text with a similarity greater than the preset similarity threshold exists, the search result will be recorded as no corresponding tenant allocation record exists.

[0045] In one possible implementation, determining the target data center code corresponding to the target business scenario classification information based on the search results includes:

[0046] If the search result indicates the existence of a corresponding tenant allocation record, then the data center code is read from the tenant allocation record, and the read data center code is used as the target data center code corresponding to the target business scenario classification information;

[0047] If the search result indicates that there is no corresponding tenant allocation record, then the initial data center code will be used as the target data center code corresponding to the target business scenario classification information;

[0048] Accordingly, the process of allocating a data center for the activation request based on the target data center code includes:

[0049] Based on the target data center code, the activation request is sent to the corresponding data center to complete the service activation by calling the data center activation interface corresponding to the target data center code;

[0050] After the service is activated, the tenant information, the target business scenario classification information, and the target data center code are written into the tenant allocation record table.

[0051] Secondly, embodiments of this application provide a data center allocation device, comprising:

[0052] The request receiving module is used to receive and parse the activation request of the IoT tenant, and when it is determined that the activation request carries business product information and / or initial business scenario classification information, the module obtains the business product information and / or initial business scenario classification information.

[0053] The data processing module is used to obtain the target business scenario classification information corresponding to the activation request based on the business product information and / or the initial business scenario classification information, according to the preset business scenario determination rules.

[0054] The data processing module is also used to match the target business scenario classification information in the data center allocation rule set to obtain the initial data center code corresponding to the target business scenario classification information.

[0055] The information retrieval module is used to obtain the tenant information of the IoT tenant, and retrieve the tenant allocation record table based on the tenant information and the initial data center code to obtain the retrieval results corresponding to the target business scenario classification information;

[0056] The central allocation module is used to determine the target data center code corresponding to the target business scenario classification information based on the search results, and to allocate a data center to the activation request based on the target data center code.

[0057] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;

[0058] The memory stores computer-executed instructions;

[0059] The processor executes computer execution instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0060] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0061] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0062] This application provides a data center allocation method, allocation device, and electronic device. By receiving and parsing IoT tenant activation requests during the IoT tenant service activation phase, and acquiring business product information and / or initial business scenario classification information when the activation request is determined to carry such information, the identification of business scenarios is based on the native data of the activation request. By obtaining the target business scenario classification information corresponding to the activation request based on business product information and / or initial business scenario classification information, and relying on preset business scenario determination rules, this application achieves automated and standardized processing of business scenario classification, which is beneficial for maintaining consistency and reusability of business scenario determination in large-scale tenant access scenarios. Next, based on the target business scenario classification information, matching is performed in the data center allocation rule set to obtain the initial data center code corresponding to the target business scenario classification information. By establishing a correspondence between business scenarios and the data center allocation rule set, different data centers can be selected for tenant services belonging to different industries and with different application scenarios, thereby supporting differentiated service levels and resource capacity configurations at the system level. Building upon this foundation, tenant information for IoT tenants is further acquired. Based on this tenant information and the initial data center code, a search is performed in the tenant allocation record table to obtain the corresponding search results. The target data center code corresponding to the target business scenario classification information is then determined accordingly. By incorporating existing historical allocation information from the tenant allocation record table into the decision-making process, the existing carrying relationships of the same tenant at the data center level are considered while adhering to business scenario allocation rules. Finally, data center allocation is performed on activation requests based on the target data center code. This ensures that the allocation results reflect the match between the business scenario and data center capabilities, maintain the continuity of tenant's historical deployments, reduce human error and additional migration costs, and improve the accuracy, stability, and overall operational efficiency of IoT tenant data center allocation. Attached Figure Description

[0063] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0064] Figure 1 A flowchart illustrating the data center allocation method provided in this application embodiment;

[0065] Figure 2 This is a flowchart illustrating a method for obtaining target business scenario classification information based on tenant information, as provided in an embodiment of this application.

[0066] Figure 3 This is a schematic diagram of the activation request processing provided in an embodiment of this application;

[0067] Figure 4This is a schematic diagram of the structure of a data center allocation device provided in an embodiment of this application;

[0068] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0069] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0070] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0071] The inventive concept of this application lies in reconstructing the data center allocation process for IoT tenants from a decentralized process relying on manual experience to a unified automated decision-making process based on business information and historical allocation information. Specifically, during the IoT tenant service activation phase, the system receives and parses the activation request from the IoT tenant. When it is determined that the activation request carries business product information and / or initial business scenario classification information, this information is obtained. By setting preset business scenario determination rules, using business product information and / or initial business scenario classification information as input, the system uniformly determines the target business scenario classification information corresponding to the activation request. This ensures that business information from different sources and with different filling habits is unified into a standardized business scenario classification system within the platform, providing a foundation for subsequent allocation.

[0072] After obtaining the target business scenario classification information, a pre-configured data center allocation rule set is used to match the target business scenario classification information within the data center allocation rule set, resulting in an initial data center code corresponding to that business scenario. This gives the selection of data centers a clear scenario-driven characteristic, allowing different business scenarios to be automatically mapped to different data center carrying strategies. Building on this, historical information from the tenant dimension is further introduced. By acquiring the tenant information of IoT tenants and retrieving the tenant allocation record table based on the tenant information and the initial data center code, the corresponding search results are obtained, reflecting whether the tenant already has an existing allocation relationship within the candidate data center range.

[0073] Based on the search results, the target data center code corresponding to the classification information of the target business scenario is determined. When an existing allocation record corresponding to the current tenant exists, the historical data center carrying relationship is used first. When no existing record exists, the initial data center code matching the business scenario is used as the target data center code. Finally, the data center is allocated to the activation request based on the target data center code. Through the above overall concept, this application ensures the matching of allocation strategy with business characteristics at the business scenario level, maintains the consistency and continuity of data center carrying at the tenant level, reduces manual intervention and operational errors, and improves the accuracy and controllability of data center allocation for IoT tenants.

[0074] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0075] Figure 1 A flowchart illustrating the data center allocation method provided in this application embodiment is shown below. Figure 1 As shown, the method includes:

[0076] S11, receive and parse the activation request from the IoT tenant, and when it is determined that the activation request carries business product information and / or initial business scenario classification information, obtain the business product information and / or initial business scenario classification information.

[0077] In this embodiment, the activation request from an IoT tenant represents the tenant's need to activate services on the IoT platform. This activation request is typically generated by the upstream business system and sent to the IoT platform. The activation request from an IoT tenant can take the form of structured fields and / or text descriptions, including at least tenant identification information and descriptions related to the tenant's business. Receiving the activation request from an IoT tenant provides a unified input entry point for the subsequent data center allocation process, allowing data center allocation to be processed directly based on the original business data in the activation process. Next, the activation request is parsed to extract content related to data center allocation. The parsing process may include structured parsing of each field in the activation request, as well as format recognition and field classification of the text content, clarifying which information describes the specific products activated by the tenant and which information represents the tenant's self-reported business scenario. By parsing the activation request, various business information originally mixed in the same request can be divided into different categories of data, providing a clear data foundation for subsequent classification judgment and rule matching.

[0078] After parsing the various fields in the activation request, it is necessary to determine whether the activation request carries business product information and / or initial business scenario classification information. Business product information describes the products or services that the IoT tenant plans to activate on the platform, such as terminal type, access service category, and application type. Initial business scenario classification information indicates the business scenario category selected or filled in by the IoT tenant during the service activation phase, such as industry category, application scenario type, and product classification identifier. By checking the existence and completeness of relevant fields in the activation request, it is determined whether the activation request currently has the basis for business scenario determination. If it is determined that the activation request carries business product information and / or initial business scenario classification information, the corresponding business product information and / or initial business scenario classification information are obtained from the parsing results and used as input data for subsequent business scenario determination. This improves the automation level of the data center allocation process and the accuracy of preconditions.

[0079] S12, based on business product information and / or initial business scenario classification information, and according to preset business scenario determination rules, obtain the target business scenario classification information corresponding to the activation request.

[0080] In this embodiment, the pre-defined business scenario determination rules are used to standardize the correspondence between the aforementioned business product information, initial business scenario classification information, and the business scenario classification system adopted internally by the platform. These rules can be configured during the deployment phase based on industry standards, the platform's product system, and historical operational experience, and can be maintained and adjusted according to business development. The business scenario determination rules may include rules for deriving business scenarios based on business product information, rules for accepting or verifying information based on initial business scenario classification information, and rules for conflict handling and priority handling when two types of information coexist. By pre-configuring the business scenario determination rules, the judgment process that originally relied on human experience can be transformed into automatically executable logic.

[0081] Based on preset business scenario determination rules, the target business scenario classification information corresponding to the activation request is obtained from business product information and / or initial business scenario classification information. This ensures that each activation request can be mapped to the platform's internal standard business scenario classification system. On the one hand, this reduces the need for manual interpretation and subjective judgment of each activation request, improving the automation and consistency of business scenario determination. On the other hand, it provides standardized input for subsequent matching of data center allocation rules based on the target business scenario classification information, which helps ensure the correspondence between data center allocation results and actual business scenarios, supporting improvements in the accuracy and controllability of the overall data center allocation method.

[0082] S13. Based on the target business scenario classification information, match it in the data center allocation rule set to obtain the initial data center code corresponding to the target business scenario classification information.

[0083] In this embodiment, the data center allocation rule set is used to describe the correspondence between business scenarios and data centers. The data center allocation rule set can consist of multiple rule entries, each rule entry associating one or more types of business scenarios with one or more available data centers. It can also be configured in conjunction with factors such as the resource capabilities, service levels, geographical distribution, and redundancy capabilities of different data centers. By pre-configuring the data center allocation rule set during the deployment or operation and maintenance phases, the data center allocation process becomes reusable and maintainable.

[0084] After obtaining the target business scenario classification information, this classification information is used as input to perform a matching process within the data center allocation rule set. Matching refers to identifying one or more rule entries in the rule set that are applicable to the current target business scenario classification information, based on pre-defined matching conditions and judgment logic. Matching conditions can be based on the business type, application domain, and scenario characteristics represented by the target business scenario classification information. When the scope of application of a rule entry covers the target business scenario classification information, that rule entry is considered to match the current business scenario. By performing matching within the rule set based on the target business scenario classification information, an automatic association between business scenarios and data center allocation strategies is achieved.

[0085] After identifying the rule entries that match the target business scenario classification information, the initial data center code for the current activation request is determined from the matched rule entries according to a preset selection strategy. The selection strategy can comprehensively consider information such as the data center attributes configured in the rule entries, business carrying strategies, and system operating status to make a reasonable selection among multiple candidate data centers. When only one matching rule exists, the data center identifier corresponding to that rule can be directly used as the initial data center code; when multiple matching rules exist, the data center identifier corresponding to one or a preferred rule can be selected as the initial data center code according to a pre-defined selection logic.

[0086] By matching target business scenario classification information with the data center allocation rule set and obtaining initial data center codes, an automatic data center recommendation mechanism driven by business scenarios is realized. This eliminates reliance on single manual judgment in data center selection, instead ensuring unified execution based on pre-configured allocation rules. This allows different business scenarios to obtain data center hosting solutions tailored to their characteristics, and provides a structured foundation for subsequent data center load balancing, service level assurance, and resource planning, thereby improving the controllability and consistency of the overall data center allocation method.

[0087] S14: Obtain the tenant information of the IoT tenant, and retrieve the tenant allocation record table based on the tenant information and the initial data center code to obtain the retrieval results corresponding to the target business scenario classification information.

[0088] In this embodiment, the tenant information of the IoT tenant is used to identify a specific tenant entity and is the basic data for distinguishing different tenants and establishing relationships between different service activation requests. Tenant information may include identification data for uniquely identifying the tenant and descriptive data for auxiliary identification; the specific content can be configured according to business needs. By obtaining the tenant information of the IoT tenant, the subsequent data center allocation process can be handled at the tenant level, rather than just at a single activation request, which is beneficial for establishing continuity between multiple service activations of the same tenant. The tenant allocation record table records the historical allocation of IoT tenants in various data centers and is a data set reflecting historical allocation relationships. This record table stores allocation information related to existing service activations, indicating whether a tenant has already established a service bearing relationship in a certain data center. By maintaining the tenant allocation record table, historical allocation information can be queried when a new activation request arrives, enabling data center allocation to comprehensively consider historical bearing relationships and current allocation strategies.

[0089] Based on tenant information and the initial data center code, a search operation is initiated in the tenant allocation record table. The search process can, according to preset search conditions and judgment rules, look for existing records related to the current tenant information and initial data center code in the tenant allocation record table, forming search results corresponding to the target business scenario classification information. These search results can be used to reflect whether there are historical allocation relationships related to the current tenant within the data center range represented by the current initial data center code, providing a basis for subsequently determining the target data center code. By obtaining the tenant information of IoT tenants and searching the tenant allocation record table based on the tenant information and initial data center code, search results corresponding to the target business scenario classification information are obtained, enabling the data center allocation process to make judgments in conjunction with historical allocation information. This can reduce disorderly new allocations when existing allocation relationships can be reused, and can also provide a decision-making basis for new allocations when no historical records exist. While ensuring that the allocation results are consistent with historical carrying relationships, it improves overall resource utilization efficiency and the continuity of data center allocation.

[0090] S15. Based on the search results, determine the target data center code corresponding to the target business scenario classification information, and allocate a data center for the activation request based on the target data center code.

[0091] In this embodiment, the search results can indicate whether a historical allocation relationship corresponding to the current tenant information already exists in the data centers related to the initial data center code, or the association status between multiple possible data centers. By generating search results in the aforementioned steps, historical information at the tenant level can be introduced into the current data center selection process, so that the determination of the target data center no longer relies solely on static allocation rules, but can also take into account the factor of historical carrying relationships. Based on the search results and the initial data center code corresponding to the target business scenario classification information, the target data center code is determined. In this process, information such as the historical allocation situation reflected in the search results, the allocation strategy of the current business scenario, and the system operating status can be comprehensively considered to compare and select different candidate data centers to form the final data center identifier suitable for the current activation request. By introducing search results when determining the target data center code, the continuity and consistency of the same tenant at the data center level can be maintained as much as possible while following the business scenario allocation strategy.

[0092] Next, the activation request is allocated to a data center based on the target data center code. Specifically, the target data center code indicates the target data center to carry the current activation request. This code is used as a routing or control parameter to submit the activation request and its associated service configuration information to the resource management system corresponding to the target data center, which then completes the corresponding resource configuration and service activation operations within the target data center. By allocating based on the target data center code, this implementation ensures that the activation request is accurately routed to the expected data center range at the physical or logical level. By determining the target data center code corresponding to the target business scenario classification information based on the search results, and allocating the activation request to a data center based on the target data center code, the data center selection process simultaneously considers the business scenario allocation strategy and the tenant's historical allocation information. This improves the overall resource utilization efficiency and service assurance capabilities of the IoT platform.

[0093] In one embodiment, the determination that the activation request carries service product information in step S11 above will be further explained here, based on the above embodiment, including:

[0094] S111: Obtain the product description information carried in the activation request, and preprocess the product description information to obtain the processed product description text.

[0095] Specifically, preprocessing can include removing redundant spaces and line breaks, standardizing the use of Chinese and English symbols, regulating capitalization, and cleaning up obviously irrelevant punctuation or control characters. This transforms the original description text from a state of messy formatting and inconsistent writing into a more structurally standardized text input. By preprocessing the product description information, the interference of noisy characters and differences in writing habits on subsequent analysis results can be reduced. This provides a more stable base text for subsequent synonym replacement and word segmentation, enabling the system to obtain comparable and consistent input when faced with diverse descriptions filled in by different business personnel.

[0096] S112, based on the preset word library, perform synonym replacement processing on the processed product description text to obtain standard description text, and perform word segmentation processing on the standard description text to obtain the word segmentation results.

[0097] Specifically, the pre-defined terminology library can contain standard terms related to IoT products, along with their common synonyms, abbreviations, or colloquial names. This is used to uniformly map different expressions used by business personnel in activation requests to standard terms agreed upon within the platform. Through synonym replacement, different expressions such as "vehicle positioning terminal," "vehicle GPS," and "vehicle location device" are unified into consistent standard terms, thereby reducing the dispersion of subsequent matching results due to differences in terminology. Subsequently, the standard description text is segmented, dividing continuous text into multiple words with independent meanings to form segmentation results. This transforms the product description from a whole paragraph of natural language into a sequence of standard terms, providing basic features for subsequent vector similarity calculations.

[0098] S113. Calculate the vector similarity between the word segmentation results and each standard word in the preset standard product list, and determine the set of candidate product names that match the word segmentation results based on the similarity results, which will serve as the product list to be activated.

[0099] Specifically, the pre-defined standard product list can be a set of products uniformly maintained by the platform, with each product name associated with a set of standard words or text feature vectors. The word segmentation results are converted into vector form and their similarity is calculated with each standard word or corresponding feature vector in the standard product list to obtain the similarity value between the word segmentation results and the words associated with each standard product. Based on the similarity results, a group of product names with a high degree of matching with the word segmentation results is selected, and these product names are grouped into a candidate product name set, which serves as the list of products to be activated for the current activation request. Through vector similarity calculation, the free text form of product descriptions is automatically mapped to a structured set of standard product names, realizing the automatic conversion from raw natural language descriptions to standard product identifiers.

[0100] S114. If no candidate product name is found in the list of products to be activated, it is determined that the activation request does not carry business product information.

[0101] S115. If there is at least one candidate product name in the product activation list, it is determined that the activation request carries business product information.

[0102] After obtaining the list of products to be activated, the system determines whether the activation request carries business product information based on whether the list contains candidate product names. If no candidate product name exists in the list, it means that no product name corresponding to the standard terminology library has been identified in the current product description information, and the activation request does not contain business product information. If at least one candidate product name exists in the list, it means that at least one usable product name has been identified in the product description information, and the activation request contains business product information. By using the emptiness of the product list as a criterion, the system automatically identifies the existence of business product information. Even when activation request fields are not standardized or the expression methods are inconsistent, the system automatically determines whether there is product information that can be used for subsequent business scenario determination, thereby reducing the workload of manual order-by-order checks and improving the quality of prerequisite information for subsequent business scenario determination and data center allocation.

[0103] In another specific embodiment, after preprocessing and standardizing the product description information, a keyword fuzzy matching method can be used to match it with the product category list to obtain the product list to be activated. Specifically, several keywords can be extracted from the standardized description text. These keywords may include product function-related terms, terminal type-related terms, and application scenario-related terms. Then, these keywords are fuzzily matched with the key identifiers corresponding to each product category in the product category list. For example, by using partial string matching, edit distance matching, or inclusion relationship matching, the matching degree of each product category at the keyword level is statistically analyzed. Based on whether the matching degree meets the preset matching conditions, the set of product names corresponding to one or more product categories that meet the conditions is selected as the product list to be activated for the current activation request.

[0104] This implementation method uses keyword fuzzy matching to match with the product category list. This allows for the identification of associated product sets even when product descriptions are not fully standardized or contain varying terminology. By performing fuzzy matching of keywords in the description text with the product category list, the method improves adaptability to diverse human descriptions, enhances the robustness of the product list acquisition process, and provides a more flexible implementation path for subsequently determining business scenario classification information based on the product list.

[0105] In one embodiment, step S12 is further described below. Based on the initial business scenario classification information, and according to preset business scenario determination rules, the target business scenario classification information corresponding to the activation request is obtained, including:

[0106] S1211, when it is determined that the activation request does not carry business product information, determine whether the activation request carries initial business scenario classification information;

[0107] S1212 If the activation request carries initial business scenario classification information, then the initial business scenario classification information is recorded as the target business scenario classification information corresponding to the activation request.

[0108] In this embodiment, when it is determined that the activation request does not carry business product information, the business scenario analysis no longer relies on business product information. Instead, it proceeds to a judgment based on the initial business scenario classification information. Therefore, it is first necessary to determine whether the activation request carries initial business scenario classification information. Initial business scenario classification information is typically selected or filled in directly by front-end business personnel during the form-filling stage. It represents the expected scenario affiliation of the IoT tenant's business within the platform, such as industry category, application scenario type, and product direction. By detecting whether the fields corresponding to the initial business scenario classification information exist in the activation request, whether they are filled in, and whether they meet the basic format requirements, it is determined whether the current activation request has directly usable scenario classification input. By judging the initial business scenario classification information under the condition of missing business product information, an ordered judgment priority is established among different information sources.

[0109] When the determination result indicates that the activation request carries initial business scenario classification information, this initial business scenario classification information is recorded as the target business scenario classification information corresponding to the activation request. In this scenario, the initial business scenario classification information, as the scenario attribution directly provided by business personnel, is considered a direct expression of the current activation request in the business scenario dimension. This initial business scenario classification information can be standardized according to the requirements of the internal business scenario classification system, and the standardized result can be used as the target business scenario classification information for subsequent data center allocation rule matching. By directly using the already filled initial business scenario classification information to determine the target business scenario classification information, even when business product information is lacking but scenario classification information is complete, a clear business scenario basis can still be provided for the current activation request.

[0110] In one embodiment, step S12 is further described below. Based on business product information and according to preset business scenario determination rules, the target business scenario classification information corresponding to the activation request is obtained, including:

[0111] S1221. Based on the list of activated products, search for the business scenario category records corresponding to each candidate product name in the preset product information table, and perform deduplication to obtain a candidate set of scenario categories.

[0112] In this embodiment, the product information table can be pre-configured to use the product name or an equivalent identifier as the index, and each record stores a business scenario classification record associated with the product, representing the product's classification relationship in dimensions such as industry affiliation, application scenario, and product category. Based on the candidate product names in the list of activated products, the corresponding business scenario classification record is searched one by one in the product information table. The retrieved business scenario classification information is summarized, and duplicate business scenario classification information is deduplicated to form a candidate set of scenario classifications without duplicates. By performing scenario mapping and deduplication on the list of activated products based on the product information table, an automatic conversion from a standardized product list to an internal business scenario classification set is achieved, ensuring the uniqueness and standardization of the business scenario classification information used in subsequent processing.

[0113] S1222, when the scenario classification candidate set contains multiple different business scenario classification information, each business scenario classification information is confirmed as multiple target business scenario classification information corresponding to the activation request.

[0114] Specifically, when the scenario classification candidate set contains multiple different business scenario classification information, it indicates that the product involved in the current activation request corresponds to multiple different business scenario types. In this case, each business scenario classification information in the scenario classification candidate set is identified as the multiple target business scenario classification information corresponding to the activation request. In this way, different business scenarios involved in the same activation request can be distinguished, and each type of business scenario classification information is retained and identified. By identifying multiple business scenario classification information as target business scenario classification information, fine-grained expression of multi-business scenario activation requests is achieved, which is beneficial for subsequent resource allocation and strategy execution for different business scenarios.

[0115] S1223, when the scenario classification candidate set contains single business scenario classification information, the single business scenario classification information is confirmed as the target business scenario classification information corresponding to the activation request.

[0116] Specifically, when the scenario classification candidate set contains only a single business scenario classification information, it indicates that the product types corresponding to the product activation list have a consistent affiliation in terms of business scenario dimension. In this case, the business scenario classification information can be directly confirmed as the target business scenario classification information corresponding to the activation request. By determining the target business scenario classification information in both multi-scenario and single-scenario scenarios based on the number of business scenario classification information in the scenario classification candidate set, unified processing of business scenarios with different complexities is achieved. This ensures the consistency and completeness between the target business scenario classification information and the product activation list, providing an accurate business scenario foundation for the subsequent data center allocation process.

[0117] In one optional implementation, when the scenario classification candidate set contains only a single business scenario classification information, a consistency check can be performed between the judgment result and the business scenario classification information filled in the activation request. Specifically, when the activation request carries initial business scenario classification information, the business scenario classification information obtained from the activation product list is compared with the initial business scenario classification information: if they are the same, it is considered that the front-end entry and automatic judgment are consistent, and no manual intervention is required; if they are inconsistent, it is considered that the judged scenario and the requested scenario conflict, and a manual review process can be triggered. It should be noted that manual review can be performed, or the method described in the previous embodiments can be used.

[0118] In the manual review process, the current activation request, the corresponding product description information, the parsed list of activated products, the business scenario classification information obtained from the product list, and the business scenario classification information filled in the activation request are all presented to the business operations personnel or reviewers. They then make a comprehensive judgment based on the tenant's actual business situation, the contractually agreed products, and internal business rules, selecting from two conflicting business scenario classifications or correcting the existing classifications. The verified business scenario classification information is recorded as the target business scenario classification information for the activation request and used for subsequent data center allocation processing. Human-machine collaboration is only introduced when there is a significant discrepancy between the automatic identification results and the front-end data entry, maintaining the overall automation efficiency of the process.

[0119] Figure 2 This is a flowchart illustrating a method for obtaining target business scenario classification information based on tenant information, provided in an embodiment of this application. The method for obtaining target business scenario classification information is explained here when the activation request does not carry business product information or initial business scenario classification information. Based on the above embodiments, as... Figure 2 As shown, it includes:

[0120] S21. When it is determined that the activation request does not carry business product information and does not carry initial business scenario classification information, obtain the tenant information of the IoT tenant.

[0121] Specifically, when it is determined that the activation request does not carry business product information or initial business scenario classification information, the business scenario cannot be directly determined based on the fields of the activation request itself. Therefore, publicly available information related to the IoT tenant is obtained based on the tenant information, supplementing the business scenario-related content from external information sources. In this case, the tenant information of the IoT tenant is obtained. This tenant information is used to identify the specific tenant entity, and may include, for example, the tenant name, registration name, or other identifying data.

[0122] S22, Based on the tenant information, obtain the tenant public text information corresponding to the tenant information by calling an external public information service.

[0123] Specifically, publicly available external information services can include publicly accessible query interfaces, business information service interfaces, or other information services that can return information such as the tenant's business scope, main business, and product descriptions. Based on tenant information, a query request is initiated to the publicly available external information service, and the returned publicly available text information from the tenant is received, such as company profiles, descriptions of business scope, and product and service descriptions. By introducing publicly available external information services, even when the activation request lacks valid fields, textual materials for determining the business scenario can still be obtained based on the business information disclosed by the tenant in public channels, thereby reducing the workload of manual data entry or manual queries.

[0124] S23. Preprocess and segment the publicly available text information of the tenant to obtain multiple text units, and identify and obtain entity information corresponding to the business scenario based on the multiple text units.

[0125] Specifically, preprocessing may include normalizing spaces, punctuation, control characters, and irrelevant symbols in the text to remove obvious noise and make the text more suitable for further analysis. After preprocessing, the tenant's publicly available text information is segmented, breaking down coherent natural language content into multiple text units. Subsequently, entity information related to the business scenario is identified from these text units, such as industry names, application areas, product categories, and service types. Specifically, based on a pre-defined entity dictionary, a set of business domain keywords, and matching rules, text units can be filtered and categorized to identify combinations of units representing industry categories, application scenario types, product categories, and typical service functions as candidate entity information. Alternatively, statistical or model-based entity recognition methods can be combined to analyze the contextual relationships of multiple text units to determine whether they belong to entity types related to the business scenario. By preprocessing, segmenting, and identifying entity information in the tenant's publicly available text information, unstructured publicly available text is transformed into structured entity information that can be used by the business scenario, providing clear semantic input for the subsequent scenario matching process.

[0126] S24. Match the entity information in the business scenario classification knowledge graph to obtain multiple candidate business scenario classification information and the confidence level of each candidate business scenario classification information.

[0127] Specifically, the business scenario classification knowledge graph is used to represent the relationship between different entity information and business scenario classifications. It can include entity nodes, business scenario nodes, and the connections between them. Using the identified entity information as the entry point, the knowledge graph searches for business scenario classification nodes associated with these entity information, forming multiple candidate business scenario classifications. Simultaneously, based on factors such as the association strength, frequency of occurrence, and path weight between entity information and business scenario classifications, a confidence score is calculated for each candidate business scenario classification, measuring its credibility as a current tenant's business scenario. By matching within the business scenario classification knowledge graph, entity clues from publicly available tenant text information are mapped to a set of candidate business scenario classifications with clearly defined confidence scores, achieving automatic association from external publicly available data to the internal scenario classification system.

[0128] S25. Based on the confidence level corresponding to each candidate business scenario classification information, filter and obtain candidate business scenario classification information with a confidence level greater than or equal to a preset threshold to obtain a scenario classification candidate set.

[0129] Specifically, a pre-set confidence threshold is used as the judgment standard. The confidence level of each candidate business scenario classification information is compared. When the confidence level of a candidate business scenario classification information is greater than or equal to the preset threshold, the candidate business scenario classification information is retained; otherwise, it is discarded. After screening, all candidate business scenario classification information that meets the condition of confidence level greater than or equal to the preset threshold is collected to form a scenario classification candidate set. Through the screening mechanism based on confidence threshold, obviously unreliable or low-relevance candidate business scenario classification information is filtered out, making the base set used to determine the target business scenario classification information more focused on scenarios highly related to the tenant's business, thereby improving the credibility of the business scenario judgment results.

[0130] S26, when the scenario classification candidate set contains multiple different business scenario classification information, each business scenario classification information is confirmed as multiple target business scenario classification information corresponding to the activation request.

[0131] S27. When the scenario classification candidate set contains single business scenario classification information, the business scenario classification information is confirmed as the target business scenario classification information corresponding to the activation request.

[0132] Specifically, based on the number of business scenario classification information in the scenario classification candidate set, the target business scenario classification information corresponding to the activation request is determined separately. When the scenario classification candidate set contains multiple different business scenario classification information, it indicates that, based on the tenant's public text information and entity recognition results, the current tenant's business may simultaneously cover multiple business scenario types. In this case, each business scenario classification information in the scenario classification candidate set is confirmed as the multiple target business scenario classification information corresponding to the activation request, so that subsequent processing can allocate resources and execute strategies separately for each business scenario.

[0133] When the scenario classification candidate set contains only a single business scenario classification information, this business scenario classification information is directly confirmed as the target business scenario classification information corresponding to the activation request. By first forming a scenario classification candidate set according to a confidence threshold, and then processing the business scenario classification information separately according to the number of business scenario classification information in the set, a unified judgment is achieved for both multi-scenario and single-scenario cases. This ensures the reliability of the business scenario classification results while also taking into account the expression needs of tenants with multiple business scenarios, providing an accurate and hierarchical business scenario foundation for subsequent data center allocation.

[0134] In summary, by utilizing a business scenario classification knowledge graph and a confidence threshold mechanism, even when an activation request lacks internal scenario fields, it can still automatically determine one or more target business scenario classification information for the activation request. This ensures that the subsequent data center allocation process has a clear business scenario basis, reduces the involvement of manual data entry and judgment, and improves the automation level and accuracy of the overall process.

[0135] In one embodiment, the business scenario classification knowledge graph can be constructed based on the platform's internal business scenario classification system. The knowledge graph uses business scenario classification information as scenario nodes and industry names, application areas, typical product names, and functional descriptions related to the business scenarios as entity nodes. Edges representing relationships are established between scenario nodes and entity nodes. The initial structure of the knowledge graph can be generated based on existing industry classification standards, the platform's preset business scenario classification table, and historical business configuration data, ensuring that the knowledge graph covers major industry categories and typical application scenarios during construction. By organizing business scenario classification information and various entity information in a graph structure, subsequent business scenario matching based on entity information can quickly locate the relevant business scenario classification information using the graph structure.

[0136] Furthermore, the business scenario classification knowledge graph can be dynamically expanded and updated by combining external data sources and business data during platform operation. During long-term operation, existing entity nodes and scenario nodes can be supplemented and corrected based on newly connected IoT tenants, actual business activation records, and manual review results. For example, when a previously unrecorded entity is repeatedly identified in a tenant's public text information, and this entity can be stably mapped to a specific business scenario classification after manual confirmation or statistical analysis, a corresponding entity node can be added to the business scenario classification knowledge graph, and a relationship between this entity node and the relevant business scenario classification information can be established. When an existing relationship is found to no longer conform to the current business reality, the corresponding relationship can also be adjusted. By iteratively updating the business scenario classification knowledge graph during operation, the knowledge graph can continuously improve with the evolution of IoT businesses, enhancing the accuracy and coverage of business scenario matching based on the knowledge graph.

[0137] Furthermore, business scenario classification information is typically a three-level classification (industry affiliation, sub-scenario, and product category), but it can also be expanded into a multi-level hierarchical structure depending on the business complexity of the IoT platform. For example, in addition to industry affiliation, sub-scenario, and product category, dimensions such as application mode, billing strategy, or security level can be added. The processing flow used in this application to determine the target business scenario classification information and the processing flow for matching data center allocation rules based on business scenario classification information are equally applicable to the expanded multi-level business scenario classification information. Simply adding the corresponding level of classification field to the business scenario classification conditions can achieve unified management and allocation control of multi-level business scenarios.

[0138] In one embodiment, the data center allocation rule set includes multiple data center allocation rules, and the data center allocation rules include business scenario classification conditions, data center codes, and priority information;

[0139] In this embodiment, the data center allocation rule set is used to centrally manage allocation strategies for different business scenarios. The data center allocation rule set includes multiple data center allocation rules, each corresponding to the allocation configuration for one or more business scenarios. The data center allocation strategies are stored in a structured manner as a rule set. This facilitates flexible maintenance of the allocation strategies as business scale and types evolve.

[0140] Data center allocation rules include business scenario classification conditions, data center codes, and priority information. Business scenario classification conditions describe the scope of business scenarios to which the rule applies. These conditions can correspond to one or more dimensions in the target business scenario classification information, such as codes for industry affiliation, sub-scenarios, or product categories, thus limiting the types of business scenarios to which the rule applies. By configuring business scenario classification conditions in data center allocation rules, requests from different business categories and application scenarios can be assigned to different allocation strategies, enabling fine-grained control of data center allocation based on business differences. Setting business scenario classification conditions allows for precise positioning of data center allocation rules at the business scenario level, enabling the rule set to cover a variety of different business scenario combinations.

[0141] Data center codes are used to identify the target or candidate data center corresponding to a given data center allocation rule. They are typically marked using a unified coding system within the platform, marking data centers, their regions, availability zones, etc. By configuring data center codes in each data center allocation rule, the data center corresponding to the current business scenario classification conditions can be clearly identified. Upon matching a rule, the specific identifier of the data center can be directly obtained for subsequent resource allocation and service activation control. Priority information is used to control the order of rule application when multiple rules simultaneously satisfy the same target business scenario classification information. Different data center allocation rules may have the same or similar business scenario classification conditions, but different priorities can be assigned based on factors such as business importance, resource capabilities, and geographical location. When performing rule matching and selection, priority information can be used as the sorting criterion, prioritizing the data center codes corresponding to rules with higher priority as candidate results.

[0142] Furthermore, an implementation method for step S13 is provided here. Based on the above embodiment, obtaining the initial data center code corresponding to the target business scenario classification information includes:

[0143] S131, Read the target business scenario classification information and obtain the industry affiliation code, sub-scenario code and product classification code from the target business scenario classification information;

[0144] S132, obtain the allocation rules for each data center, and obtain the business scenario classification conditions, data center codes and priority information corresponding to the data center allocation rules;

[0145] S133, determine whether the industry affiliation, sub-scenario and product classification in each business scenario classification condition match the industry affiliation code, sub-scenario code and product classification code respectively, and add the matching data center allocation rules to the candidate set;

[0146] S134. From the candidate set, select the data center allocation rule with the highest priority according to the priority information, and determine the data center code corresponding to the data center allocation rule with the highest priority as the initial data center code corresponding to the target business scenario classification information.

[0147] In this embodiment, the target business scenario classification information is first parsed to read the encoded information representing the business scenario hierarchy. The target business scenario classification information can be encoded according to a pre-defined business scenario classification system, including at least industry affiliation codes, sub-scenario codes, and product classification codes. By reading the target business scenario classification information and obtaining the industry affiliation codes, sub-scenario codes, and product classification codes respectively, the abstract business scenario identifier is decomposed into multiple classification dimensions with clear semantics, enabling subsequent data center allocation rules to be matched at a fine-grained business scenario level.

[0148] After parsing the target business scenario classification information, the corresponding business scenario classification conditions, data center codes, and priority information are read from each data center allocation rule. The business scenario classification conditions describe the scope of business scenarios to which the rule applies, the data center code indicates the target data center corresponding to the rule, and the priority information indicates the selection order when multiple rules apply.

[0149] Next, using industry affiliation codes, sub-scenario codes, and product classification codes as benchmarks, each business scenario classification condition in the data center allocation rules is evaluated. Specifically, it is determined whether the industry affiliation, sub-scenario, and product classification conditions in each business scenario classification condition match the aforementioned industry affiliation codes, sub-scenario codes, and product classification codes, respectively. When a data center allocation rule is determined to meet the matching relationship in these classification dimensions, it is considered a rule applicable to the current target business scenario classification information and added to the candidate set. Through the above matching process, a set of candidate rules that match the target business scenario classification information can be filtered from the data center allocation rule set, achieving a contraction from the full rule set to the set of rules available for the current business scenario. This improves the accuracy and relevance of the rule matching results.

[0150] After obtaining a candidate set containing one or more data center allocation rules, the rules in the candidate set are compared based on the priority information carried by each candidate rule, and the data center allocation rule with the highest priority is selected as the final allocation basis for the current business scenario. Subsequently, the corresponding data center code is read from the data center allocation rule with the highest priority, and this data center code is determined as the initial data center code corresponding to the target business scenario classification information. By selecting an optimal rule from the candidate set based on priority information and determining the initial data center code, orderly decision-making among multiple candidate data centers is achieved, enabling the data center allocation process to reflect the preset business strategy while maintaining determinism and controllability at the execution level.

[0151] In one example, the data center allocation rule set can simultaneously contain the following two data center allocation rules: The first rule's business scenario classification conditions are industry affiliation code E001, sub-scenario code F001, and product classification code G001, with a corresponding data center code of DC1_A1 and a priority of 1; the second rule's business scenario classification conditions are industry affiliation code E001, sub-scenario code F001, and any product classification code, with a corresponding data center code of DC1_A2 and a priority of 2. When the target business scenario classification information is E001_F001_G001, it satisfies both the exact matching condition of the first rule and the arbitrary product classification condition of the second rule. In this case, the system selects the first rule with higher priority based on the priority information, determining DC1_A1 as the initial data center code. This method ensures a unique initial data center code even when rule conditions have inclusion or overlap, avoiding conflicts between different rules.

[0152] In one embodiment, an implementation of step S14 is provided. Based on the above embodiment, obtaining the retrieval results corresponding to the target business scenario classification information includes:

[0153] S141, Based on the preset tenant allocation record table, according to the initial data center code, select the tenant allocation records that are the same as the initial data center code to obtain a candidate tenant allocation record set;

[0154] S142, perform text preprocessing on the tenant information and the historical allocation records in the candidate tenant allocation record set to obtain the tenant information text and the historical tenant information text;

[0155] S143, Based on preset semantic matching rules, calculate the similarity between the tenant information text and the information text of each historical tenant;

[0156] S144, If there is historical tenant information text with a similarity greater than or equal to the preset similarity threshold, the search result will be recorded as having a corresponding tenant allocation record.

[0157] S145. If there is no historical tenant information text with a similarity greater than the preset similarity threshold, the search result will be recorded as no corresponding tenant allocation record exists.

[0158] In this embodiment, the tenant allocation record table is a pre-maintained set of historical allocation data used to record the existing allocation relationships of IoT tenants in various data centers. Based on the pre-set tenant allocation record table, the records are filtered according to the initial data center code, retaining only the tenant allocation records whose data center code is the same as the initial data center code, thus obtaining a candidate tenant allocation record set. By using the initial data center code as the filtering condition, the search scope is limited to the historical records related to the currently recommended data center. This reduces the scale and complexity of subsequent matching calculations and ensures that the search results are directly related to the current candidate data center.

[0159] After obtaining the candidate tenant allocation record set, to improve the accuracy of tenant information matching, text preprocessing is required for both the tenant information and the historical allocation records in the candidate tenant allocation record set, resulting in tenant information text and historical tenant information text. Text preprocessing can include standardizing spaces, punctuation marks, capitalization, and common prefixes and suffixes in tenant names, and unifying equivalent expressions under different writing conventions. This transforms the original tenant information and the tenant information in historical records into a standardized text format that is easy to compare. By performing text preprocessing on the tenant information and the tenant information in historical allocation records, interference caused by differences in writing format and additional modifiers is reduced, allowing subsequent similarity calculations to focus more on the differences in the tenant identifiers themselves.

[0160] After text preprocessing, the similarity between the tenant information text and each historical tenant information text is calculated. Pre-defined semantic matching rules comprehensively consider character-level similarity, word-level similarity, and overall semantic closeness to quantify the similarity between the tenant information text and historical tenant information text. For example, they can consider the degree of keyword overlap in names, the consistency of different field combinations, and the correspondence of similar words or abbreviations. By calculating the similarity between the tenant information text and each historical tenant information text in the candidate tenant allocation record set, the subjective judgment of whether they belong to the same tenant is transformed into a comparable similarity value. By setting semantic matching rules, even when faced with situations where tenant names are not completely consistent, or when there are abbreviations or extended descriptions, the historical records corresponding to the current tenant information can still be accurately identified.

[0161] If historical tenant information text with a similarity greater than or equal to a preset similarity threshold exists, it indicates that a historical record highly matching the current tenant information exists in the candidate tenant allocation record set. In this case, the search result is recorded as having a corresponding tenant allocation record, indicating that the current tenant already has an existing allocation relationship in the data center corresponding to the initial data center code. If no historical tenant information text with a similarity greater than the preset similarity threshold exists, it is considered that no historical record matching the current tenant information has been found within the data center range. In this case, the search result is recorded as not having a corresponding tenant allocation record. By combining similarity and the preset threshold, automatic judgment of search results is achieved. Even when there are slight differences or inconsistencies in the tenant information, it is still possible to reliably distinguish between cases with existing allocation records and cases without allocation records, providing a basis for decision-making on whether to continue using historical allocation results or adopt the initial data center code.

[0162] Next, based on the above embodiments, the target data center code corresponding to the target business scenario classification information is determined, including:

[0163] S1511, If ​​the search result shows that there is a corresponding tenant allocation record, then read the data center code from the tenant allocation record and use the read data center code as the target data center code corresponding to the target business scenario classification information;

[0164] S1512, If the search result is that there is no corresponding tenant allocation record, then the initial data center code is used as the target data center code corresponding to the target business scenario classification information.

[0165] In this embodiment, when the search result indicates the existence of a corresponding tenant allocation record, it means that a historical allocation relationship related to the current IoT tenant has been stored within the data center range corresponding to the initial data center code. At this time, the data center code is read from the tenant allocation record and directly used as the target data center code corresponding to the target business scenario classification information. By using the existing data center code in the historical tenant allocation record, the current activation request continues to use the existing bearer location in data center selection, thereby ensuring the continuity and consistency of the deployment of the same IoT tenant at the data center level, reducing the dispersion of the same tenant's business across multiple data centers, and reducing the complexity of subsequent business migration, data synchronization, and operation and maintenance management.

[0166] When the search result indicates that no corresponding tenant allocation record exists, it means that no existing allocation record matching the current tenant information has been found within the data center range corresponding to the initial data center code. In other words, the IoT tenant has not yet established a corresponding historical bearer relationship in the current candidate data centers. In this case, the initial data center code is directly used as the target data center code corresponding to the target business scenario classification information. The initial data center code originates from the aforementioned matching result based on the target business scenario classification information and the data center allocation rule set, reflecting the recommended selection of data centers according to the business scenario allocation strategy. Using the initial data center code when no historical allocation record exists ensures that the data center allocation result follows the preset business scenario allocation rules, allowing new tenants or new business scenarios to still receive a data center bearer solution matching their business characteristics during the initial allocation. By combining the search result with the initial data center code for determining the target data center code, a coordinated balance between rule recommendation and historical allocation is achieved. This helps to ensure that the data center allocation result matches the business scenario while reducing unnecessary resource reconstruction and business migration, thereby improving the stability and resource utilization efficiency of the overall data center allocation process.

[0167] In one embodiment, data center allocation for the activation request based on the target data center code includes:

[0168] S1521, Based on the target data center code, the activation request is sent to the corresponding data center to complete the service activation by calling the data center activation interface corresponding to the target data center code;

[0169] S1522 After the service is activated, the tenant information, target business scenario classification information and target data center code are written into the tenant allocation record table.

[0170] In this embodiment, the target data center code is used as an indicator parameter to call the data center activation interface corresponding to that target data center code, sending the activation request to the corresponding data center for processing. The data center activation interface can be a service access interface provided by each data center to the outside world, used to receive activation requests from the IoT platform and complete resource creation, configuration distribution, and service binding operations locally. When calling the data center activation interface, information related to service activation in the activation request can be sent to the target data center, enabling the target data center to establish a service-bearing environment for the corresponding IoT tenant based on this information. By selecting the appropriate data center activation interface based on the target data center code and sending the activation request, the connection between data center allocation decisions and specific service activation execution is achieved, ensuring that the allocation results are actually implemented at the underlying resource level.

[0171] After the target data center completes the service activation request, the tenant information, target service scenario classification information, and target data center code involved in this service activation are written into the tenant allocation record table. The tenant allocation record table, as a data structure recording the historical allocation relationships of IoT tenants, stores the current service activation result by adding or updating records. By associating tenant information with target service scenario classification information and target data center code, historical allocation information can be directly read from the tenant allocation record table when processing new services, changing services, or querying service carrying status for that tenant, enabling cross-request allocation relationship tracing. By writing relevant information into the tenant allocation record table after service activation, the current data center allocation result is incorporated into a unified historical allocation management system. This provides data support for subsequent data center selection based on historical records and improves the traceability and maintainability of the service activation process, helping to maintain consistency between data center allocation strategies and actual carrying status.

[0172] Figure 3 This is a schematic diagram illustrating the activation request processing flow provided in an embodiment of this application. Figure 3 As shown, the overall process of this application is explained, including: first, receiving a tenant activation request (i.e., activation request), obtaining the customer name (i.e., the tenant name field in the tenant information), business scenario classification information (i.e., initial business scenario classification information), and activation product description (i.e. activation product description information), and calling the sub-process of extracting or parsing the activation product description to obtain the activation product list (i.e., the process of determining that the activation request carries business product information and generating the activation product list).

[0173] If a list of products to be activated can be obtained, the product table (i.e., product information table) is queried based on the product names (i.e., candidate product names) in the list to obtain a set of business scenario categories (i.e., candidate scenario categories) corresponding to multiple products. When multiple different business scenario categories exist, the set of business scenario categories can be split to generate multiple new tenant activation requests, which are then sent to the tenant activation request splitting process (i.e., the implementation method of splitting activation requests based on multiple target business scenario category information), and the current process ends. When only a single business scenario category exists, the business scenario category obtained based on the product (i.e., the target business scenario category information determined based on business product information) is compared with the business scenario category of the original tenant activation request (i.e., the initial business scenario category information). If they match, the business scenario category is directly used as the allocation basis to enter the tenant allocation sub-process; if they do not match, the business scenario category obtained based on the product is used as the default business scenario category, and the process enters the tenant allocation sub-process.

[0174] If the product list is not obtained in the aforementioned judgment, the process continues to check whether the business scenario category of the tenant's activation request is missing, i.e., whether the activation request carries initial business scenario category information. If the business scenario category is not missing, the business scenario category in the tenant's activation request is directly used as the allocation basis, and the tenant allocation sub-process begins. If the business scenario category is missing, the process of obtaining business scenario category information from public channels is invoked (i.e., the process of inferring the business scenario based on the tenant information by calling external public information services and using the business scenario category knowledge graph), and the business scenario category obtained from public channels is used as the default business scenario category and as the allocation basis, and the process continues to determine whether the missing information has been filled in. If the missing information has been filled in, the process continues to the tenant allocation sub-process. If the missing information is still not filled in, the business scenario is filled in according to the business scenario category associated with the key business channel logic configuration (i.e., the key business channel logic configuration and human-machine collaborative processing process in the specification, which may include manually filling in the business scenario field or opening the tenant based on the business scenario category corresponding to the product name), and the business scenario is filled in. The filled business scenario category is used as the final allocation basis, and the process continues to the tenant allocation sub-process to complete the subsequent data center allocation.

[0175] Figure 4 This is a schematic diagram of the structure of the data center allocation device provided in the embodiments of this application, such as... Figure 4 As shown, the data center allocation device 40 provided in this embodiment includes:

[0176] The request receiving module 401 is used to receive and parse the activation request of the IoT tenant, and when it is determined that the activation request carries business product information and / or initial business scenario classification information, it obtains the business product information and / or initial business scenario classification information.

[0177] Data processing module 402 is used to obtain the target business scenario classification information corresponding to the activation request based on business product information and / or initial business scenario classification information, according to preset business scenario determination rules;

[0178] The data processing module 402 is also used to match the target business scenario classification information in the data center allocation rule set to obtain the initial data center code corresponding to the target business scenario classification information.

[0179] The information retrieval module 403 is used to obtain the tenant information of the IoT tenant, and retrieve the tenant allocation record table based on the tenant information and the initial data center code to obtain the retrieval results corresponding to the target business scenario classification information;

[0180] The central allocation module 404 is used to determine the target data center code corresponding to the target business scenario classification information based on the search results, and to allocate a data center to the activation request based on the target data center code.

[0181] The data center allocation device 40 provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0182] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 5 As shown, the electronic device 50 provided in this embodiment includes at least one processor 501 and a memory 502. Optionally, the electronic device 50 further includes a communication component 503. The processor 501, memory 502, and communication component 503 are connected via a bus 504.

[0183] In a specific implementation, at least one processor 501 executes computer execution instructions stored in memory 502, causing at least one processor 501 to perform the above-described method.

[0184] The specific implementation process of processor 501 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0185] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0186] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0187] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0188] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0189] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A data center allocation method, characterized in that, include: Receive and parse the activation request from the IoT tenant, and when it is determined that the activation request carries business product information and / or initial business scenario classification information, obtain the business product information and / or initial business scenario classification information; Based on the business product information and / or initial business scenario classification information, and according to the preset business scenario determination rules, the target business scenario classification information corresponding to the activation request is obtained; Based on the target business scenario classification information, a matching process is performed in the data center allocation rule set to obtain the initial data center code corresponding to the target business scenario classification information; Obtain the tenant information of the IoT tenant, and based on the tenant information and the initial data center code, retrieve the tenant allocation record table to obtain the retrieval results corresponding to the target business scenario classification information; Based on the search results, the target data center code corresponding to the target business scenario classification information is determined, and the activation request is allocated a data center based on the target data center code.

2. The method according to claim 1, characterized in that, It is determined that the activation request carries business product information, including: Obtain the product description information carried in the activation request, and preprocess the product description information to obtain the processed product description text; Based on a preset vocabulary library, the processed product description text is subjected to synonym replacement processing to obtain standard description text, and the standard description text is subjected to word segmentation processing to obtain word segmentation results. The vector similarity is calculated between the word segmentation results and each standard word in the preset standard product list, and a set of candidate product names matching the word segmentation results is determined based on the similarity results, which serves as the list of products to be activated. If no candidate product name is found in the list of products to be activated, it is determined that the activation request does not contain business product information. If there is at least one candidate product name in the list of products to be activated, then it is determined that the activation request carries business product information.

3. The method according to claim 2, characterized in that, Based on the initial business scenario classification information, and according to preset business scenario determination rules, the target business scenario classification information corresponding to the activation request is obtained, including: If it is determined that the activation request does not carry business product information, then it is determined whether the activation request carries initial business scenario classification information. If the activation request carries initial business scenario classification information, then the initial business scenario classification information is recorded as the target business scenario classification information corresponding to the activation request.

4. The method according to claim 2, characterized in that, Based on the aforementioned business product information, and according to preset business scenario determination rules, the target business scenario classification information corresponding to the activation request is obtained, including: Based on the list of activated products, the business scenario category records corresponding to each candidate product name are searched in the preset product information table, and deduplication is performed to obtain a candidate set of scenario categories. When the scenario classification candidate set contains multiple different business scenario classification information, each business scenario classification information is respectively confirmed as multiple target business scenario classification information corresponding to the activation request; Alternatively, if the scenario classification candidate set contains single business scenario classification information, the business scenario classification information will be confirmed as the target business scenario classification information corresponding to the activation request.

5. The method according to claim 1, characterized in that, Also includes: When it is determined that the activation request does not carry business product information and does not carry initial business scenario classification information, the tenant information of the IoT tenant is obtained; Based on the tenant information, the publicly available text information of the tenant corresponding to the tenant information is obtained by calling an external public information service; The publicly available text information of the tenant is preprocessed and segmented to obtain multiple text units, and entity information corresponding to the business scenario is identified and obtained based on the multiple text units. Based on the entity information, a matching process is performed in the business scenario classification knowledge graph to obtain multiple candidate business scenario classification information and the confidence level corresponding to each candidate business scenario classification information. Based on the confidence level corresponding to each candidate business scenario classification information, filter and obtain candidate business scenario classification information with a confidence level greater than or equal to a preset threshold to obtain a scenario classification candidate set; When the scenario classification candidate set contains multiple different business scenario classification information, each business scenario classification information is respectively confirmed as multiple target business scenario classification information corresponding to the activation request; Alternatively, if the scenario classification candidate set contains single business scenario classification information, the business scenario classification information will be confirmed as the target business scenario classification information corresponding to the activation request.

6. The method according to claim 1, characterized in that, The data center allocation rule set includes multiple data center allocation rules, which include business scenario classification conditions, data center codes, and priority information. Accordingly, the step of matching the target business scenario classification information in the data center allocation rule set to obtain the initial data center code corresponding to the target business scenario classification information includes: Read the target business scenario classification information to obtain the industry affiliation code, sub-scenario code, and product classification code from the target business scenario classification information; Obtain the allocation rules for each data center, and obtain the business scenario classification conditions, data center codes, and priority information corresponding to the data center allocation rules; Determine whether the industry affiliation, sub-scenario, and product classification in each business scenario classification condition match the industry affiliation code, sub-scenario code, and product classification code, respectively, and add the matching data center allocation rules to the candidate set; From the candidate set, the data center allocation rule with the highest priority is selected according to the priority information, and the data center code corresponding to the data center allocation rule with the highest priority is determined as the initial data center code corresponding to the target business scenario classification information.

7. The method according to claim 1, characterized in that, The step of retrieving the tenant allocation record table based on the tenant information and the initial data center code to obtain the retrieval results corresponding to the target business scenario classification information includes: Based on the preset tenant allocation record table, according to the initial data center code, the tenant allocation records with the same initial data center code are filtered out to obtain a candidate tenant allocation record set; The tenant information and the historical allocation records in the candidate tenant allocation record set are preprocessed to obtain the tenant information text and the historical tenant information text. Based on preset semantic matching rules, the similarity between the tenant information text and the information text of each historical tenant is calculated; If there is historical tenant information text with a similarity greater than or equal to a preset similarity threshold, the search result will be recorded as having a corresponding tenant allocation record. If no historical tenant information text with a similarity greater than the preset similarity threshold exists, the search result will be recorded as no corresponding tenant allocation record exists.

8. The method according to claim 7, characterized in that, The step of determining the target data center code corresponding to the target business scenario classification information based on the search results includes: If the search result indicates the existence of a corresponding tenant allocation record, then the data center code is read from the tenant allocation record, and the read data center code is used as the target data center code corresponding to the target business scenario classification information; If the search result indicates that there is no corresponding tenant allocation record, then the initial data center code will be used as the target data center code corresponding to the target business scenario classification information; Accordingly, the process of allocating a data center for the activation request based on the target data center code includes: Based on the target data center code, the activation request is sent to the corresponding data center to complete the service activation by calling the data center activation interface corresponding to the target data center code; After the service is activated, the tenant information, the target business scenario classification information, and the target data center code are written into the tenant allocation record table.

9. A data center distribution device, characterized in that, include: The request receiving module is used to receive and parse the activation request of the IoT tenant, and when it is determined that the activation request carries business product information and / or initial business scenario classification information, the module obtains the business product information and / or initial business scenario classification information. The data processing module is used to obtain the target business scenario classification information corresponding to the activation request based on the business product information and / or the initial business scenario classification information, according to the preset business scenario determination rules. The data processing module is used to match the target business scenario classification information in the data center allocation rule set to obtain the initial data center code corresponding to the target business scenario classification information. The information retrieval module is used to obtain the tenant information of the IoT tenant, and retrieve the tenant allocation record table based on the tenant information and the initial data center code to obtain the retrieval results corresponding to the target business scenario classification information; The central allocation module is used to determine the target data center code corresponding to the target business scenario classification information based on the search results, and to allocate a data center to the activation request based on the target data center code.

10. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1 to 8.

Images