Dependence configuration method and device for batch jobs, electronic equipment, medium and program product

Abstract

The invention provides a dependency configuration method and device for batch jobs, electronic equipment, a medium and a program product, relates to application of a large model in batch job processing, and can be applied to the technical field of artificial intelligence. The method comprises the following steps: acquiring a dependency configuration request of a user for a target batch job; pre-generated cue word protocol information is obtained, and the cue word protocol information is generated at least based on the batch job dependency configuration rule information and the batch job dependency relationship data; the cue word specification information is updated based on the update content of the batch job dependency configuration rule information; inputting the cue word specification information, the dependency configuration request and target batch job dependency relationship data associated with the target batch job into a large language model to generate a dependency configuration result; and verifying the dependency configuration result, and executing dependency configuration operation under the condition that the verification result represents that verification is passed.

Description

Technical Field

[0001] This application relates to the application of large models in batch job processing, to the field of artificial intelligence technology, and more specifically to a batch job dependency configuration method, apparatus, device, medium and program product. Background Technology

[0002] With the development of data warehousing and offline analytics, the number of batch jobs within enterprises continues to grow, creating complex upstream and downstream dependencies between jobs. To ensure the timeliness and consistency of data output, scheduling platforms typically need to configure job dependencies so that downstream execution is triggered only after upstream execution is completed. However, existing solutions mostly rely on manual maintenance. Development and operations personnel need to manually determine and configure dependencies based on input and output data, and in cross-system and cross-topic scenarios, they also need to supplement information with documentation and experience. This approach is inefficient and costly to maintain when the scale of jobs increases and changes frequently, and it lacks a global perspective, easily leading to problems such as circular and redundant dependencies. At the same time, the numerous dependency configuration elements make manual operation prone to errors, which may lead to scheduling anomalies, data inconsistencies, or even delays in critical business operations, making it difficult to meet the actual needs of managing large-scale job dependencies. Summary of the Invention

[0003] In view of the above problems, this application provides a method, apparatus, device, medium and program product for batch operation dependency configuration.

[0004] According to a first aspect of this application, a dependency configuration method for batch jobs is provided. The method includes: obtaining a user's dependency configuration request for a target batch job; obtaining pre-generated prompt word specification information, wherein the prompt word specification information is generated based at least on batch job dependency configuration rule information and batch job dependency relationship data, the batch job dependency relationship data being used to characterize the global dependency relationship between multiple batch jobs; updating the prompt word specification information based on the updated content of the batch job dependency configuration rule information; inputting the prompt word specification information, the dependency configuration request, and the target batch job dependency relationship data associated with the target batch job into a large language model to generate a dependency configuration result; and validating the dependency configuration result, and performing a dependency configuration operation if the validation result indicates that the validation has passed.

[0005] According to an embodiment of this application, the prompt word specification information is updated based on the updated content of the batch job dependency configuration rule information, including: obtaining the updated content, performing structured parsing on the updated content to obtain a set of rule change items; determining specification difference information based on the correspondence between the set of rule change items and the specification clauses in the prompt word specification information; and performing incremental updates on the prompt word specification information based on the specification difference information.

[0006] According to an embodiment of this application, determining the specification difference information based on the correspondence between the set of rule change items and the specification clauses in the prompt word specification information includes: determining the scope of influence of rule changes based on the set of rule change items, wherein the scope of influence of rule changes is used to represent the affected dependency configuration fields, dependency types, or job types; determining a subset of target specification clauses in the prompt word specification information based on the correspondence between the scope of influence of rule changes and the specification clauses; and determining the specification difference information based on the subset of target specification clauses and the scope of influence of rule changes, wherein the specification difference information includes operation type information and clause content information, wherein the operation type information is used to indicate the addition, modification, or deletion operation to be performed on the subset of target specification clauses.

[0007] According to embodiments of this application, the prompt word specification information is generated at least based on batch job dependency configuration rule information and batch job dependency relationship data, including: extracting configuration constraint information from the batch job dependency configuration rule information to form a specification clause set, wherein the configuration constraint information includes an output field set, field value constraints, job type constraints, dependency relationship constraints, and / or format constraints; extracting job definition information and dependency relationship structure information from the batch job dependency relationship data to generate a structured job context; and combining the specification clause set and the structured job context to generate the prompt word specification information.

[0008] According to an embodiment of this application, the step of inputting the prompt word specification information, the dependency configuration request, and the target batch job dependency relationship data associated with the target batch job into a large language model to generate a dependency configuration result includes: determining candidate upstream job data and / or candidate downstream job data associated with the target batch job based on the batch job dependency relationship data; constructing the target batch job dependency relationship data based on the candidate upstream job data and / or the candidate downstream job data; converting the target batch job dependency relationship data into a structured context; arranging the structured context according to the paragraph structure specified by the prompt word specification information to obtain an arrangement result; and combining the arrangement result with the prompt word specification information and the dependency configuration request to obtain a model input, inputting the model input into the large language model to generate the dependency configuration result.

[0009] According to an embodiment of this application, determining candidate upstream job data and / or candidate downstream job data associated with the target batch job includes: obtaining a global dependency graph constructed based on the batch job dependency data; determining a preset hop count neighborhood subgraph associated with the target batch job based on the global dependency graph; filtering job nodes in the preset hop count neighborhood subgraph based on data output identifier matching operation and / or scheduling window matching operation to obtain a candidate job set; and determining the candidate upstream job data and / or the candidate downstream job data based on the candidate job set.

[0010] According to an embodiment of this application, the verification of the dependency configuration result includes circular dependency detection, specifically including: constructing a target dependency graph based on the dependency configuration result, performing loop detection on the target dependency graph to obtain a circular dependency detection result; and determining the verification result based on the circular dependency detection result, wherein, if the verification result indicates that the verification fails, location information is generated, and the location information includes the job link segment that caused the circular dependency.

[0011] According to an embodiment of this application, the execution of dependency configuration operation includes: converting the dependency configuration result into structured configuration data; formatting the structured configuration data based on the batch job dependency configuration rule information to obtain target configuration data; generating a configuration version identifier based on the target configuration data; and writing the configuration version identifier and the target configuration data into the batch job dependency configuration storage or submitting them to the batch job dependency configuration system.

[0012] A second aspect of this application provides a dependency configuration apparatus for batch jobs. The apparatus includes: a data acquisition module, configured to: acquire a user's dependency configuration request for a target batch job; a prompt word acquisition module, configured to: acquire pre-generated prompt word specification information, wherein the prompt word specification information is generated at least based on batch job dependency configuration rule information and batch job dependency relationship data, the batch job dependency relationship data being used to characterize the global dependency relationship between multiple batch jobs; the prompt word specification information is updated based on the update content of the batch job dependency configuration rule information; a dependency configuration result generation module, configured to: input the prompt word specification information, the dependency configuration request, and the target batch job dependency relationship data associated with the target batch job into a large language model to generate a dependency configuration result; and a verification module, configured to: verify the dependency configuration result, and if the verification result indicates that the verification is successful, perform a dependency configuration operation.

[0013] A third aspect of this application provides an electronic device comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the method described above.

[0014] A fourth aspect of this application also provides a computer-readable storage medium having a computer program or instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0015] The fifth aspect of this application also provides a computer program product, including a computer program or instructions that, when executed by a processor, implement the steps of the above-described method.

[0016] According to embodiments of this application, by introducing a large language model, under the constraints of pre-generated prompt word specifications that can be dynamically adjusted with rule updates, the system automatically understands and generates batch job dependency configuration requests submitted by users. This achieves automated configuration of batch job dependencies, reducing the workload of manually analyzing upstream and downstream relationships and manually filling in configurations, thereby improving the efficiency and consistency of dependency configuration. Simultaneously, by combining the generated results with global batch job dependency data and performing verification after generation, dependency configuration operations are only executed if the verification passes, effectively reducing the probability of circular dependencies, erroneous dependencies, and other problems. Overall, embodiments of this application, while ensuring configuration accuracy and controllability, reduce the complexity and operational costs of maintaining batch job dependencies, and improve the stability and reliability of the batch job scheduling system. Attached Figure Description

[0017] The above-mentioned contents, other objects, features and advantages of this application will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0018] Figure 1 The illustration schematically depicts an application scenario of a batch job dependency configuration method, apparatus, device, medium, and program product according to embodiments of this application;

[0019] Figure 2 A flowchart illustrating a dependency configuration method for batch jobs according to an embodiment of this application is shown schematically.

[0020] Figure 3 A flowchart illustrating a method for generating prompt word specification information according to some exemplary embodiments of this application is shown schematically;

[0021] Figure 4 This illustration schematically shows a batch job dependency configuration system according to some exemplary embodiments of this application;

[0022] Figure 5 This schematically illustrates a structural block diagram of a batch job dependency configuration apparatus according to an embodiment of this application; and

[0023] Figure 6 A block diagram of an electronic device suitable for implementing a batch operation dependency configuration method according to an embodiment of this application is illustrated schematically. Detailed Implementation

[0024] The embodiments of this application will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of this application. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of this application for ease of explanation. However, it will be apparent that one or more embodiments may be implemented without these specific details. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of this application.

[0025] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0026] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0027] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0028] With the continuous development of data warehousing, offline computing, and report analysis, the number of batch jobs within enterprises is constantly increasing, and the chain relationship between jobs—input data, processing logic, and output data—is becoming increasingly complex. To ensure the timeliness and consistency of data output, batch scheduling platforms typically need to configure upstream and downstream dependencies for each job. This allows the scheduling system to trigger downstream job execution only after upstream data is ready, and to handle failures and delays by blocking, retrying, and issuing alerts, thereby achieving end-to-end data link management. In actual production environments, a business theme often consists of multiple data collection, cleaning, aggregation, indicator calculation, and report generation jobs linked together. These jobs are frequently added or modified due to iterative requirements, adjustments in definitions, and changes in data sources, and job dependencies evolve accordingly. How to efficiently, accurately, and sustainably maintain large-scale job dependencies has become a fundamental issue in batch scheduling governance.

[0029] Currently, the maintenance of batch job dependencies still relies heavily on manual configuration. Development and operations personnel typically need to determine upstream and downstream relationships based on the job's processing content, such as identifying input files or data tables that the job depends on, output files or data tables generated by the job, and intermediate result data provided by the job. Based on this, they then manually fill in configuration items such as the list of preceding jobs, trigger conditions, dependency types, and exception handling strategies for each target job in the batch platform. For cross-system, cross-topic, and cross-departmental data links, maintenance personnel may also need to supplement dependency information through documentation, verbal communication, historical experience, or ad-hoc troubleshooting to ultimately form a workable scheduling dependency network. While this approach is intuitive and has low implementation costs, it can meet basic needs in scenarios with small job scales and infrequent changes. However, when the number of jobs reaches a certain scale, the dependencies form a network structure, and there are frequent changes and multi-team collaborations, manual maintenance gradually reveals its significant limitations.

[0030] First, manual maintenance of dependencies is inefficient. Maintenance personnel need to check the upstream inputs and downstream outputs of each job individually to confirm the correspondence before configuring them on the platform. This process relies on individual experience and understanding, and it's difficult to establish stable batch processing capabilities. Once changes occur, such as data source migration, field definition adjustments, or task splitting and merging, multiple dependency configurations on the related job chain need to be updated synchronously. Maintenance costs increase exponentially, and configurations often lag behind changes, thus affecting the timeliness of scheduling plans.

[0031] Secondly, existing methods often focus on a single job or application perspective, lacking unified verification and governance capabilities from a global job perspective. Maintenance personnel typically only focus on the direct upstream and downstream of the target job, making it difficult to fully perceive its position and scope of influence in the global dependency network during configuration. In this situation, problems such as circular dependencies, redundant dependencies, repeated triggering, and broken links are more likely to occur after configuration. For example, some jobs may be refactored or replaced in historical iterations, but old dependencies may not be cleaned up in time, resulting in meaningless preceding relationships in the scheduling link; or multiple business teams may maintain their own local links without a unified view and consistency rules, easily forming implicit cycles or multiple dependencies at cross-topic junctions, causing scheduling to fail or unnecessary waiting.

[0032] Secondly, manual maintenance is prone to errors in handling large-scale job dependencies, leading to job scheduling anomalies. Dependency configurations typically include multiple elements such as job name, execution window, trigger conditions, dependency granularity, and success criteria. Any configuration error in any detail can trigger a chain reaction: at best, downstream jobs may execute prematurely and read incomplete data, resulting in data inconsistency; at worst, it can cause critical link blockages, overall output delays, and impact core scenarios such as business reports, risk control accounting, and operational analysis. Furthermore, as the number of jobs increases, the complexity of the dependency network significantly increases, and the cost of error investigation and repair also rises. It often requires repeated location, verification, and rollback across multiple jobs, affecting system stability and operational efficiency.

[0033] Based on this, embodiments of this application provide a dependency configuration method for batch jobs. The method includes: obtaining a user's dependency configuration request for a target batch job; obtaining pre-generated prompt word specification information, wherein the prompt word specification information is generated at least based on batch job dependency configuration rule information and batch job dependency relationship data, and the batch job dependency relationship data is used to represent the global dependency relationship between multiple batch jobs; updating the prompt word specification information based on the updated content of the batch job dependency configuration rule information; inputting the prompt word specification information, the dependency configuration request, and the target batch job dependency relationship data associated with the target batch job into a large language model to generate a dependency configuration result; and validating the dependency configuration result, and performing a dependency configuration operation if the validation result indicates that the validation is successful. According to embodiments of this application, by introducing a large language model, under the constraints of pre-generated prompt word specifications that can be dynamically adjusted with rule updates, the system automatically understands and generates batch job dependency configuration requests submitted by users. This achieves automated configuration of batch job dependencies, reducing the workload of manually analyzing upstream and downstream relationships and manually filling in configurations, thereby improving the efficiency and consistency of dependency configuration. Simultaneously, by combining the generated results with global batch job dependency data and performing verification after generation, dependency configuration operations are only executed if the verification passes, effectively reducing the probability of circular dependencies, erroneous dependencies, and other problems. Overall, embodiments of this application, while ensuring configuration accuracy and controllability, reduce the complexity and operational costs of maintaining batch job dependencies, and improve the stability and reliability of the batch job scheduling system.

[0034] It should be noted that the batch job dependency configuration method, apparatus, device, medium, and program product defined in this application can be used in the fields of artificial intelligence technology, cloud computing technology, and fintech, and can also be used in a variety of fields other than artificial intelligence technology, cloud computing technology, and fintech. The application fields of the batch job dependency configuration method, apparatus, device, medium, and program product provided in the embodiments of this application are not limited.

[0035] In the technical solution of this application, the user information (including but not limited to user personal information, user image information, user device information, such as location information) and data (including but not limited to data used for analysis, stored data, and displayed data) involved are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of related data all comply with relevant laws, regulations, and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entry points for users to choose to authorize or refuse.

[0036] In scenarios where personal information is used for automated decision-making, the methods, devices, and systems provided in this application all offer users corresponding operation entry points, allowing them to choose to agree to or reject the automated decision results; if the user chooses to reject, the process proceeds to expert decision-making. Here, "expert decision-making" refers to the decision-making activities of personnel who specialize in a particular field, possess specialized experience, knowledge, and skills, and have reached a certain level of professional expertise.

[0037] Figure 1 The illustration schematically depicts an application scenario of a batch job dependency configuration method, apparatus, device, medium, and program product according to embodiments of this application.

[0038] like Figure 1 As shown, application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 serves as a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0039] Users can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 via the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social media platform software, etc. (for example only).

[0040] In the embodiments of this application, the first terminal device 101 can be an example of the first device, and the second terminal device 102 and / or the third terminal device 103 can be an example of at least one second device. The first device and the second device can communicate collaboratively through an internal client mechanism to implement the data distribution and rendering logic described in the batch job dependency configuration method.

[0041] In some embodiments, the first device and at least one second device may be different display modules, windows or screens on the same computing terminal (such as a host), or multiple physical devices that work together through a network, such as different client instances deployed on a desktop computer, tablet terminal or mobile device respectively.

[0042] The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be various electronic devices with displays and support web browsing, including but not limited to smart mobile terminals, tablet computers, laptop computers, and desktop computers.

[0043] Server 105 can be a server that provides various services, such as a backend management server that supports websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103 (this is just an example). The backend management server can analyze and process data such as received user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal devices.

[0044] It should be noted that the batch job dependency configuration method provided in this application embodiment can generally be executed by server 105. Correspondingly, the batch job dependency configuration device provided in this application embodiment can generally be located in server 105. The batch job dependency configuration method provided in this application embodiment can also be executed by a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105. Correspondingly, the batch job dependency configuration device provided in this application embodiment can also be located in a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105.

[0045] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0046] The following will be based on Figure 1 The described scene, through Figures 2-4 The dependency configuration method for batch jobs in the disclosed embodiments is described in detail.

[0047] Figure 2 A flowchart illustrating a dependency configuration method for batch jobs according to an embodiment of this application is shown schematically.

[0048] like Figure 2 As shown, the batch job dependency configuration method 200 of this embodiment includes operations S210 to S240.

[0049] In operation S210, obtain the user's dependency configuration request for the target batch job.

[0050] In the embodiments of this application, dependency configuration requests can be submitted by users through batch job management interfaces, configuration pages, or APIs to express the user's intention to establish or adjust upstream and downstream dependencies for a target batch job. Dependency configuration requests can include adding dependencies, modifying existing dependencies, or confirming or supplementing existing dependencies. The content and format are not limited, as long as they reflect the dependency requirements between the target batch job and other batch jobs.

[0051] In some embodiments, the dependency configuration request may include job identification information of the target batch job, such as job name, job number, or the business theme to which the job belongs, to clarify the specific job object targeted by this configuration. Simultaneously, the dependency configuration request may also include user-described dependency information, such as which data results the target batch job depends on, the time granularity of the dependency, and the conditions for determining dependency completion. This descriptive information can be input in natural language or a combination of structured fields and natural language, allowing users to express their configuration intent without needing to understand complex dependency configuration rules.

[0052] In embodiments of this application, the dependency configuration request may also carry contextual information related to the target batch job, such as the current scheduling cycle, job type, data processing stage, or runtime environment identifier of the target batch job.

[0053] In some embodiments, when submitting a dependency configuration request, a user may specify only the target batch job itself without explicitly listing the specific upstream or downstream jobs. The system will then automatically identify potential dependent objects in subsequent processes by combining global batch job dependency data.

[0054] In the embodiments of this application, dependency configuration requests can also originate from different business roles or use cases. For example, during the development phase, developers can submit initial dependency configuration requests based on newly created jobs; during the operation and maintenance phase, operations personnel can submit dependency correction requests based on scheduling anomalies or business adjustments; in business change scenarios, business personnel can also trigger the reconfiguration of related batch job dependencies by describing changes in business processes. The system does not limit the source of dependency configuration requests and can uniformly handle configuration requirements from different roles.

[0055] In some embodiments, the dependency configuration request may also include constraints on dependencies, such as requiring the avoidance of certain types of dependencies, or limiting the number, hierarchy, or execution order of dependencies. This type of constraint information can serve as an important reference condition when generating subsequent dependency configuration results, ensuring that the generated results meet business needs while conforming to established scheduling rules and management requirements.

[0056] In operation S220, pre-generated prompt word specification information is obtained. The prompt word specification information is generated based at least on batch job dependency configuration rule information and batch job dependency relationship data. The batch job dependency relationship data is used to represent the global dependency relationship between multiple batch jobs. The prompt word specification information is updated based on the updated content of the batch job dependency configuration rule information.

[0057] In the embodiments of this application, the prompt word specification information can be pre-built and stored during the system initialization phase or during operation to form a unified constraint on the subsequent dependency configuration generation process. This prompt word specification information originates from the organization and abstraction of batch job dependency configuration rule information, and is supplemented by batch job dependency relationship data. This ensures that the prompt word specification information not only reflects the configuration rule requirements but also reflects the overall dependency structure that already exists or has historically formed among multiple batch jobs, thereby providing a constraint background that more closely resembles the real scheduling environment for the subsequent generation process.

[0058] In some embodiments, prompt word specification information can be generated based on batch job dependency configuration rule information. Batch job dependency configuration rule information may include allowed dependency types, dependency quantity limits, dependency direction constraints, dependency relationships between job types, and filling specifications for configuration fields. By summarizing and organizing this rule information, the system can form a set of general configuration constraint descriptions and transform them into prompt word specification information suitable for model processing, ensuring that the subsequently generated dependency configuration results naturally conform to the established rule requirements.

[0059] In embodiments of this application, the prompt word specification information can also be generated in conjunction with batch job dependency data. Batch job dependency data is used to characterize the overall dependency relationship of multiple batch jobs in historical execution or current configuration, such as upstream and downstream relationships between jobs, common dependency chain structures, and proven feasible dependency patterns.

[0060] In some embodiments, the prompt word specification information can be updated as batch job dependency configuration rule information changes. When the system detects an adjustment to the dependency configuration rules, such as adding rules, modifying existing constraints, or deprecating some configuration requirements, the system can synchronously update the prompt word specification information based on the updated rule content.

[0061] In the embodiments of this application, the update of the prompt word specification information can be performed incrementally. The system can parse the updated rule content, identify which configuration constraints have changed, and then replace, supplement, or delete the corresponding prompt word specification content without having to rebuild the entire specification information.

[0062] In some embodiments, prompt word specification information can also be organized according to different business scenarios or job categories. For example, different types of batch jobs may be subject to different constraints in dependency configuration. The system can prepare differentiated specification content for different job categories under a unified specification framework. When obtaining prompt word specification information, the corresponding specification content can be selected according to the attributes of the target batch job, thereby improving the targeting of subsequent dependency configuration generation.

[0063] In the embodiments of this application, the generation and updating process of prompt word specification information can involve a large language model. When constructing prompt word specification information, the system can use batch job dependency configuration rule information and batch job dependency relationship data as input, which are then processed and summarized by the large language model. The large language model generates prompt word specification content that conforms to the semantics of the rules and has a structured expression. In this way, prompt word specification information can better adapt to natural language understanding and generation scenarios while maintaining rule consistency.

[0064] In some embodiments, when the batch job dependency configuration rule information is updated, the system can use the rule content before and after the update as context input to the large language model. The large language model performs semantic analysis on the rule changes and adjusts the original prompt word specification information accordingly. Adjustment methods may include supplementing, replacing, or deleting the original specification content to ensure that the updated prompt word specification information is consistent with the latest rules. This approach avoids long-term lag in prompt word specifications due to rule updates, which could affect the accuracy of subsequent dependency configuration generation results.

[0065] In operation S230, the prompt word specification information, dependency configuration request, and target batch job dependency relationship data associated with the target batch job are input into the large language model to generate dependency configuration results.

[0066] In the embodiments of this application, by uniformly inputting multi-source information into the model, the model can not only understand the user's configuration intent when generating dependency configurations, but also refer to the established configuration constraints and the global job dependency structure at the same time, thereby forming a controlled and context-aware reasoning path during the generation process.

[0067] In some embodiments, prompt word specification information, when input into the model, can serve as a constraining context to guide the model in adhering to uniform configuration specifications and rules when understanding dependency configuration requests. Target batch job dependency data is used to provide the model with job topology information from a global perspective, enabling the model to avoid making judgments based solely on local information when generating dependency configuration results.

[0068] In the embodiments of this application, the target batch job dependency data may include upstream and downstream job information that are directly or indirectly related to the target batch job, such as common dependencies in historical configurations, execution order of adjacent jobs, and typical dependency chain structures.

[0069] In some embodiments, before inputting the above information into the large language model, the system can preprocess the dependency configuration request and target batch job dependency data, such as uniformly organizing key information like job identifier, job type, and scheduling cycle, to make the input content clearer and more orderly. The prompt word reduction information can then be combined with the above content according to a pre-designed paragraph structure to form an input context suitable for model processing, improving the model's consistency in understanding the input information.

[0070] In the embodiments of this application, after receiving the combined input information, the large language model can infer the dependency requirements of the target batch job based on its understanding of natural language semantics and its ability to comprehensively analyze contextual information, and generate a dependency configuration result. The generated dependency configuration result may include the set of upstream jobs that the target batch job needs to depend on, the direction of the dependency relationship, the triggering conditions of the dependency, and the execution order, so that the dependency configuration result can directly reflect the position of the target batch job in the overall job system.

[0071] In some embodiments, when generating dependency configuration results, the large language model can also refer to the constraints contained in the prompt word specification information to self-constrain the generated results. For example, when the prompt word specification information stipulates that a certain type of job is not allowed to have dependencies, the model can actively avoid configuring dependencies for that type of job during the generation process, thereby reducing the probability of failing results in the subsequent verification stage.

[0072] In the embodiments of this application, the large language model can be trained based on general text data during the pre-training stage. To make the large language model more suitable for batch job dependency configuration scenarios, fine-tuning can also be performed based on the pre-training. During the fine-tuning process, sample data related to batch job scheduling can be provided to the large language model, such as job dependency descriptions, dependency configuration rule texts, historical dependency configuration examples, and corresponding configuration results, so that the model can gradually learn the specific semantics and expression habits in the batch job dependency configuration domain. In this way, the large language model can more accurately grasp the business meaning of job dependencies when understanding user dependency configuration requests.

[0073] In some embodiments, the fine-tuning process can be conducted in a supervised manner. This involves inputting exemplary dependency configuration requests into the large language model and providing corresponding expected configuration results as a reference, allowing the model to gradually converge to a generation pattern that conforms to the dependency configuration rules during training. The fine-tuning data can come from historical manual configuration records, verified feasible dependency configuration schemes, or rule-driven standard examples, thereby ensuring the consistency between the training data and the actual scheduling environment.

[0074] In operation S240, the dependency configuration result is validated. If the validation result indicates that the validation is successful, the dependency configuration operation is executed.

[0075] In the embodiments of this application, by introducing a verification step after generating the dependency configuration result, the rationality and consistency of the generated result can be checked before the configuration is formally written or submitted, thereby preventing configurations that do not conform to existing scheduling rules or global dependency structures from taking effect directly. This verification process, as an intermediate step between generation and execution, enables the entire dependency configuration process to have the necessary security buffer and quality control capabilities.

[0076] In some embodiments, the validation of dependency configuration results can be based on pre-defined dependency validation rules. These rules can originate from batch job dependency configuration rule information and are used to constrain the legality of dependencies. For example, they can determine whether the dependency direction conforms to the scheduling order and verify whether the dependency relationship satisfies the conventions between job types. In this way, the system can systematically validate the generated results without relying on manual, item-by-item checks.

[0077] In the embodiments of this application, the verification process can be performed in conjunction with the position of the target batch job in the global dependency relationship. For example, the system can construct the dependency structure related to the target batch job based on the dependency configuration results and determine whether there are any unreasonable dependency links. If the verification result indicates that the verification fails, corresponding location information can be generated to indicate the possible location of the problem or the associated job relationship, thereby providing a reference for subsequent processing.

[0078] In some embodiments, when the verification result indicates that the verification passed, the system can proceed to the dependency configuration operation phase. The dependency configuration operation may include converting the dependency configuration result into a platform-recognizable configuration data format and writing it to the batch job dependency configuration storage, or directly submitting it to the batch job configuration system.

[0079] In the embodiments of this application, when performing dependency configuration operations, a corresponding version identifier can be generated for each configuration to distinguish configuration results generated at different times or from different requests. By introducing version identifiers, changes to dependency configurations can be clearly tracked during subsequent operation and maintenance or adjustments, enhancing the controllability of configuration management. At the same time, this approach also facilitates the retrospective or comparison of historical configurations when needed.

[0080] In some embodiments, dependency configuration operations that pass verification can also be judged in conjunction with the existing configuration status. For example, when the target batch job already has dependency configurations, the system can compare the differences between the newly generated dependency configuration results and the currently effective configurations before executing the operation, and then decide whether to overwrite, append, or replace them.

[0081] In the embodiments of this application, a failed validation does not directly trigger dependency configuration operations. Instead, the validation result can be fed back to subsequent processes. For example, the reason for the validation failure can be displayed to the user along with the generated dependency configuration result for reference or further adjustment of the dependency configuration request. In this way, the user can more intuitively understand the relationship between the generated result and the rule constraints, thereby modifying the configuration requirements when necessary.

[0082] In some embodiments, the verification process can also support the combined use of multiple verification strategies. For example, structural verification can be prioritized in some scenarios, while supplementary verification can be performed by combining time dimensions or job types in other scenarios. By flexibly configuring verification strategies, the verification process can adapt to batch job dependency configuration scenarios of different scales and complexities.

[0083] In the embodiments of this application, after verifying the dependency configuration result, the system can not only decide whether to execute the dependency configuration operation based on the verification result, but also use the verification result as feedback information for subsequent process optimization. When the verification result indicates that the verification fails, the system can record the corresponding verification conclusion and location information, and associate it with the prompt word specification information and dependency configuration request used when generating the dependency configuration result.

[0084] In some embodiments, the system can display the verification results and related location information to the user, allowing the user to confirm whether the generated result meets expectations. When the user confirms a problem, the user can describe or annotate the problem, and the problem description can be received by the system as feedback information. Based on this feedback information, the system can adjust the subsequent dependency configuration generation process, such as strengthening specific constraints in subsequent generation processes or avoiding the repeated generation of the same type of configuration result.

[0085] In the embodiments of this application, the verification results and user feedback information can also be used to update the prompt word specification information. For example, when multiple verification results indicate that the same type of problem occurs repeatedly, the system can adjust the prompt word specification information based on this type of problem, so that the prompt word specification can more clearly express the relevant constraints in the subsequent generation process. In this way, verification feedback can continuously act on the prompt word specification generation and dependency configuration generation process, forming a closed-loop optimization mechanism in which generation, verification, and feedback interact.

[0086] According to embodiments of this application, by introducing pre-generated prompt word specification information that can be dynamically adjusted with rule updates, the dependency configuration generation process is always constrained by the latest configuration rules and global job dependencies, avoiding inconsistencies or inaccuracies in the generated results due to rule changes. Furthermore, the prompt word specification information, dependency configuration requests, and dependency relationship data associated with the target batch job are jointly input into a large language model, enabling the generation process to simultaneously integrate user intent, rule constraints, and the global dependency structure, thereby automatically generating dependency configuration results that better suit the actual scheduling environment. Finally, the generated dependency configuration results are validated, and dependency configuration operations are executed only if the validation passes, effectively reducing the risk of unreasonable dependencies entering the scheduling system. Overall, the above steps work together to automate, standardize, and controllable batch job dependency configuration, significantly reducing manual maintenance costs, improving the accuracy and consistency of dependency configuration, and enhancing the stability and operational efficiency of the batch job scheduling system in complex scenarios.

[0087] The following describes the batch job dependency configuration method of this application embodiment in detail by way of preferred embodiments.

[0088] In the embodiments of this application, when the batch job depends on changes in configuration rule information, the system can perform corresponding update processing on the prompt word specification information. Specifically, the system first obtains the rule update content, which may include, for example, newly added configuration constraints, adjustments to existing constraints, or deletion of some rules. To facilitate subsequent processing, the system can perform structured parsing of the update content, breaking down the rule update content into several rule change items, thereby forming a set of rule change items, so that different types of rule changes can be expressed in a unified data structure.

[0089] After obtaining the set of rule changes, the system can determine the specification differences based on the correspondence between the set of rule changes and the specification clauses in the prompt word specification information. This correspondence is used to characterize the association between each specification clause and a specific configuration rule, enabling the system to identify which specification clauses are affected by the rule changes.

[0090] In the embodiments of this application, the specification difference information can be used to describe the changes that need to occur in the prompt word specification information before and after the rule update, such as new specification clauses, clauses that need to be modified, or specification clauses that need to be removed. Based on this specification difference information, the system can perform incremental updates on the prompt word specification information, adjusting only the affected parts, thereby maintaining the stability of the overall structure of the prompt word specification information.

[0091] In this way, the prompt word specification information can continuously evolve as the batch job dependency configuration rules change, avoiding the problem of specification content lagging or inconsistent due to rule updates. At the same time, the incremental update mechanism can reduce the complexity of specification maintenance and improve update efficiency.

[0092] In the embodiments of this application, when updating the prompt word specification information based on the rule change item set, the process of determining specification difference information can be further refined. Specifically, the system can first analyze the scope of impact of rule changes based on the rule change item set. The scope of impact of rule changes is used to characterize the dependency configuration fields, dependency relationship types, or batch job types that may be affected by rule changes, thereby providing clear boundaries for the subsequent positioning of specification clauses.

[0093] After determining the scope of the rule change's impact, the system can combine the correspondence between each clause in the prompt message specification information and the configuration rules to filter out a subset of target specification clauses related to the scope of the rule change's impact from the prompt message specification information. In this way, clauses related to the current rule change can be accurately located from the overall specification content, avoiding unnecessary adjustments to irrelevant specification clauses.

[0094] In the embodiments of this application, after obtaining the target subset of regulatory clauses, the system can further analyze the target subset of regulatory clauses in conjunction with the scope of influence of rule changes to determine regulatory difference information. Regulatory difference information may include operation type information and clause content information. The operation type information is used to indicate operations such as adding, modifying, or deleting from the target subset of regulatory clauses, while the clause content information describes the specific content or adjustment direction of the corresponding regulatory clauses after the update.

[0095] Through the above methods, specification difference information can reflect the specific impact of rule changes on prompt word specification information in a structured form, enabling the system to update specification clauses in a targeted manner based on clear operation instructions. This mechanism not only improves the controllability and traceability of the prompt word specification update process, but also reduces the maintenance costs caused by overall specification reconstruction.

[0096] In the embodiments of this application, the generation of prompt word specification information can be combined with batch job dependency configuration rule information and batch job dependency relationship data.

[0097] Figure 3 The flowchart illustrating a method for generating prompt word specification information according to some exemplary embodiments of this application is shown schematically.

[0098] like Figure 3 As shown, the method for generating the prompt word specification information includes operations S310 to S330.

[0099] In operation S310, configuration constraint information is extracted from the batch job dependency configuration rule information to form a set of specification clauses. The configuration constraint information includes output field set, field value constraint, job type constraint, dependency constraint and / or format constraint.

[0100] In embodiments of this application, the system can first extract configuration constraint information related to dependency configuration from batch job dependency configuration rule information, and form a set of specification clauses based on the configuration constraint information. The configuration constraint information can be used to limit the expression of dependency configuration results at the field level and structure level, such as specifying the set of fields to be output in the dependency configuration results, the allowed value range of each field, the dependency relationships between different job types, the direction and quantity restrictions of dependency relationships, and the format requirements of the overall configuration content, etc.

[0101] In operation S320, job definition information and dependency structure information are extracted from batch job dependency data to generate a structured job context.

[0102] In embodiments of this application, the system can also extract job definition information and dependency structure information from batch job dependency data to generate a structured job context. For example, job definition information may include the basic attributes of the batch job, its business theme, scheduling cycle, and job type; dependency structure information can be used to describe the existing or historically formed upstream and downstream relationships and dependency link structures between multiple batch jobs.

[0103] In operation S330, the set of specification terms and the structured job context are combined to generate prompt word specification information.

[0104] In the embodiments of this application, after obtaining the set of specification clauses and the structured job context, the system can combine the two to generate prompt word specification information. This combination process can use the specification clauses as binding descriptions and the structured job context as background descriptions, so that the generated prompt word specification information has both rule constraint attributes and context guidance attributes. The prompt word specification information formed in this way provides a unified and stable constraint foundation for the subsequent dependency configuration generation process, which is conducive to improving the consistency between the generated results and the rule requirements and global dependencies.

[0105] In the embodiments of this application, the prompt word specification information can be generated or supplemented based on other data related to batch job dependency configuration, in addition to being generated based on batch job dependency configuration rule information and batch job dependency relationship data.

[0106] For example, the prompt word specification information can be generated based on historical dependency configuration data. Historical dependency configuration data can reflect dependency configuration results that have been effective or verified in different business scenarios. The system can extract common dependency patterns, stable dependency chain structures, and frequently occurring configuration combinations from historical dependency configuration data, and transform this information into specification terms or constraint prompts, incorporating them into the prompt word specification information, thereby guiding the generated results towards mature and stable configuration patterns in the subsequent generation process.

[0107] For example, the prompt word specification information can also be generated based on job execution characteristic data. Job execution characteristic data may include the execution duration distribution, execution frequency, failure retry status, and scheduling window characteristics of batch jobs. By analyzing these execution characteristics, the system can identify which types of jobs are more suitable for establishing dependencies and which jobs should remain independent in scheduling, and reflect the corresponding constraints or guidance information in the prompt word specification to improve the rationality of the generated dependency configuration results at the scheduling level.

[0108] In some embodiments, prompt word specification information can also be generated in conjunction with business process description data. Business process description data can be used to characterize the sequential relationship or logical association between business processing steps, such as the connection order between business stages such as data collection, data cleaning, indicator calculation, and result output. By incorporating business process description data into prompt word specification information, the system can better reflect the inherent logic of the business process when generating dependency configuration results, ensuring that the generated dependencies are consistent with the actual business processing order.

[0109] In the embodiments of this application, before inputting the prompt word specification information, dependency configuration requests, and target batch job dependency data associated with the target batch job into the large language model, the system can filter and organize the dependency data related to the target batch job. Specifically, candidate upstream job data and candidate downstream job data associated with the target batch job can be determined based on the batch job dependency data. The candidate upstream job data is used to characterize the set of batch jobs that may provide input data or prerequisite results for the target batch job, and the candidate downstream job data is used to characterize the set of batch jobs that may depend on the execution results of the target batch job.

[0110] After obtaining candidate upstream and downstream job data, the system can construct target batch job dependency data based on this candidate data. Target batch job dependency data is used to characterize the local dependency structure of the target batch jobs within the current job hierarchy, enabling subsequent generation processes to focus on dependencies directly or indirectly related to the target batch jobs without processing all global dependency data, thereby reducing processing complexity.

[0111] In embodiments of this application, the system can convert the constructed target batch job dependency data into a structured context. The structured context can adopt a unified data organization format to clearly express job identifiers, job types, dependency directions, and relationships. In some embodiments, the structured context can be arranged according to a predefined paragraph structure in the prompt word specification information, ensuring a stable and recognizable organization of the context content in the model input.

[0112] In the embodiments of this application, by arranging the structured context according to the paragraph structure specified by the prompt word specification information, the dependency background information and rule constraint information of the target batch task can be orderly combined to form a clear context description. This arrangement result helps the large language model quickly distinguish different types of information when processing input content, and improves the consistency of understanding of dependency background and constraint conditions.

[0113] In some embodiments, after obtaining the orchestration result, the system can combine the orchestration result with prompt word specification information and dependency configuration requests to form the model input. The model input can simultaneously include a description of the dependency configuration task, a constraint description of the generated result, and contextual information of the target batch job in the dependency system, enabling the large language model to comprehensively consider multiple factors in a unified input.

[0114] In the embodiments of this application, the system provides the generated model input to a large language model, which then generates dependency configuration results based on semantic understanding and contextual analysis of the input content. Through the above processing flow, the dependency configuration results generated by the large language model can reflect the reasonable dependencies between the target batch job and related jobs under the combined effect of rule constraints and dependency context. While enhancing the flexibility of generation, it also improves the stability and interpretability of dependency configuration results, reduces the probability of generation bias, and improves the accuracy and generation efficiency of batch job dependency configuration.

[0115] In embodiments of this application, when determining candidate upstream and downstream job data associated with a target batch job, a global dependency graph can be constructed based on the batch job dependency data. The global dependency graph is used to depict the existing dependencies between multiple batch jobs in a graph structure, where jobs are nodes and dependencies are connections between nodes, thereby enabling an understanding of the dependency topology between batch jobs from a holistic perspective.

[0116] In some embodiments, the system can determine a neighborhood subgraph with a preset hop count associated with the job node corresponding to the target batch job in the global dependency graph. The preset hop count is used to limit the hierarchical range of expansion upstream or downstream in the dependency graph, for example, only considering job nodes that have direct or limited-level indirect dependencies with the target batch job. By introducing a neighborhood subgraph with a preset hop count, the system can ensure that dependency analysis covers relevant jobs while avoiding including job nodes with low correlation to the target batch job in the candidate range, thereby reducing the complexity of subsequent processing.

[0117] In the embodiments of this application, after obtaining a neighborhood subgraph with a preset number of hops, the system can further filter job nodes in the neighborhood subgraph to form a candidate job set. During the filtering process, job nodes can be judged based on data output identifier matching operations and scheduling window matching operations. The data output identifier matching operation is used to determine whether the data identifier produced by the job node is related to the data identifier required by the target batch job; the scheduling window matching operation is used to determine whether the execution time window of the job node has a reasonable connection relationship with the scheduling window of the target batch job.

[0118] The candidate job set obtained after screening can serve as the basis for determining candidate upstream and downstream job data. The system can determine whether a job node in the candidate job set belongs to the upstream or downstream job of the target batch job based on its relative position in the global dependency graph, and accordingly generate candidate upstream and downstream job data. In this way, under the constraints of global dependencies, the system can determine the range of candidate jobs highly correlated with the target batch job by combining data output characteristics and scheduling characteristics, thus providing a more accurate contextual basis for subsequent dependency configuration generation.

[0119] In embodiments of this application, the verification of dependency configuration results may include the detection of circular dependencies. Specifically, the system can construct a target dependency graph based on the generated dependency configuration results to depict the batch jobs involved in the dependency configuration results and their dependencies. This target dependency graph can use jobs as nodes and dependencies as connections, thereby expressing the generated dependency configuration state in a graph structure.

[0120] In some embodiments, after constructing the target dependency graph, the system can perform a cycle detection operation on the target dependency graph. The cycle detection operation is used to determine whether there are closed dependency paths in the target dependency graph, that is, dependency links formed between multiple jobs. By traversing and analyzing the target dependency graph, the system can obtain circular dependency detection results, which are used to characterize whether circular dependencies exist and their specific manifestations.

[0121] In the embodiments of this application, the system can determine the verification result based on the circular dependency detection result. When the circular dependency detection result indicates that there is no circular dependency, the verification result can be determined as passing; when the circular dependency detection result indicates that there is a circular dependency, the verification result can be determined as failing. In the case where the verification result indicates failing, the system can also generate corresponding location information.

[0122] In some embodiments, the location information may include job chain segments that cause circular dependencies. Job chain segments describe the job nodes involved in the circular dependency and their dependency order, enabling relevant personnel to intuitively understand the formation path of the circular dependency. By generating location information containing job chain segments, the system can provide clear guidance for subsequent processing, helping to quickly locate and understand circular dependency problems, and providing a reference for adjusting dependency configurations or optimizing generation logic.

[0123] It should be noted that the above-described detection of circular dependencies is only one implementation method for verifying dependency configuration results, used to exemplify the specific form of the verification process. The embodiments of this application do not limit the verification process to only including circular dependency detection. In practical applications, other verification methods can be combined to comprehensively judge the dependency configuration results. For example, the completeness and consistency of fields in the dependency configuration results can be verified, the legality of dependencies between different job types can be judged, the number of dependencies, dependency levels, or execution order can be verified to conform to configuration rules, or the consistency between generated dependencies and existing dependency configurations can be compared, etc.

[0124] In the embodiments of this application, if the verification result indicates that the verification is successful, the dependency configuration operation can be performed. Specifically, the system can first convert the generated dependency configuration result into structured configuration data, so that the dependency relationship is expressed in a unified data structure form, which facilitates subsequent processing and storage. In this process, the system can format the structured configuration data according to the batch job dependency configuration rule information, such as organizing the field order, field completeness, and configuration expression form, thereby obtaining target configuration data that meets the configuration specification requirements.

[0125] After obtaining the target configuration data, the system can generate a corresponding configuration version identifier based on the target configuration data. The configuration version identifier is used to distinguish the dependency configuration results formed at different times or in different generation processes, so that each dependency configuration operation has independent version identification information. By introducing the configuration version identifier, changes to dependency configurations can be tracked in subsequent configuration management or operation and maintenance processes, enhancing the controllability of configuration management.

[0126] In the embodiments of this application, after generating the configuration version identifier, the system can write the target configuration data carrying the configuration version identifier into the batch job dependency configuration storage, or submit the target configuration data to the batch job dependency configuration system. Through this method, the generated and verified dependency configuration results can be formally incorporated into the batch job scheduling environment as the basis for subsequent scheduling and execution. This execution process achieves a seamless connection from the generation of dependency configuration results to the actual configuration implementation, forming a complete closed loop in the dependency configuration process.

[0127] Corresponding to the above-described batch job dependency configuration method, embodiments of this application also provide a batch job dependency configuration system and a batch job dependency configuration device.

[0128] Figure 4 The illustration shows a schematic diagram of a batch job dependency configuration system according to some exemplary embodiments of this application.

[0129] like Figure 4 As shown, the batch job dependency configuration system may include components such as a prompt word specification generation module, a dependency configuration generation module, a configuration verification and feedback module, and a batch job dependency configuration system. These components form a collaborative working relationship through data flow and control flow.

[0130] In this system, users can input dependency configuration requests for target batch jobs. The prompt word specification generation module is connected to the batch dependency configuration specification and the batch configuration database to generate or update prompt word specification information. This module can adjust existing prompt word specifications based on rule updates when dependency configuration rules change, ensuring that the specification information used in subsequent generation processes remains consistent with the current rules.

[0131] The dependency configuration generation module receives dependency configuration requests and prompt word specification information from the prompt word specification generation unit. It then combines this information with batch job dependency relationship data to generate the dependency configuration results corresponding to the target batch job. The dependency configuration generation module can output the generated dependency configuration results to the configuration verification and feedback module.

[0132] The configuration verification feedback module is used to verify the dependency configuration results output by the dependency configuration generation module. The verification process can be based on preset dependency configuration rules to judge the legality and consistency of the generated results, and generate corresponding verification results and location information if the verification fails. The configuration verification feedback module can also provide verification results to the user for viewing, and if the verification passes, transmit the dependency configuration results to the batch job dependency configuration system.

[0133] The configuration verification feedback module can also automatically determine the dependency configuration results based on other batch dependency rules, ensuring the rationality and consistency of the configuration results from multiple dimensions. An example is shown below:

[0134] Table 1 Examples of Judgment Rules

[0135]

[0136] The batch job dependency configuration system receives validated dependency configuration results and writes them to the corresponding configuration storage or uses them for subsequent batch job scheduling and execution. Through the collaborative work of the above units, the system realizes a complete process from user dependency configuration request input, prompt word specification generation, automatic dependency configuration generation, validation feedback to final configuration implementation.

[0137] Figure 5 A schematic block diagram of a batch job dependency configuration apparatus according to an embodiment of this application is shown.

[0138] like Figure 5 As shown, the batch job dependency configuration device 500 of this embodiment includes a data acquisition module 510, a prompt word acquisition module 520, a dependency configuration result generation module 530, and a verification module 540.

[0139] The data acquisition module 510 can be used to acquire the user's dependency configuration request for the target batch job. In one embodiment, the data acquisition module 510 can be used to perform the operation S210 described above, which will not be repeated here.

[0140] The prompt word acquisition module 520 can be used to acquire pre-generated prompt word specification information. This prompt word specification information is generated at least based on batch job dependency configuration rule information and batch job dependency relationship data. The batch job dependency relationship data is used to characterize the global dependency relationships between multiple batch jobs. The prompt word specification information is updated based on the updated content of the batch job dependency configuration rule information. In one embodiment, the prompt word acquisition module 520 can be used to perform the operation S220 described above, which will not be repeated here.

[0141] The dependency configuration result generation module 530 can be used to input prompt word specification information, dependency configuration requests, and target batch job dependency relationship data associated with the target batch job into the large language model to generate dependency configuration results. In one embodiment, the dependency configuration result generation module 530 can be used to perform the operation S230 described above, which will not be repeated here.

[0142] The verification module 540 can be used to verify the dependency configuration result. If the verification result indicates that the verification is successful, the dependency configuration operation is performed. In one embodiment, the verification module 540 can be used to perform the operation S240 described above, which will not be repeated here.

[0143] According to an embodiment of this application, the prompt word acquisition module 520 can also be used to acquire updated content, perform structured parsing on the updated content to obtain a set of rule change items; determine the specification difference information based on the correspondence between the set of rule change items and the specification clauses in the prompt word specification information; and perform incremental updates on the prompt word specification information based on the specification difference information.

[0144] According to an embodiment of this application, the prompt word acquisition module 520 can also be used to determine the scope of influence of rule changes based on the set of rule change items, where the scope of influence of rule changes is used to represent the affected dependency configuration fields, dependency types, or job types; determine a subset of target specification clauses in the prompt word specification information based on the correspondence between the scope of influence of rule changes and specification clauses; and determine specification difference information based on the subset of target specification clauses and the scope of influence of rule changes, whereby the specification difference information includes operation type information and clause content information, and the operation type information is used to indicate whether to perform add, modify, or delete operations on the subset of target specification clauses.

[0145] According to an embodiment of this application, the prompt word acquisition module 520 can also be used to extract configuration constraint information from batch job dependency configuration rule information to form a set of specification clauses, wherein the configuration constraint information includes a set of output fields, field value constraints, job type constraints, dependency constraints and / or format constraints; extract job definition information and dependency structure information from batch job dependency data to generate a structured job context; and combine the set of specification clauses and the structured job context to generate prompt word specification information.

[0146] According to an embodiment of this application, the dependency configuration result generation module 530 can also be used to determine candidate upstream job data and / or candidate downstream job data associated with the target batch job based on batch job dependency relationship data; construct target batch job dependency relationship data based on candidate upstream job data and / or candidate downstream job data; convert the target batch job dependency relationship data into a structured context; arrange the structured context according to the paragraph structure specified by the prompt word specification information to obtain an arrangement result; and combine the arrangement result, prompt word specification information, and dependency configuration request to obtain model input, input the model input into the large language model, and generate dependency configuration results.

[0147] According to an embodiment of this application, the dependency configuration result generation module 530 can also be used to obtain a global dependency graph constructed based on the batch job dependency data, determine a preset hop count neighborhood subgraph associated with the target batch job based on the global dependency graph; filter job nodes in the preset hop count neighborhood subgraph based on data output identifier matching operation and / or scheduling window matching operation to obtain a candidate job set; and determine candidate upstream job data and / or candidate downstream job data based on the candidate job set.

[0148] According to an embodiment of this application, the verification module 540 can also be used to construct a target dependency graph based on the dependency configuration result, perform loop detection on the target dependency graph to obtain a circular dependency detection result; and determine the verification result based on the circular dependency detection result, wherein, if the verification result indicates that the verification fails, location information is generated, and the location information includes the job link segment that caused the circular dependency.

[0149] According to an embodiment of this application, the verification module 540 can also be used to convert the dependency configuration result into structured configuration data, format the structured configuration data based on the batch job dependency configuration rule information to obtain target configuration data; generate a configuration version identifier based on the target configuration data; and write the configuration version identifier and the target configuration data into the batch job dependency configuration storage or submit them to the batch job dependency configuration system.

[0150] According to embodiments of this application, any multiple modules among the data acquisition module 510, prompt word acquisition module 520, dependency configuration result generation module 530, and verification module 540 can be merged into one module, or any one of these modules can be split into multiple modules. Alternatively, at least some of the functions of one or more of these modules can be combined with at least some of the functions of other modules and implemented in one module. According to embodiments of this application, at least one of the data acquisition module 510, prompt word acquisition module 520, dependency configuration result generation module 530, and verification module 540 can be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or implemented in hardware or firmware by any other reasonable means of integrating or packaging the circuitry, or implemented in software, hardware, or firmware, or in any suitable combination of any of these three implementation methods. Alternatively, at least one of the data acquisition module 510, the prompt word acquisition module 520, the dependency configuration result generation module 530, and the verification module 540 can be implemented at least partially as a computer program module, which can perform corresponding functions when the computer program module is run.

[0151] Figure 6 A block diagram of an electronic device suitable for implementing a batch operation dependency configuration method according to an embodiment of this application is illustrated schematically.

[0152] like Figure 6 As shown, an electronic device 600 according to an embodiment of this application includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage portion 608 into a random access memory (RAM) 603. The processor 601 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 601 may also include onboard memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of this application.

[0153] RAM 603 stores various programs and data required for the operation of electronic device 600. Processor 601, ROM 602, and RAM 603 are interconnected via bus 604. Processor 601 executes various operations of the method flow according to embodiments of this application by executing programs in ROM 602 and / or RAM 603. It should be noted that the programs may also be stored in one or more memories other than ROM 602 and RAM 603. Processor 601 may also execute various operations of the method flow according to embodiments of this application by executing programs stored in said one or more memories.

[0154] According to embodiments of this application, the electronic device 600 may further include an input / output (I / O) interface 605, which is also connected to a bus 604. The electronic device 600 may also include one or more of the following components connected to the input / output (I / O) interface 605: an input section 606 including a keyboard, mouse, etc.; an output section 607 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the input / output (I / O) interface 605 as needed. A removable medium 611, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 610 as needed so that computer programs read from it can be installed into the storage section 608 as needed.

[0155] This application also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of this application.

[0156] According to embodiments of this application, the computer-readable storage medium can be a non-volatile computer-readable storage medium, such as including but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of this application, the computer-readable storage medium may include ROM 602 and / or RAM 603 and / or one or more memories other than ROM 602 and RAM 603 described above.

[0157] Embodiments of this application also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code is used to enable the computer system to implement the batch job dependency configuration method provided in the embodiments of this application.

[0158] When the computer program is executed by the processor 601, it performs the functions defined in the system / apparatus of this application embodiment. According to the embodiments of this application, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0159] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and downloaded and installed via the communication section 609, and / or installed from the removable medium 611. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0160] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 609, and / or installed from the removable medium 611. When the computer program is executed by the processor 601, it performs the functions defined in the system of this application embodiment. According to the embodiments of this application, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0161] According to embodiments of this application, program code for executing the computer programs provided in the embodiments of this application can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0162] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0163] Those skilled in the art will understand that the features described in the various embodiments of this application can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in this application. In particular, the features described in the various embodiments of this application can be combined and / or combined in various ways without departing from the spirit and teachings of this application. All such combinations and / or combinations fall within the scope of this application.

[0164] The embodiments of this application have been described above. However, these embodiments are merely illustrative and not intended to limit the scope of this application. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. Without departing from the scope of this application, those skilled in the art can make various substitutions and modifications, all of which should fall within the scope of this application.

Claims

1. A dependency configuration method for batch jobs, characterized in that, The method comprises: obtaining a user's dependency configuration request for a target batch job; obtaining pre-generated prompt word regulation information, wherein the prompt word regulation information is generated based at least on batch job dependency configuration rule information and batch job dependency relationship data, the batch job dependency relationship data is used to represent the global dependency relationship between a plurality of batch jobs; the prompt word regulation information is updated based on the update content of the batch job dependency configuration rule information; inputting the prompt word regulation information, the dependency configuration request and the target batch job dependency relationship data associated with the target batch job into a large language model to generate a dependency configuration result; and verifying the dependency configuration result, and executing a dependency configuration operation if the verification result indicates that the verification is passed.

2. The method of claim 1, wherein, The prompt word regulation information is updated based on the update content of the batch job dependency configuration rule information, comprising: obtaining update content, structurally analyzing the update content to obtain a rule change item set; determining regulation difference information based on the correspondence between the rule change item set and the regulation clauses in the prompt word regulation information; and performing incremental update on the prompt word regulation information based on the regulation difference information.

3. The method of claim 2, wherein, The determination of the regulation difference information based on the correspondence between the rule change item set and the regulation clauses in the prompt word regulation information comprises: determining a rule change impact range based on the rule change item set, the rule change impact range being used to represent the affected dependency configuration field, dependency type or job type; determining a target regulation clause subset in the prompt word regulation information based on the correspondence between the rule change impact range and the regulation clauses; and determining the regulation difference information based on the target regulation clause subset and the rule change impact range, the regulation difference information comprising operation type information and clause content information, the operation type information being used to indicate that the target regulation clause subset is executed with an adding, modifying or deleting operation.

4. The method of claim 1, wherein, The prompt word regulation information is generated based at least on batch job dependency configuration rule information and batch job dependency relationship data, comprising: extracting configuration constraint information from the batch job dependency configuration rule information to form a regulation clause set, wherein the configuration constraint information comprises an output field set, field value constraint, job type constraint, dependency relationship constraint and / or format constraint; extracting job definition information and dependency relationship structure information from the batch job dependency relationship data to generate a structured job context; and combining the regulation clause set and the structured job context to generate the prompt word regulation information.

5. The method of claim 1, wherein, The inputting of the prompt word regulation information, the dependency configuration request and the target batch job dependency relationship data associated with the target batch job into a large language model to generate a dependency configuration result comprises: based on the batch job dependency relationship data, determining candidate upstream job data and / or candidate downstream job data associated with the target batch job; construct the target batch job dependency relationship data based on the candidate upstream job data and / or the candidate downstream job data; convert the target batch job dependency relationship data into a structured context, and arrange the structured context according to a paragraph structure specified by the prompt word specification information to obtain an arrangement result; and combine the arrangement result, the prompt word specification information, and the dependency configuration request to obtain a model input, input the model input into the large language model, and generate the dependency configuration result.

6. The method of claim 5, wherein, The determination of the candidate upstream job data and / or the candidate downstream job data associated with the target batch job comprises: obtaining a global dependency relationship graph constructed based on the batch job dependency relationship data, and determining a preset hop number neighborhood subgraph associated with the target batch job based on the global dependency relationship graph; in the preset hop number neighborhood subgraph, filtering job nodes based on data output identification matching operations and / or scheduling window matching operations to obtain a candidate job set; and determining the candidate upstream job data and / or the candidate downstream job data based on the candidate job set.

7. The method of claim 1, wherein, The verification of the dependency configuration result comprises cyclic dependency detection, specifically comprising: constructing a target dependency relationship graph based on the dependency configuration result, performing ring detection on the target dependency relationship graph to obtain a cyclic dependency detection result; and determining the verification result based on the cyclic dependency detection result, wherein in the case that the verification result represents that the verification fails, positioning information is generated, and the positioning information comprises a job link segment causing the cyclic dependency.

8. The method of claim 1, wherein, The execution of the dependency configuration operation comprises: converting the dependency configuration result into structured configuration data, performing format processing on the structured configuration data based on the batch job dependency configuration rule information to obtain target configuration data; generating a configuration version identifier based on the target configuration data; and writing the configuration version identifier and the target configuration data into a batch job dependency configuration storage or submitting them to a batch job dependency configuration system.

9. A device for dependency configuration of batch jobs, characterized in that The apparatus comprises: a data acquisition module configured to acquire a dependency configuration request of a user for a target batch job; a prompt word acquisition module configured to acquire pre-generated prompt word specification information, wherein the prompt word specification information is generated based on at least batch job dependency configuration rule information and batch job dependency relationship data, the batch job dependency relationship data is used to represent a global dependency relationship between a plurality of batch jobs, and the prompt word specification information is updated based on updated content of the batch job dependency configuration rule information; a dependency configuration result generation module configured to input the prompt word specification information, the dependency configuration request, and target batch job dependency relationship data associated with the target batch job into a large language model to generate a dependency configuration result; and a verification module configured to verify the dependency configuration result, and execute a dependency configuration operation in the case that a verification result represents that the verification passes.

10. An electronic device comprising: one or more processors; a memory for storing one or more computer programs, characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 8.

11. A computer readable storage medium having stored thereon a computer program or instructions, characterized in that, The computer program or instructions, when executed by a processor, implement the steps of the method according to any one of claims 1 to 8.

12. A computer program product comprising computer programs or instructions, characterized in that, The computer program or instructions, when executed by a processor, implement the steps of the method according to any one of claims 1 to 8.

Images