Task processing method and device, electronic equipment and medium

Abstract

The invention discloses a task processing method and device, electronic equipment and a medium, and the method comprises the steps: responding to an obtained to-be-processed task set, classifying the to-be-processed task set, and obtaining a plurality of to-be-processed task subsets; target processing is executed for each to-be-processed task subset so as to complete processing of the to-be-processed task set, and the target processing comprises the steps that according to the service relation of each to-be-processed task in the to-be-processed task subset, multiple to-be-processed tasks in the to-be-processed task subset are recombined, and to-be-processed task units are obtained, the business relationship is used for representing relevance between the to-be-processed tasks; and processing the to-be-processed task in the to-be-processed task unit. The method can be applied to the business fields of financial science and technology, medical health, old-age care and the like, and the task processing efficiency is improved in the mode that the to-be-processed task set is classified and recombined to form the to-be-processed task units.

Description

Technical Field

[0001] The present application relates to the field of task processing, and in particular to a task processing method, device, electronic device and medium. Background Art

[0002] With the rapid development of information technology, various business systems are increasingly demanding task processing capabilities. In enterprise-level systems, distributed computing platforms, and big data processing scenarios, task scheduling and execution have become key factors in system efficiency and stability.

[0003] Traditional task processing methods cannot fully utilize the parallel processing capabilities of multi-node clusters, resulting in low utilization of computing resources such as CPU and memory. The task chain execution time increases linearly with the number of tasks. When faced with high-concurrency and high-throughput business scenarios, task processing efficiency is low. Summary of the Invention

[0004] The main purpose of the embodiments of the present application is to provide a task processing method, device, electronic device and medium, aiming to solve the problem of low task processing efficiency of existing task processing methods.

[0005] To achieve the above objectives, a first aspect of an embodiment of the present application provides a task processing method, the method comprising:

[0006] In response to the acquired set of pending tasks, classify the set of pending tasks to obtain a plurality of subsets of pending tasks;

[0007] Target processing is performed on each subset of tasks to be processed to complete processing of the set of tasks to be processed, wherein the target processing includes:

[0008] reorganizing the plurality of pending tasks in the pending task subset according to the business relationship of each pending task in the pending task subset to obtain a pending task unit, wherein the business relationship is used to characterize the relevance between the pending tasks;

[0009] Processing the pending tasks in the pending task unit.

[0010] In some implementations, processing the pending tasks in the pending task unit includes:

[0011] Determining target resource requirement information of the task unit to be processed, where the target resource requirement information is used to indicate the computing resources required to process the task unit to be processed;

[0012] Based on the target resource requirement information, the pending tasks in the pending task unit are processed.

[0013] In some embodiments, the processing the to-be-processed task in the to-be-processed task unit based on the target resource requirement information comprises:

[0014] obtaining node resource information of a plurality of task processing nodes associated with the to-be-processed task set, the node resource information being used to indicate available computing resources of the task processing nodes;

[0015] determining a target node from the plurality of task processing nodes according to the target resource requirement information of the to-be-processed task unit and the node resource information of each of the task processing nodes;

[0016] processing the to-be-processed task in the to-be-processed task unit by using the target node.

[0017] In some embodiments, the determining the target node from the plurality of task processing nodes according to the target resource requirement information of the to-be-processed task unit and the node resource information of each of the task processing nodes comprises:

[0018] determining availability of each of the task processing nodes according to the computing resources indicated by the node resource information of each of the task processing nodes;

[0019] screening the plurality of task processing nodes according to the availability of each of the task processing nodes to obtain a first candidate node set;

[0020] for each of the to-be-processed task units, performing: calculating a matching degree between the to-be-processed task unit and each of the task processing nodes in the first candidate node set according to the resource requirement of the to-be-processed task unit and the computing resources indicated by the node resource information of each of the task processing nodes in the first candidate node set, the matching degree being used to represent a matching degree of the available computing resources in the corresponding node and the required computing resources of the to-be-processed task unit;

[0021] taking the task processing node with the highest matching degree as the target node of the to-be-processed task unit.

[0022] In some embodiments, the classifying the to-be-processed task set to obtain a plurality of to-be-processed task subsets in response to the obtained to-be-processed task set comprises:

[0023] obtaining the to-be-processed task set, the to-be-processed task set comprising a plurality of to-be-processed tasks;

[0024] respectively extracting features of each of the to-be-processed tasks to obtain a business feature corresponding to each of the to-be-processed tasks, the business feature being used to indicate a business type to which the to-be-processed task belongs;

[0025] The set of tasks to be processed is classified according to the business characteristics corresponding to each of the tasks to be processed to obtain a plurality of subsets of the tasks to be processed.

[0026] In some embodiments, after reorganizing the plurality of tasks to be processed in the subset of tasks to be processed according to the business relationship of each task to be processed in the subset of tasks to be processed to obtain a task unit to be processed, the method includes:

[0027] Obtaining the urgency level of each pending task in the pending task unit;

[0028] Inputting the urgency of the tasks to be processed and the first resource requirement information corresponding to each task to be processed into a preset weight distribution model, and outputting the weight corresponding to each task to be processed by the weight distribution model;

[0029] The processing order of the tasks to be processed in the task unit to be processed is updated according to the weight corresponding to each task to be processed, to obtain an updated task unit to be processed.

[0030] In some embodiments, performing target processing on each subset of tasks to be processed to complete processing of the set of tasks to be processed includes:

[0031] Controlling a plurality of the subsets of tasks to be processed to execute the target processing in parallel to complete the processing of the set of tasks to be processed;

[0032] The processing of the pending tasks in the pending task unit includes:

[0033] According to the preset execution logic corresponding to the task unit to be processed, multiple tasks to be processed in the task unit to be processed are serially processed in sequence to complete the processing of the task unit to be processed.

[0034] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application provides a task processing device, the device comprising:

[0035] a classification module, configured to classify the acquired set of tasks to be processed in response to the acquired set of tasks to be processed, and obtain a plurality of subsets of tasks to be processed;

[0036] A processing module is configured to perform target processing on each subset of tasks to be processed to complete processing of the set of tasks to be processed, wherein the target processing includes:

[0037] reorganizing the plurality of pending tasks in the pending task subset according to the business relationship of each pending task in the pending task subset to obtain a pending task unit, wherein the business relationship is used to characterize the relevance between the pending tasks;

[0038] Processing the pending tasks in the pending task unit.

[0039] To achieve the above-mentioned purpose, the third aspect of an embodiment of the present application proposes an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the task processing method described in the first aspect above.

[0040] To achieve the above objectives, the fourth aspect of an embodiment of the present application proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the task processing method described in the first aspect above.

[0041] To achieve the above objectives, an embodiment of the present application may provide a computer program product for implementation. When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device implements the task processing method described in the first aspect above.

[0042] The task processing method, device, electronic device and medium proposed in the present application, in response to the acquired set of pending tasks, classify the set of pending tasks to obtain multiple subsets of pending tasks, and divide a number of pending tasks into multiple subsets of pending tasks through classification operations, providing a structured basis for subsequent processing and improving processing efficiency; performing target processing on each subset of pending tasks to complete the processing of the set of pending tasks, the target processing includes: reorganizing the multiple pending tasks in the subset of pending tasks according to the business relationship of each pending task in the subset of pending tasks to obtain a pending task unit, and the business relationship is used to characterize the correlation between the pending tasks; processing the pending tasks in the task unit; reorganizing according to the business relationship between the pending tasks to form a pending task unit, so that task processing is more in line with actual business logic, enhancing the consistency and accuracy of processing, thereby realizing accurate scheduling and efficient utilization of computing resources, avoiding resource waste or conflict, and improving the overall task processing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a flowchart of a task processing method provided in an embodiment of the present application;

[0044] Figure 2 yes Figure 1 Schematic diagram of the process in step S102;

[0045] Figure 3 yes Figure 1 Schematic diagram of the process in step S203;

[0046] Figure 4 yes Figure 1 Flowchart of step S101 in FIG.

[0047] Figure 5 is a structural diagram of a task processing device provided in an embodiment of the present application;

[0048] Figure 6 This is a schematic diagram of the hardware structure of the electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0050] It should be noted that although the device schematics illustrate functional module divisions and the flowcharts illustrate logical sequences, in certain circumstances, the steps shown or described may be performed in a sequence that differs from the module divisions in the device or the sequence in the flowcharts. The terms "first," "second," and so on, in the specification, claims, and drawings, are used to distinguish similar items and are not necessarily used to describe a specific sequence or precedence.

[0051] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein are for the purpose of describing the embodiments of this application only and are not intended to limit this application.

[0052] With the rapid development of information technology, various business systems are increasingly demanding task processing capabilities. In enterprise-level systems, distributed computing platforms, and big data processing scenarios, task scheduling and execution have become key factors in system efficiency and stability.

[0053] Traditional task processing methods cannot fully utilize the parallel processing capabilities of multi-node clusters, resulting in low utilization of computing resources such as CPU and memory. The task chain execution time increases linearly with the number of tasks. When faced with high-concurrency and high-throughput business scenarios, task processing efficiency is low.

[0054] Based on this, the embodiments of the present application provide a task processing method, device, electronic device and medium, aiming to solve the problem of low task processing efficiency of existing task processing methods.

[0055] The task processing method provided in the embodiment of the present application is specifically illustrated through the following embodiments. First, the task processing method in the embodiment of the present application is described.

[0056] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to achieve optimal results.

[0057] Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0058] The task processing method, device, electronic device and medium provided in the embodiments of the present application relate to the field of task processing. The task processing method provided in the embodiments of the present application can be applied to a terminal, can be applied to a server side, or can be software running in a terminal or a server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or as a server cluster or distributed system composed of multiple physical servers, or as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the task processing method, etc., but is not limited to the above forms.

[0059] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments in which tasks are performed by remote processing devices connected via a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0060] It should be noted that in each specific embodiment of the present application, when it comes to the need to perform relevant processing based on data related to the user's identity or characteristics, such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of such data will comply with relevant laws, regulations, and standards. In addition, when the embodiment of the present application needs to obtain the user's sensitive personal information, the user's separate permission or consent will be obtained through a pop-up window or by jumping to a confirmation page. After clearly obtaining the user's separate permission or consent, the necessary user-related data for the normal operation of the embodiment of the present application will be obtained.

[0061] Figure 1 This is a flowchart of the task processing method provided in the embodiment of the present application. Figure 1 The task processing method provided in the embodiment of the present application may include but is not limited to steps S101 to S102.

[0062] Step S101 : In response to the acquired set of tasks to be processed, the set of tasks to be processed is classified to obtain a plurality of subsets of tasks to be processed.

[0063] In this step, the pending task set may refer to a collection of tasks that need to be processed, which may be implemented by using a task queue or database storage method. The pending task set may be classified by identifying the commonalities between the pending tasks, such as task type, priority, required resources, or business areas. The classification method may be based on preset classification rules or on a machine learning model, such as using a clustering algorithm to classify the pending task set according to the characteristics of the pending tasks.

[0064] Specifically, all pending tasks of the current system are obtained, all pending tasks are taken as a pending task set, and the pending task set is classified to obtain multiple pending task subsets. Among them, the pending task set is classified to obtain multiple pending task subsets by analyzing the business characteristics of each pending task, and the pending tasks with similar business characteristics are classified into the same pending task subset. The pending tasks can be classified according to their priorities, and the pending tasks with the same priority are classified into the same pending task subset. The pending tasks can also be classified according to their resource requirements, and the pending tasks with the same resource requirements are classified into the same pending task subset. The pending tasks with complementary resource requirements can also be classified into the same pending task subset. No specific limitations are made here.

[0065] For example, in the financial insurance industry, insurance companies receive a large number of customer applications, claims requests, and loan approval tasks waiting to be processed. The above pending tasks are grouped into a pending task set, and the above pending task set is classified according to the business type to which the pending tasks belong, to obtain a high-risk loan approval task subset, an insurance claim task subset, and an anti-fraud review task subset.

[0066] For example, in the health care industry, the medical intelligent system receives a large number of patient medical record analysis, imaging diagnosis, and health testing tasks waiting to be processed. The above-mentioned pending tasks are grouped into a pending task set, and the pending task set is divided into a CT imaging diagnosis task subset, a medical record analysis task subset, and a health monitoring data processing task subset according to the business type described in the pending task.

[0067] Step S102: Target processing is performed on each subset of tasks to be processed to complete processing of the set of tasks to be processed. The target processing includes:

[0068] reorganizing the plurality of pending tasks in the pending task subset according to the business relationship of each pending task in the pending task subset to obtain a pending task unit, wherein the business relationship is used to characterize the relevance between the pending tasks;

[0069] Processing the pending tasks in the pending task unit.

[0070] In this step, business relationships may include dependencies between tasks to be processed, data flow relationships, or execution order, etc. Specifically, they can be modeled using a task dependency graph or a business rule engine, and task reorganization can be achieved through graph structure analysis or dependency analysis technology.

[0071] Specifically, target processing is performed for each subset of pending tasks. Target processing consists of two main steps: task reorganization and task processing. During the task reorganization phase, tasks within the pending task subset are reorganized based on their business relationships to create pending task units. Business relationships characterize the relationships between tasks, which can include factors such as data dependencies and execution order. Task reorganization allows closely related tasks to be grouped together to improve processing efficiency.

[0072] For example, in the financial insurance industry, after obtaining the subset of high-risk loan approval pending tasks and the subset of insurance claims pending tasks, the high-risk loan approval type may include multiple pending tasks such as customer credit assessment, income verification, and asset proof, and the insurance claims type may include multiple pending tasks such as accident investigation, medical record verification, and insurance terms matching. According to the business process sequence of these pending tasks, the pending tasks in each pending task subset are reorganized. For example, the high-risk loan approval pending task unit includes pending tasks of credit assessment, income verification, asset certification, and approval decision, and the insurance claims pending task unit includes pending tasks according to accident investigation, medical record verification, insurance terms matching, claim amount calculation, etc.

[0073] The credit assessment model that performs loan approval in the high-risk loan approval pending task unit requires higher CPU (Central Processing Unit) resources, and the image recognition model requires GPU (Graphics Processing Unit) resources. The insurance claim pending task unit needs to utilize large-capacity storage resources such as stored customer historical data. Therefore, more CPU resources and memory are allocated to the high-risk loan approval pending task unit for processing, and more memory is allocated to the insurance claim pending task unit for processing.

[0074] In this implementation, in response to the acquired set of pending tasks, the set of pending tasks is classified to obtain multiple subsets of pending tasks, and a number of pending tasks are divided into multiple subsets of pending tasks through classification operations, providing a structured basis for subsequent processing and improving processing efficiency; target processing is performed on each subset of pending tasks to complete the processing of the set of pending tasks, and the target processing includes: reorganizing the multiple pending tasks in the subset of pending tasks according to the business relationship of each pending task in the subset of pending tasks to obtain a pending task unit, and the business relationship is used to characterize the correlation between the pending tasks; processing the pending tasks in the task unit; reorganizing according to the business relationship between the pending tasks to form a pending task unit, so that task processing is more in line with actual business logic, enhancing the consistency and accuracy of processing, thereby realizing accurate scheduling and efficient utilization of computing resources, avoiding resource waste or conflict, and improving the overall task processing efficiency.

[0075] In some implementations, the processing of the pending tasks in the pending task unit in step S102 may include, but is not limited to, the following:

[0076] Determining target resource requirement information of the task unit to be processed, where the target resource requirement information is used to indicate the computing resources required to process the task unit to be processed;

[0077] Based on the target resource requirement information, the pending tasks in the pending task unit are processed.

[0078] In this implementation, the target resource requirement information may refer to the type of computing resources required by the task unit to be processed, such as CPU, memory, GPU, storage resources, and network resources, etc., which is specifically estimated by analyzing the complexity of all tasks to be processed in the task unit to be processed, historical resource usage, or using a prediction model. For example, a linear regression or neural network model may be used to predict the resource requirements of the tasks to be processed, and the resource requirements of all tasks to be processed in the task unit to be processed are integrated to obtain the target resource requirement information of the task unit to be processed.

[0079] Task processing can refer to the processing of pending task units, which can specifically include allocating computing resources to the pending task units, scheduling task execution, monitoring task progress, and handling task failures. For example, resource scheduling algorithms (such as greedy algorithms, genetic algorithms, or simulated annealing algorithms) can be used to optimize resource allocation, or task queue systems (such as Kafka and RabbitMQ) can be used to manage the execution order of tasks.

[0080] Specifically, during the task processing phase, the target resource requirement information for the task unit to be processed is first determined. This information indicates the computing resources required to process the task unit, and can include the specific requirements for resources such as CPU, memory, and storage. The tasks in the task unit to be processed are then processed based on this resource requirement information.

[0081] In this implementation, the target resource requirement information of the task unit to be processed is determined, and the target resource requirement information is used to indicate the computing resources required to process the task unit to be processed; based on the target resource requirement information, the tasks to be processed in the task unit to be processed are processed to achieve accurate scheduling and efficient utilization of computing resources, avoid resource waste or conflict, and improve the overall task processing efficiency.

[0082] In some embodiments, as Figure 2 As shown, the above-mentioned processing of the pending tasks in the pending task unit based on the target resource demand information may include but is not limited to steps S201 to S203.

[0083] Step S201: Obtain node resource information of multiple task processing nodes associated with the set of tasks to be processed, where the node resource information is used to indicate computing resources available to the task processing nodes.

[0084] Step S202: Determine a target node among the plurality of task processing nodes according to the target resource requirement information of the task unit to be processed and the node resource information of each task processing node.

[0085] Step S203: Use the target node to process the pending tasks in the pending task unit.

[0086] In this implementation, the node resource information may include quantitative indicators such as the number of CPU cores, memory capacity, and network bandwidth.

[0087] The target resource demand information can be generated by analyzing the data volume, computational complexity, and dependency of the tasks to be processed within the task unit to be processed.

[0088] The target node determination process can be achieved by comparing the matching degree between resource requirements and available resources of the node, for example, by adopting a resource gap minimization algorithm or a load balancing strategy.

[0089] Specifically, when the task unit to be processed is generated, the resource usage status of each task processing node is collected in real time to form a node resource information set; at the same time, the resource demand modeling of the task unit to be processed is performed to extract the type and quantity of computing resources required.

[0090] By performing a multi-dimensional matching calculation between resource requirements and available node resources, the node that meets resource constraints and has the most remaining resources is selected as the target node. For example, if a task unit requires an 8-core CPU and 32GB of memory, and Node A has 10 CPU cores and 40GB of memory remaining, while Node B has 6 CPU cores and 64GB of memory remaining, Node A will be prioritized for executing the task unit.

[0091] After the target node is determined, the pending task unit is scheduled to the node for execution, and the node resource status is continuously monitored during the execution of the pending task. If insufficient resources occur, the dynamic migration mechanism is triggered.

[0092] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0093] Obtain node resource information for multiple task processing nodes associated with the pending task set. Node resource information indicates the computing resources available to the task processing nodes. For example, for a distributed computing cluster consisting of 10 task processing nodes, the resource monitoring module can collect real-time resource metrics such as CPU usage, memory usage, and disk I / O (input / output read / write) for each node.

[0094] Obtain target resource requirement information for the pending task unit. This target resource requirement information indicates the computing resources required to process the pending task unit. Specifically, the resource requirements, such as the number of CPU cores and memory size, required for the pending task unit can be estimated based on the historical execution data of each task in the pending task unit.

[0095] Based on the target resource requirement information of the task unit to be processed and the node resource information of each task processing node, a target node is determined from among the multiple task processing nodes. Furthermore, a best matching algorithm can be used to match the resource requirements of the task unit to be processed with the available resources of each node, and the node with the highest resource matching degree is selected as the target node.

[0096] The target node is used to process the pending tasks in the pending task unit. In this way, the tasks can be assigned to the most appropriate node for execution, improving resource utilization efficiency.

[0097] In this implementation, the task units to be processed can be intelligently scheduled and allocated according to their resource requirements and the resource status of each node, so as to make full use of the computing resources in the system and avoid the situation where some nodes are overloaded while other nodes are idle, thereby improving the task processing efficiency and throughput of the entire system; at the same time, by allocating the task units to be processed to the most suitable nodes for execution, the task execution time can be reduced and the user experience can be improved.

[0098] In some embodiments, as Figure 3 As shown, in step S203, according to the target resource requirement information of the task unit to be processed and the node resource information of each task processing node, a target node is determined among multiple task processing nodes, which may include but is not limited to steps S301 to S303.

[0099] Step S301: Determine the availability corresponding to each task processing node according to the computing resources indicated by the node resource information of each task processing node.

[0100] Step S302: Filter the plurality of task processing nodes according to the availability corresponding to each task processing node to obtain a first candidate node set.

[0101] Step S303: For each of the task units to be processed, executing: based on the resource requirements corresponding to the task unit to be processed and the computing resources indicated by the node resource information corresponding to each task processing node in the first candidate node set, calculating the matching degree between the task unit to be processed and each task processing node in the first candidate node set, wherein the matching degree is used to indicate the degree of matching between the computing resources available in the corresponding node and the computing resources required by the task unit to be processed;

[0102] The task processing node with the highest matching degree is used as the target node of the task unit to be processed.

[0103] In this implementation, the availability of a task processing node can be quantified by the difference between the total amount of computing resources in the node resource information and the real-time occupied resources, and the availability screening threshold is dynamically adjusted according to the resource requirements of the task processing unit.

[0104] The first candidate node set may be generated by adopting a resource availability grading mechanism, for example, including nodes whose available computing resources are greater than a preset threshold into the set.

[0105] The matching degree calculation can adopt a multi-dimensional resource weight algorithm, for example, the number of CPU cores, memory capacity, and disk IO bandwidth are weighted and summed according to a preset ratio, and the cosine similarity calculation is performed with the resource requirements of the task processing unit.

[0106] Specifically, the availability of task processing nodes is dynamically evaluated by real-time monitoring of the CPU utilization, memory remaining, and network bandwidth occupancy of the nodes, and a set of nodes whose total available resources meet the minimum requirements of the task processing unit are selected.

[0107] For each pending task unit, the resource requirements for CPU core count, memory capacity, and storage space are extracted and compared against the available resources of each node in the first candidate node set. The match is calculated using a normalized approach, converting the difference between the task requirement and the node resources into a score range of 0 to 1. A weighted summation is then performed to obtain a comprehensive match score. For example, if a task requires an 8-core CPU and the node has 10 available cores, the CPU match score is 0.8; if the memory requirement is 16GB and the node has 20GB available, the memory match score is 0.8. The final match score is the weighted average of the two scores.

[0108] By selecting the node with the highest matching degree, we ensure that the task processing unit obtains the optimal resource adaptation and avoids resource fragmentation or node overload.

[0109] In this implementation, the most suitable processing node is dynamically selected based on the real-time resource status of the task processing node and the resource requirements of the task unit to be processed, thereby improving the utilization efficiency of computing resources and avoiding the problem of uneven resource allocation. At the same time, through matching degree calculation, the rationality of task allocation is ensured, which helps to improve the overall task processing efficiency. In addition, this embodiment has good scalability and can adapt to task processing scenarios of different scales and types.

[0110] In other embodiments, determining a target node from among multiple task processing nodes based on the target resource requirement information of the task unit to be processed and the node resource information of each task processing node in step S203 may include, but is not limited to, the following:

[0111] The target resource requirement information of the task unit to be processed and the node resource information corresponding to the multiple task processing nodes are input into a pre-trained node dynamic allocation model, and the node dynamic allocation model outputs the target node corresponding to each task unit to be processed.

[0112] In this implementation, the node dynamic allocation model can be obtained by training historical task processing data, where the historical task processing data includes a mapping relationship between to-be-processed task units with different resource requirements and node resource states.

[0113] The input layer of the node dynamic allocation model receives the resource demand vector of the task unit to be processed and the node resource status matrix, extracts features through a multi-layer neural network, and the output layer generates the matching probability distribution of each node.

[0114] During the training process of the dynamic node allocation model, a backpropagation algorithm can be used to optimize model parameters, enabling the model to learn the dynamic matching rules between resource requirements and node status. During the model deployment phase, the resource requirement parameters of the current task unit and the real-time node resource data are received in real time, and the target node selection result is generated through forward calculation.

[0115] Specifically, when a task unit to be processed is generated, its resource requirement information is quantified into a feature vector containing the number of CPU cores, memory capacity, and processing delay requirements. The resource information of the task processing node is synchronously collected and constructed into a data matrix containing node identification, available computing resources, and load status.

[0116] The resource demand vector and node resource status matrix are normalized and input into the node dynamic allocation model. The model extracts local resource matching features through the convolution layer, and the fully connected layer integrates the global resource constraints. Finally, the adaptation weight value of each node is output, and the node with the highest weight value is selected as the target node to achieve dynamic adaptation of resource demand and node capabilities.

[0117] In this implementation, model reasoning replaces manual rule calculation, which can complete matching decisions for large-scale node clusters within milliseconds, while adapting to real-time fluctuations in node resource status to improve task processing efficiency in high-concurrency, high-throughput business scenarios.

[0118] In some embodiments, as Figure 4As shown, in step S101 , in response to the acquired set of tasks to be processed, the set of tasks to be processed is classified to obtain multiple subsets of tasks to be processed, which may include but is not limited to steps S401 to S403 .

[0119] Step S401: Acquire the pending task set, which includes a plurality of pending tasks;

[0120] Step S402: extracting features from each of the tasks to be processed to obtain a business feature corresponding to each of the tasks to be processed, where the business feature indicates the business type to which the task to be processed belongs.

[0121] Step S403: classify the set of tasks to be processed according to the business characteristics corresponding to each of the tasks to be processed to obtain a plurality of subsets of the tasks to be processed.

[0122] In this implementation, the feature extraction step identifies the business type by parsing task attributes or context information, for example, extracting keywords, data formats, or interface types from task metadata.

[0123] Business types can be classified according to task keywords or task themes, or according to task sources. For example, tasks from a CRM system are classified as customer service, and tasks from a procurement system are classified as purchase order. They can also be set according to actual application scenario requirements and are not specifically limited here.

[0124] The classification step can use a clustering algorithm or a classifier based on preset rules, for example, to classify tasks with the same data source or processing logic into the same subset.

[0125] The feature extraction and classification steps form a closed loop, business features provide the basis for classification, and the classification results optimize subsequent task reorganization and resource allocation.

[0126] Specifically, after obtaining a set of pending tasks, each is independently analyzed to extract business characteristics. For example, by parsing the service identifier in the task parameters to determine whether it belongs to order processing or log analysis. Business characteristics are stored as structured data, such as hash tables or vectors.

[0127] The classification process groups tasks based on feature similarity. For example, the K-means algorithm is used to group tasks with similar feature vector distances into the same subset. For example, in an e-commerce system, tasks containing user and product IDs are classified into the order processing subset, while tasks containing timestamps and error codes are classified into the log monitoring subset. This ensures that the tasks within the task subset are highly cohesive, reduces resource competition caused by the mixing of different types of tasks, and improves the parallel efficiency of the target processing stage.

[0128] For example, in the financial insurance industry, there is a pending task set in the insurance core business system, which contains 38 pending tasks that originally need to be processed serially. According to the actual situation of the insurance core business system, the 38 pending tasks are classified into six pending subsets. Among them, the first pending task subset includes pending tasks of dividend offset, survival benefit repayment, survival benefit offset, and renewal transfer to actual collection; the second pending task subset includes pending tasks of maturity, dividend distribution, and dividend transfer withdrawal; the third pending task subset includes pending tasks of renewal transfer to actual collection (renewal processing) The fourth subset of pending tasks includes the pending tasks of canceling the surcharge due, extending the effectiveness of preservation, deciding on the effectiveness of expiration, taking effect of the second verification, and transferring renewal to actual collection; the fifth subset of pending tasks includes the processing of self-advancement day, processing of suspension day, and supplementary extraction of files for renewal batches; the sixth subset of pending tasks includes the pending tasks of canceling batch acceptance and automatic adjustment of failed financial transfers. By splitting and classifying the 38 pending tasks into six subsets of pending tasks, these 38 pending tasks can be processed in parallel according to the pending subsets, thereby greatly improving the task processing efficiency of the insurance core business system.

[0129] In this implementation, tasks with similar business attributes are grouped into the same subset, facilitating the subsequent adoption of differentiated processing strategies for different types of tasks and improving overall task processing efficiency. Furthermore, task classification through feature extraction and clustering is more objective and accurate than manual classification, better reflecting the inherent relevance between tasks.

[0130] In some embodiments, after the plurality of pending tasks in the pending task subset are reorganized according to the business relationship of each pending task in the pending task subset to obtain a pending task unit in step S102, the task processing method provided in the embodiment of the present application may include, but is not limited to, the following contents:

[0131] Obtaining the urgency level of each pending task in the pending task unit and first resource requirement information corresponding to each pending task, where the first resource requirement information is used to indicate the computing resources required to process the pending task;

[0132] Inputting the urgency of the tasks to be processed and the first resource requirement information corresponding to each task to be processed into a preset weight distribution model, and outputting the weight corresponding to each task to be processed by the weight distribution model;

[0133] The processing order of the tasks to be processed in the task unit to be processed is updated according to the weight corresponding to each task to be processed, to obtain an updated task unit to be processed.

[0134] In this implementation, the urgency level may be determined by the task deadline or the business priority tag.

[0135] The first resource requirement information includes CPU core number, memory capacity, and storage space indicators.

[0136] The weight allocation model adopts a linear weighted algorithm. The urgency and the first resource demand information are mapped into numerical parameters respectively, and a comprehensive weight value is generated after normalization processing.

[0137] The processing order is updated by comparing the weight values ​​and placing the tasks with higher weight values ​​at the front of the execution queue of the task unit to be processed.

[0138] For example, when a pending task unit contains three pending tasks, Task A's urgency parameter is 0.8, and its first resource requirement information parameter is 0.3; Task B's urgency parameter is 0.5, and its first resource requirement information parameter is 0.6; and Task C's urgency parameter is 0.6, and its first resource requirement information parameter is 0.4. The weight distribution model sets the urgency coefficient to 0.7 and the resource requirement coefficient to 0.3. The calculated weight of Task A is 0.7 × 0.8 + 0.3 × 0.3 = 0.65, the weight of Task B is 0.7 × 0.5 + 0.3 × 0.6 = 0.53, and the weight of Task C is 0.7 × 0.6 + 0.3 × 0.4 = 0.54. The updated processing order becomes Task A to Task C to Task B, ensuring that high-urgency tasks receive computing resources first while balancing the execution timing of tasks with higher resource consumption. This dynamic adjustment mechanism ensures that the processing order of each pending task within the pending task unit meets both timeliness requirements and resource efficiency.

[0139] As a further example, as a preferred embodiment, the solution of the present application is specifically implemented as follows:

[0140] Obtain the urgency level and first resource requirement information corresponding to each pending task in the pending task unit. The first resource requirement information indicates the computing resources required to process the pending task. For example, for a data processing task, its urgency level may be set to "high," and the first resource requirement information may include a requirement for a 2-core CPU and 4GB of memory.

[0141] The urgency of the pending tasks and the first resource requirement information corresponding to each pending task are input into a preset weight allocation model. The weight allocation model then outputs a weight corresponding to each pending task. Specifically, the weight allocation model can be a model trained based on a machine learning algorithm, which receives the urgency and resource requirement as input features and outputs a weight value between 0 and 1.

[0142] The processing order of the tasks to be processed in the task unit to be processed is updated according to the weight corresponding to each task to be processed, thereby obtaining an updated task unit to be processed. Furthermore, the tasks to be processed can be sorted from high to low according to the weight to form a new processing queue, which is the updated task unit to be processed.

[0143] In this implementation, the task processing order is dynamically adjusted based on task urgency and resource requirements, improving the flexibility and efficiency of task processing. This prioritizes important and urgent tasks while also taking resource requirements into account, avoiding processing delays caused by irrational resource allocation. This allows for better adaptation to complex and changing task processing scenarios, improving overall response speed and resource utilization.

[0144] In other embodiments, after the target node is used to process the pending tasks in the pending task unit in step S204, the task processing method provided in the embodiment of the present application may include, but is not limited to, the following contents:

[0145] Monitoring the processing result of each task to be processed in the task to be processed unit;

[0146] When the processing result indicates that an exception occurs in the target to-be-processed task, obtaining a predecessor task and a successor task of the target to-be-processed task, wherein the predecessor task is a task to be processed before the target to-be-processed task, and the successor task is a task to be processed after the target to-be-processed task;

[0147] Using the predecessor task and the successor task as dependent tasks of the target task to be processed;

[0148] Re-execute the dependent task to re-execute the target to-be-processed task.

[0149] In this implementation, the monitoring processing results can be achieved through log analysis or status code detection.

[0150] The predecessor task and the successor task of the abnormal task (target to-be-processed task) can be determined according to the task execution sequence or the dependency graph.

[0151] The re-execution of dependent tasks is done by reinserting the tasks into the processing queue through the task scheduler and assigning them to available nodes with priority.

[0152] To identify task dependencies, a directed acyclic graph data structure can be used to record the order of tasks. When an exception is triggered, the structure is traversed to obtain related tasks.

[0153] Specifically, when the target node processes a task unit, the execution status of each task is recorded in real time. If an exception is detected in the target pending task, its predecessor task and successor task are immediately extracted from the task dependency relationship. The predecessor task is the last successful task before the execution of the abnormal task, and the successor task is the first task to be executed after the execution of the abnormal task. The dependent tasks are marked as pending retry status and reallocated to idle nodes for processing. For example, in an order processing system, if the inventory deduction task fails, its predecessor order verification task and the subsequent logistics notification task are automatically obtained, and only these three tasks are re-executed to avoid repeated processing of payment or user information verification tasks. Through the local retry mechanism, redundant calculations are reduced and the efficiency of abnormal recovery is improved.

[0154] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0155] After the target node processes the pending tasks in the pending task unit, the processing results of each pending task in the pending task unit are monitored. Specifically, the execution status, output results, resource usage, and other information of each pending task can be obtained through regular polling or real-time monitoring.

[0156] Furthermore, if the processing result indicates that the target pending task has an exception, the predecessor task and successor task of the target pending task are obtained. A predecessor task is the previous pending task of the target pending task, and a successor task is the next pending task of the target pending task. For example, by analyzing a task dependency graph or a task execution sequence, upstream and downstream tasks directly related to the abnormal task are determined.

[0157] Therefore, the predecessor and successor tasks are considered as dependent tasks of the target task to be processed. These dependent tasks may affect the input data of the abnormal task or be affected by the output of the abnormal task, so they need to be considered together.

[0158] Finally, re-execute the dependent tasks to re-execute the target pending task. Specifically, re-execution can be done in the original execution order, starting with the predecessor task, then executing the target pending task, and finally executing the successor task. During the re-execution process, more computing resources or optimized algorithms can be used to improve the execution success rate.

[0159] This implementation improves the reliability and robustness of task processing by monitoring the processing results of pending tasks in real time, promptly identifying and addressing anomalies and resuming normal operation by re-executing related dependent tasks. Furthermore, by re-executing only necessary dependent tasks, rather than the entire set of tasks, unnecessary resource waste can be reduced, improving overall efficiency.

[0160] In some implementations, in step S102 , target processing is performed on each subset of tasks to be processed to complete processing of the set of tasks to be processed, which may include but is not limited to step S501 .

[0161] Step S501: Control multiple subsets of the pending tasks to execute the target processing in parallel to complete the processing of the pending task sets;

[0162] The processing of the pending tasks in the pending task unit includes:

[0163] According to the preset execution logic corresponding to the task unit to be processed, multiple tasks to be processed in the task unit to be processed are serially processed in sequence to complete the processing of the task unit to be processed.

[0164] In this implementation, parallel execution of target processing maximizes the computing power of the distributed system by allocating different subsets of tasks to be processed to multiple processing nodes for simultaneous execution.

[0165] When the number of processing nodes is insufficient to cover all pending task units, the task allocation order is dynamically adjusted through the priority sorting mechanism, giving priority to processing high-priority task units to avoid critical path blocking.

[0166] Serial processing is specifically manifested as processing tasks one by one according to the preset execution logic within a single task unit to be processed, ensuring that tasks with dependencies are executed in the correct order.

[0167] Priority determination can be based on task type, deadline, or resource demand intensity. For example, data processing tasks have a higher priority than log cleaning tasks, and real-time computing tasks have a higher priority than batch tasks.

[0168] Specifically, after obtaining the set of tasks to be processed, it is first classified into multiple subsets; each subset is reorganized to form a task unit, and the tasks within the unit are sorted according to business relationships. When the number of available nodes is less than the number of task units, the priority parameters of each unit are extracted, for example, by analyzing the task type identifier or the weight table in the preset rule base; high-priority units are preferentially assigned to idle nodes, and low-priority units enter the waiting queue.

[0169] When processing a unit, a node executes the tasks within it one by one in a pre-set order, for example, executing data cleaning tasks before aggregate computing tasks. Once a node completes processing the current unit, it extracts the next priority unit from the queue and continues. This serial processing eliminates state conflicts between tasks, while priority allocation ensures that system resources are allocated to critical tasks, optimizing overall processing efficiency under resource-constrained conditions.

[0170] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0171] When receiving a set of pending tasks containing user order data, inventory verification requests, and logistics scheduling instructions, the feature extraction engine first identifies the business type label of each task and divides the task set into an order processing subset, an inventory management subset, and a logistics scheduling subset.

[0172] Three independent task processing threads were then established within the distributed computing cluster, with each subset assigned to a different thread for parallel processing. When processing the order processing subset, payment requests, address verification operations, and product packaging instructions for the same order number were sequentially grouped into task units according to the pre-defined order verification-payment confirmation-shipment preparation process. These units were then assigned to the server nodes in the subset for serial processing according to the process steps.

[0173] In this implementation, under the premise of maintaining the integrity of the business logic within the task unit, the overall processing throughput is significantly improved through multi-subset parallel execution, effectively solving the technical contradiction that traditional task processing methods cannot coordinate parallel and serial execution logic, thereby greatly improving task processing efficiency.

[0174] In other embodiments, in step S501, controlling multiple subsets of pending tasks to execute target processing in parallel to complete processing of the pending task set may include, but is not limited to, the following:

[0175] When the number of the plurality of task processing nodes associated with the to-be-processed task set is less than the number of the to-be-processed task units, obtaining a priority corresponding to each of the to-be-processed task units;

[0176] According to the priority corresponding to each of the to-be-processed task units, each of the to-be-processed task units is allocated to a corresponding target node for processing.

[0177] In this implementation, parallel processing can allocate the classified task subsets to different threads or processes through a task scheduler.

[0178] Priority evaluation can generate a priority score based on the business type, resource consumption or deadline of the task unit to be processed.

[0179] Node allocation prioritizes high-priority tasks to idle nodes based on the ranking results. When there are insufficient nodes to handle all pending tasks simultaneously, the pending tasks are sorted in descending order of score, and the load status of each node is updated in real time to dynamically adjust the allocation strategy.

[0180] In this implementation, after receiving the classified subset of pending tasks, multiple parallel processing threads are started. Each thread reorganizes the subset of pending tasks into pending task units, and calls a preset scoring algorithm to generate a priority scoring value based on the business type identification, historical resource consumption records and preset urgency parameters of the pending task units.

[0181] The task units with a score higher than the set threshold are inserted into the head of the priority queue, and the task units with a score lower than the threshold are inserted into the tail of the queue.

[0182] The system periodically collects CPU utilization, remaining memory, and network bandwidth data for each node. When it detects that a node's resources have reached their available threshold, it removes a task unit from the head of the priority queue and assigns it to that node for execution. If node resources remain insufficient, the task unit's waiting time in the queue triggers a timeout retry mechanism, and the priority is reassessed before the next round of allocation begins.

[0183] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0184] When the number of task processing nodes associated with a pending task set is insufficient to process all pending task units simultaneously, a priority score is generated for each task unit by analyzing the task unit's business type and the priority strategy in the preset rule base. The priority score is calculated based on the average urgency of the tasks within the task unit, the total resource requirements, and the business type weight coefficient.

[0185] All pending tasks are then sorted by priority, from high to low, to form a task queue. The available resource capacity of the task processing nodes is monitored in real time. Based on the order of the task queue, the highest-priority pending task is dynamically assigned to a node with sufficient available resources. When a node completes processing its current task, its freed resources are immediately used to process the next highest-priority task in the queue.

[0186] This implementation effectively addresses the issue of decreased task processing efficiency in scenarios with limited computing resources. Through a dynamic priority-based task allocation mechanism, high-priority task units are prioritized for computing resources, reducing latency in critical service links. This approach also enables dynamic adaptation and recycling of computing resources, preventing idle or overloaded node resources and maintaining the overall throughput of the task processing system within limited node constraints.

[0187] Figure 5 This is a schematic diagram of the structure of the task processing device provided in the embodiment of the present application. Figure 5 The embodiment of the present application further provides a task processing device 800, which can implement the above-mentioned task processing method. The task processing device 800 includes:

[0188] A classification module 801 is configured to classify the acquired set of tasks to be processed in response to the acquired set of tasks to be processed to obtain a plurality of subsets of tasks to be processed;

[0189] The processing module 802 is configured to perform target processing on each subset of tasks to be processed to complete processing of the set of tasks to be processed. The target processing includes:

[0190] reorganizing the plurality of pending tasks in the pending task subset according to the business relationship of each pending task in the pending task subset to obtain a pending task unit, wherein the business relationship is used to characterize the relevance between the pending tasks;

[0191] Processing the pending tasks in the pending task unit.

[0192] In some implementations, the processing module 802 includes:

[0193] An information determination submodule, configured to determine target resource requirement information of the task unit to be processed, wherein the target resource requirement information is used to indicate the computing resources required to process the task unit to be processed;

[0194] The first processing submodule is configured to process the pending tasks in the pending task unit based on the target resource requirement information.

[0195] In some implementations, the processing module 802 includes:

[0196] A first acquisition submodule is configured to acquire node resource information of a plurality of task processing nodes associated with the set of tasks to be processed, wherein the node resource information is used to indicate computing resources available to the task processing nodes;

[0197] a node determination submodule, configured to determine a target node among the plurality of task processing nodes according to the target resource requirement information of the task unit to be processed and the node resource information of each of the task processing nodes;

[0198] The second processing submodule is configured to process the pending tasks in the pending task unit using the target node.

[0199] In some embodiments, the node determination submodule includes:

[0200] a computing unit, configured to determine the availability corresponding to each of the task processing nodes according to the computing resources indicated by the node resource information of each of the task processing nodes;

[0201] a screening unit, configured to screen the plurality of task processing nodes according to the availability corresponding to each of the task processing nodes to obtain a first candidate node set;

[0202] an execution unit, configured to, for each of the to-be-processed task units, calculate, based on resource requirements corresponding to the to-be-processed task unit and computing resources indicated by node resource information corresponding to each task processing node in the first candidate node set, a matching degree between the to-be-processed task unit and each task processing node in the first candidate node set, wherein the matching degree is used to indicate a degree of matching between computing resources available in the corresponding node and computing resources required by the to-be-processed task unit;

[0203] The task processing node with the highest matching degree is used as the target node of the task unit to be processed.

[0204] In some implementations, the node determination submodule further includes:

[0205] An allocation unit is used to input the target resource requirement information of the task unit to be processed and the node resource information corresponding to the multiple task processing nodes into a pre-trained node dynamic allocation model, and the node dynamic allocation model outputs the target node corresponding to each task unit to be processed.

[0206] In some embodiments, the classification module 801 includes:

[0207] A second acquisition submodule is used to acquire the pending task set, where the pending task set includes a plurality of pending tasks;

[0208] A feature extraction submodule is used to extract features from each of the tasks to be processed to obtain business features corresponding to each of the tasks to be processed, where the business features are used to indicate the business type to which the task to be processed belongs;

[0209] The task classification submodule is used to classify the set of tasks to be processed according to the business characteristics corresponding to each of the tasks to be processed, so as to obtain a plurality of subsets of the tasks to be processed.

[0210] In some implementations, the task processing device 800 further includes:

[0211] A resource acquisition module, configured to acquire the urgency level of each pending task in the pending task unit;

[0212] A weight determination module, configured to input the urgency of the tasks to be processed and the first resource requirement information corresponding to each task to be processed into a preset weight allocation model, and output the weight corresponding to each task to be processed by the weight allocation model;

[0213] The sequence updating module is used to update the processing sequence of the tasks to be processed in the task unit to be processed according to the weight corresponding to each task to be processed, so as to obtain an updated task unit to be processed.

[0214] In some implementations, the processing module 802 further includes:

[0215] A monitoring submodule, configured to monitor the processing result of each task to be processed in the task to be processed unit;

[0216] A task acquisition submodule is used to acquire a predecessor task and a successor task of the target to-be-processed task when the processing result indicates that an abnormality occurs in the target to-be-processed task, wherein the predecessor task is the previous to-be-processed task of the target to-be-processed task and the successor task is the next to-be-processed task of the target to-be-processed task;

[0217] A task determination submodule, configured to use the predecessor task and the successor task as dependent tasks of the target task to be processed;

[0218] The re-execution submodule is used to re-execute the dependent task to re-execute the target task to be processed.

[0219] In some implementations, the processing module 802 further includes:

[0220] A third processing submodule is used to control the plurality of task subsets to be processed to execute the target processing in parallel, so as to complete the processing of the task set to be processed;

[0221] The processing of the pending tasks in the pending task unit includes:

[0222] According to the preset execution logic corresponding to the task unit to be processed, multiple tasks to be processed in the task unit to be processed are serially processed in sequence to complete the processing of the task unit to be processed.

[0223] In some embodiments, the second processing submodule includes:

[0224] a priority obtaining unit, configured to obtain the priority corresponding to each of the to-be-processed task units when the number of the plurality of task processing nodes associated with the to-be-processed task set is less than the number of the to-be-processed task units;

[0225] The node allocation unit is used to allocate each of the to-be-processed task units to a corresponding target node for processing according to the priority corresponding to each of the to-be-processed task units.

[0226] The specific implementation of the task processing device 800 is substantially the same as the specific embodiment of the task processing method described above, and will not be described in detail here.

[0227] The present application also provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the task processing method described above when executing the computer program. The electronic device can be any intelligent terminal, including a desktop computer, a tablet computer, a mobile phone, and an in-vehicle computer.

[0228] See also Figure 6 , Figure 6 This is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application. The electronic device includes:

[0229] The processor 901 can be implemented as a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application.

[0230] The memory 902 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called by the processor 901 to execute the task processing method of the embodiments of this application.

[0231] Input / output interface 903, used to implement information input and output;

[0232] Communication interface 904, used to implement communication interaction between this device and other devices, which can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WiFi, Bluetooth, etc.);

[0233] Bus 905 , which transmits information between various components of the device (e.g., processor 901 , memory 902 , input / output interface 903 , and communication interface 904 );

[0234] The processor 901 , the memory 902 , the input / output interface 903 and the communication interface 904 are connected to each other in communication within the device via a bus 905 .

[0235] An embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned task processing method is implemented.

[0236] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0237] In addition, the embodiments of the present application may be implemented by providing a computer program product. When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device implements the task processing method in the above embodiment.

[0238] The task processing method, device, electronic device and medium provided in the embodiments of the present application, in response to the acquired set of pending tasks, classify the set of pending tasks to obtain multiple subsets of pending tasks, and divide a number of pending tasks into multiple subsets of pending tasks through classification operations, thereby providing a structured basis for subsequent processing and improving processing efficiency; performing target processing on each subset of pending tasks to complete the processing of the set of pending tasks, and the target processing includes: reorganizing the multiple pending tasks in the subset of pending tasks according to the business relationship of each pending task in the subset of pending tasks to obtain a pending task unit, wherein the business relationship is used to characterize the correlation between the pending tasks; processing the pending tasks in the task unit; reorganizing according to the business relationship between the pending tasks to form a pending task unit, so that task processing is more in line with actual business logic, enhancing the consistency and accuracy of processing, thereby realizing accurate scheduling and efficient utilization of computing resources, avoiding resource waste or conflict, and improving the overall task processing efficiency.

[0239] The embodiments described in the embodiments of this application are intended to more clearly illustrate the technical solutions of the embodiments of this application and do not constitute a limitation on the technical solutions provided by the embodiments of this application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0240] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0241] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, i.e., they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment.

[0242] Those skilled in the art will appreciate that all or some of the steps in the methods, systems, and functional modules / units in the devices disclosed above may be implemented as software, firmware, hardware, or appropriate combinations thereof.

[0243] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0244] It should be understood that in this application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0245] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the above-mentioned units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0246] The units described above as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0247] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0248] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes multiple instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: various media that can store programs, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0249] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but are not intended to limit the scope of the present invention. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and essence of the present invention should be within the scope of the present invention.

Claims

1. A task processing method, characterized in that: The method comprises: In response to the acquired set of pending tasks, classify the set of pending tasks to obtain a plurality of subsets of pending tasks; Target processing is performed on each subset of tasks to be processed to complete processing of the set of tasks to be processed, wherein the target processing includes: reorganizing the plurality of pending tasks in the pending task subset according to the business relationship of each pending task in the pending task subset to obtain a pending task unit, wherein the business relationship is used to characterize the relevance between the pending tasks; Processing the pending tasks in the pending task unit.

2. The method according to claim 1, characterized in that The processing of the pending tasks in the pending task unit includes: Determining target resource requirement information of the task unit to be processed, where the target resource requirement information is used to indicate the computing resources required to process the task unit to be processed; Based on the target resource requirement information, the pending tasks in the pending task unit are processed.

3. The method according to claim 2, characterized in that The processing of the pending tasks in the pending task unit based on the target resource demand information includes: Acquire node resource information of a plurality of task processing nodes associated with the set of tasks to be processed, wherein the node resource information is used to indicate computing resources available to the task processing nodes; Determining a target node among the plurality of task processing nodes according to the target resource requirement information of the task unit to be processed and the node resource information of each task processing node; The target node is used to process the pending tasks in the pending task unit.

4. The method according to claim 3, characterized in that The determining of a target node among the plurality of task processing nodes according to the target resource requirement information of the task unit to be processed and the node resource information of each task processing node includes: Determining the availability of each task processing node according to the computing resources indicated by the node resource information of each task processing node; Screening the plurality of task processing nodes according to the availability corresponding to each task processing node to obtain a first candidate node set; For each of the to-be-processed task units, executing: calculating, based on the resource requirements corresponding to the to-be-processed task unit and the computing resources indicated by the node resource information corresponding to each task processing node in the first candidate node set, a matching degree between the to-be-processed task unit and each task processing node in the first candidate node set, wherein the matching degree is used to represent the degree of matching between the computing resources available in the corresponding node and the computing resources required by the to-be-processed task unit; The task processing node with the highest matching degree is used as the target node of the task unit to be processed.

5. The method according to claim 1, wherein In response to the acquired set of pending tasks, the set of pending tasks is classified to obtain a plurality of subsets of pending tasks, including: Acquire the pending task set, where the pending task set includes a plurality of pending tasks; Extracting features of each of the tasks to be processed to obtain business features corresponding to each of the tasks to be processed, wherein the business features are used to indicate the business type to which the tasks to be processed belong; The set of tasks to be processed is classified according to the business characteristics corresponding to each of the tasks to be processed to obtain a plurality of subsets of the tasks to be processed.

6. The method according to claim 2, characterized in that After reorganizing the plurality of tasks to be processed in the subset of tasks to be processed according to the business relationship of each task to be processed in the subset of tasks to be processed to obtain a task unit to be processed, the method includes: Obtaining the urgency level of each pending task in the pending task unit; Inputting the urgency of the tasks to be processed and the first resource requirement information corresponding to each task to be processed into a preset weight distribution model, and outputting the weight corresponding to each task to be processed by the weight distribution model; The processing order of the tasks to be processed in the task unit to be processed is updated according to the weight corresponding to each task to be processed, to obtain an updated task unit to be processed.

7. The method according to claim 1, characterized in that The performing target processing on each subset of the to-be-processed tasks to complete the processing of the to-be-processed task set includes: Controlling a plurality of the subsets of tasks to be processed to execute the target processing in parallel to complete the processing of the set of tasks to be processed; The processing of the pending tasks in the pending task unit includes: According to the preset execution logic corresponding to the task unit to be processed, multiple tasks to be processed in the task unit to be processed are serially processed in sequence to complete the processing of the task unit to be processed.

8. A task processing device, characterized in that: The device comprises: a classification module, configured to classify the acquired set of tasks to be processed in response to the acquired set of tasks to be processed, and obtain a plurality of subsets of tasks to be processed; A processing module is configured to perform target processing on each subset of tasks to be processed to complete processing of the set of tasks to be processed, wherein the target processing includes: reorganizing the plurality of pending tasks in the pending task subset according to the business relationship of each pending task in the pending task subset to obtain a pending task unit, wherein the business relationship is used to characterize the relevance between the pending tasks; Processing the pending tasks in the pending task unit.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the task processing method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the task processing method according to any one of claims 1 to 7 is implemented.

Images