A method and apparatus for adjusting the priority of a transaction job

Abstract

The application discloses a kind of priority adjustment method and device of transaction operation task, comprising: according to the single score of each performance index of each historical transaction operation task and actual comprehensive score, target weight of each performance index in objective function model is determined by using chaos particle swarm genetic algorithm iterative solution objective function model;The current single score of each performance index of current transaction operation task is obtained, and the current comprehensive score of current transaction operation task is calculated according to the current single score and the target weight of each performance index, and the priority of current transaction operation task is adjusted according to the current comprehensive score.Chaos mapping and particle swarm strategy are fused into genetic algorithm to solve, and the target weight of each performance index is obtained, so as to improve the accuracy of the target weight solved, and the priority is automatically adjusted based on the current comprehensive score determined by the target weight, to improve the priority adjustment efficiency and accuracy.

Description

Technical Field

[0001] This invention relates to the field of financial technology, and in particular to a method and apparatus for prioritizing transaction tasks. Background Technology

[0002] With the rapid development of information technology, financial institutions generate a large number of transaction tasks every day that need to be processed. The efficiency of these transaction tasks directly affects the timeliness of data generation and the accuracy of business decisions, while the priority of transaction tasks directly determines the efficiency of operation.

[0003] Currently, the priority adjustment of trading tasks mainly relies on the experience and judgment of managers. However, in large-scale task scheduling scenarios, manual configuration is inefficient, prone to errors due to negligence, and affects system stability. Furthermore, manual adjustment has a time lag, making it difficult to ensure the timely execution of critical tasks. Moreover, the manual approach lacks a unified evaluation standard, and different managers have different adjustment strategies, resulting in uneven allocation of system resources for trading tasks. Therefore, the existing priority adjustment method cannot meet the actual needs of financial institutions. Summary of the Invention

[0004] This invention provides a method and apparatus for adjusting the priority of trading tasks, so as to achieve accurate adjustment of the priority of trading tasks.

[0005] According to a first aspect of the present invention, a method for prioritizing transaction tasks is provided, the method comprising: Obtain the actual overall score of historical transaction jobs, as well as at least one performance indicator and its corresponding individual score, wherein the performance indicator includes at least one of the following: job importance guarantee level, job runtime, job computing resource consumption, and cluster remaining resources; Based on the individual scores of each performance indicator of each historical trading task and the actual comprehensive score, a chaotic particle swarm genetic algorithm is used to iteratively solve the objective function model to determine the target weight of each performance indicator in the objective function model. The objective function model is used to characterize the minimization error between the predicted comprehensive score and the actual comprehensive score of the historical trading task. Obtain the current individual scores of each performance metric for the current trading task, calculate the current overall score of the current trading task based on the current individual scores and the target weights of each performance metric, and adjust the priority of the current trading task based on the current overall score.

[0006] According to another aspect of the present invention, a priority adjustment device for transaction operation tasks is provided, the device comprising: The score acquisition module is used to acquire the actual comprehensive score of historical transaction jobs, as well as at least one performance indicator and its corresponding individual score. The performance indicator includes at least one of the following: job importance guarantee level, job runtime, job computing resource consumption, and cluster remaining resources. The objective weight solution module is used to iteratively solve the objective function model using a chaotic particle swarm genetic algorithm based on the individual scores of each performance indicator of each historical trading task and the actual comprehensive score, so as to determine the objective weight of each performance indicator in the objective function model. The objective function model is used to characterize the minimized error between the predicted comprehensive score and the actual comprehensive score of the historical trading task. The priority adjustment module is used to obtain the current individual scores of each performance indicator of the current trading task, calculate the current comprehensive score of the current trading task based on the current individual scores and the target weights of each performance indicator, and adjust the priority of the current trading task based on the current comprehensive score.

[0007] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: one or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any embodiment of the present invention.

[0008] According to another aspect of the present invention, a storage medium for computer-executable instructions is provided, on which a computer program is stored, which, when executed by a processor, implements the method as described in any of the embodiments of the present invention.

[0009] The technical solution of this invention constructs an objective function model based on information from historical trading tasks, and integrates chaotic mapping and particle swarm optimization into a genetic algorithm for solving the model. This yields the target weights for each performance indicator. Chaotic mapping increases ergodicity, while particle weights accelerate convergence, thereby improving the accuracy of the obtained target weights. Based on the target weights and the current comprehensive score determined by the current individual scores of each performance indicator, the priority is automatically adjusted, thus improving the efficiency and accuracy of priority adjustment for the current trading task.

[0010] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a flowchart of a method for adjusting the priority of transaction tasks according to Embodiment 1 of the present invention; Figure 2 This is a flowchart of another method for adjusting the priority of transaction tasks according to Embodiment 2 of the present invention; Figure 3 This is a schematic diagram of the structure of a priority adjustment device for transaction operation tasks according to Embodiment 3 of the present invention; Figure 4 This is a structural block diagram of an electronic device provided in Embodiment 4 of the present invention. Detailed Implementation

[0013] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0014] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, apparatus, product, or terminal device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or terminal devices.

[0015] Example 1 Figure 1This is a flowchart of a method for adjusting the priority of trading tasks according to Embodiment 1 of the present invention. This embodiment is applicable to situations where the priority of trading tasks needs to be adjusted. This method can be executed by a device for adjusting the priority of trading tasks. This device can be implemented in hardware and / or software, and can be integrated into an electronic device with data processing capabilities. Figure 1 As shown, the method includes: S101, obtain the actual overall score of historical transaction tasks, as well as at least one performance indicator and its corresponding individual score.

[0016] Optionally, the actual comprehensive score of historical trading tasks, as well as at least one performance indicator and its corresponding individual score, are obtained, including: obtaining the actual comprehensive score marked for each historical trading task, and collecting the raw data of each performance indicator of each historical trading task; obtaining the pre-configured score calculation strategy for each performance indicator, and calculating the raw data according to the score calculation strategy for each performance indicator to obtain the individual score corresponding to the performance indicator.

[0017] Specifically, this implementation retrieves the actual comprehensive scores of each historical transaction task from the database. These historical transaction tasks are primarily those with scores manually marked. Additionally, it collects raw data on various performance metrics for each historical transaction task. These metrics include task importance level, task runtime, task computational resource consumption, and remaining cluster resources. Task importance level refers to the importance of the transaction task; for example, a payment system task is significantly more important than a regular log analysis task. Task runtime refers to the duration of the transaction task execution; typically, more resources are reserved for long tasks. Task computational resource consumption refers to the resource capacity used after the transaction task is completed. Remaining cluster resources refer to the remaining server space resources when executing the transaction task; in resource-constrained situations, smaller tasks are usually prioritized. This implementation is merely illustrative and does not limit the specific types of performance metrics.

[0018] In addition, in this embodiment, corresponding score calculation strategies are pre-configured for each performance indicator. Therefore, after obtaining the raw data of each performance indicator, the raw data will be calculated according to the corresponding score calculation strategy to obtain the corresponding individual score. For example, for job runtime, corresponding individual score mapping relationships are set for different job runtime ranges. Therefore, after obtaining the specific duration of the runtime, the corresponding individual score can be determined according to the specific duration and the above mapping relationship. Of course, this embodiment is only an example. The calculation method for individual scores of other performance indicators is roughly the same, and will not be elaborated in this embodiment.

[0019] S102, based on the individual scores and actual comprehensive scores of each performance indicator of each historical transaction task, the objective function model is iteratively solved using the chaotic particle swarm genetic algorithm to determine the target weight of each performance indicator in the objective function model.

[0020] Optionally, before iteratively solving the objective function model using the chaotic particle swarm genetic algorithm based on the individual scores and actual comprehensive scores of each performance indicator for each historical trading task, the algorithm further includes: calculating the product of the individual scores of each performance indicator and the corresponding dynamic weights for each historical trading task, where the dynamic weights are variables; and summing the product results of each performance indicator to obtain the predicted comprehensive score of the historical trading task.

[0021] Specifically, in this embodiment, the predicted comprehensive score of each historical transaction task is calculated based on the individual scores of each performance indicator and the corresponding dynamic weights, as shown in the following formula (1): in, This refers to the predicted composite score. This refers to the score for a single item in the importance guarantee level of the task. This refers to the score for the duration of the task. This refers to the individual score for calculating resource consumption in a task. This refers to the individual score of the remaining resources in the cluster. This refers to the dynamic weight corresponding to the importance level of the operation. This refers to the dynamic weight corresponding to the job's runtime. This refers to the dynamic weight corresponding to the computational resource consumption of the task. This refers to the dynamic weights corresponding to the remaining resources of the cluster. Since the actual comprehensive scores marked by historical transaction tasks are known, according to the above formula, it can be known that the individual scores of each transaction task are objective data and are related to the actual operation of the transaction task. Therefore, the predicted comprehensive score is related to the dynamic weights. How to calculate an optimal set of dynamic weights to ensure that the error between the predicted comprehensive score and the actual comprehensive score is the main problem to be solved. Thus, when there is a new transaction task, the comprehensive score can be calculated based on the obtained optimal set of dynamic weights. Therefore, in this embodiment, the objective function model shown in the following formula (2) can be constructed based on the individual scores of each performance index of each historical transaction task and the actual comprehensive score: in, Indicates the number of historical assignments. This indicates that the k-th historical transaction is in the... Individual scores for each performance metric i =1,2,3,4 (1 represents the job's importance level, 2 represents the job's runtime, 3 represents the job's computational resource consumption, and 4 represents the cluster's remaining resources) These are the job importance assurance level, job runtime, job computing resource consumption, and dynamic weights of remaining cluster resources. The objective function model described above represents the actual composite score marked by the historical trading task, and is used to characterize the minimization error between the predicted composite score and the actual composite score of the historical trading task.

[0022] Optionally, a chaotic particle swarm optimization (PSO) genetic algorithm is used to iteratively solve the objective function model to determine the target weights of each performance index in the objective function model. This includes: initializing parameters for the objective function model, where the parameters include the maximum number of iterations and the population size; randomly generating a set number of chaotic variables using chaotic mapping and mapping the chaotic variables to the actual value range of the weights, where the set number is greater than the population size; calculating the fitness of each chaotic variable and sorting them in ascending order of fitness to obtain a sequence of chaotic variables, retaining the chaotic variables at the front of the sequence that match the population size as selected individuals, and constructing an initial population based on the selected individuals, where the individuals in the initial population represent the dynamic weight set corresponding to all performance indices; iteratively solving the objective solution using a genetic algorithm and a particle swarm optimization algorithm based on the initial population, and extracting the target weights of each performance index from the objective solution.

[0023] Specifically, this embodiment solves the aforementioned objective function model to determine the target weights of each performance index within the model. The solution process utilizes a hybrid intelligent algorithm combining chaotic mapping, particle swarm optimization, and genetic algorithms, iteratively solving to obtain the target weights. First, the objective function model is initialized with parameters, including the maximum number of iterations and the population size *m*, which determines the number of candidate solutions (i.e., the number of different weight combinations) competing in each iteration. In this embodiment, chaotic variables are introduced into the optimization variables, and the range of chaotic motion is expanded to encompass the value range of the optimization variables. Encoding is then performed, where *n*×m different initial values ​​are randomly selected, where *n* is a positive integer greater than 1. Through chaotic mapping, *n*×m chaotic variables are obtained, and these generated chaotic variables are mapped to the actual value range of the weights (e.g., between 0 and 1). In this implementation, the adaptiveness of each chaotic variable is calculated, and the chaotic variables are sorted in ascending order of adaptiveness to obtain a sequence. From this sequence, m individuals are selected from the front end to form an initial population. Each individual in the initial population represents a set of dynamic weights corresponding to all performance indicators. This implementation applies chaotic mapping to the generation of the initial population, thus generating a high-quality initial population at the beginning of the iterative solution. This allows the algorithm to start its search from a better starting point, thereby improving the efficiency and accuracy of subsequent solutions.

[0024] Optionally, the target weights are obtained by iteratively solving using a genetic algorithm and a particle swarm optimization algorithm based on the initial population. This includes: calculating the fitness of each individual in the current population for each iteration, and updating the current population elite set by comparing the current best with the historical best based on the fitness; obtaining an updated population using a genetic algorithm for the current population elite set, and performing hierarchical perturbation using a particle swarm optimization algorithm on the updated population, wherein the genetic operations include selection, crossover, and mutation; stopping the iteration to obtain the target solution when the maximum number of iterations is reached or the target solution no longer improves after multiple consecutive iterations.

[0025] Optionally, a particle swarm optimization algorithm is used to perform hierarchical perturbation on the updated population, including: dividing individuals in the updated population into three types based on fitness, namely elite individuals in the head region, tail individuals in the tail region, and intermediate individuals in the middle region; performing small perturbations on elite individuals to conduct fine-grained searches near the target solution; keeping intermediate individuals unchanged and directly entering the next generation; determining whether tail individuals are trapped in local traps, and if so, regenerating the perturbation point set for tail individuals; otherwise, updating the velocity and position of tail individuals using the particle swarm optimization algorithm.

[0026] Specifically, in this embodiment, after constructing a high-quality initial population, a genetic algorithm and a particle swarm optimization algorithm are used for iterative solving to obtain a global target solution while avoiding getting trapped in local optima. During the solution process, for each iteration, the best individual (target solution) in this generation is selected based on the fitness of each individual in the current population. The calculated target solution is compared with historical target solutions, and the current population elite set is updated based on the comparison results, thus ensuring that the historical target solution found by the algorithm is not accidentally overwritten by subsequent operations. In this embodiment, standard genetic algorithm operations are performed on the current population elite set to simulate the biological evolution process, specifically including selection, crossover, and mutation. The selection operation selects better individuals, crossover allows two individuals to exchange genes (weights) to generate new individuals, and mutation randomly changes the base of some individuals to increase diversity. Since the principle of genetic algorithm operation is not the focus of this application, it will not be elaborated in this embodiment.

[0027] In this process, after using a genetic algorithm to obtain an updated population from the current elite set, a particle swarm optimization (PSO) algorithm is employed for hierarchical perturbation to further improve the accuracy of the target solution. Specifically, different perturbation strategies are used for different types of individuals. For elite individuals, which are already performing well, only minor adjustments are made in their vicinity to try and find a more perfect extreme point. For tail-end individuals trapped in local traps, they are directly eliminated and replaced with new chaotic variables to escape local optima and explore new regions. For tail-end individuals not trapped in local traps, the velocity and position update formulas of the PSO algorithm are used to accelerate their approach to the current target solution, thus speeding up convergence. Middle-ranking individuals remain unchanged to maintain population stability. For example, individuals in the updated population are sorted according to their fitness, and then classified based on the sorting results: the top 10% are elite individuals, the bottom 20% are tail-end individuals, and the middle 70% are middle-ranking individuals. Different strategies are employed for different types of individuals. Specifically, elite individuals undergo minor perturbations to utilize chaotic variables for a refined search near the target solution, attempting to find a better solution (local development). Intermediate individuals remain unperturbed to maintain stability and follow the main group. Tail-end individuals are first assessed for chaotic operation conditions, i.e., whether they are trapped in a local trap. If trapped, these tail-end individuals are eliminated, and a new perturbation point set is generated. If not trapped, particle swarm optimization is used to update their velocity and position, guiding them towards a better region (global exploration). Therefore, this implementation achieves refined searching for high-quality solutions through layered perturbations, while allowing inferior solutions to quickly find new paths or eliminating them, and keeping intermediate solutions unchanged. Furthermore, this implementation stops iterating to obtain the target solution when the maximum number of iterations is reached or when the target solution no longer improves after multiple consecutive iterations. The target solution includes the four performance indicators mentioned above: job importance guarantee level, job runtime, job computational resource consumption, and the target weight corresponding to the remaining cluster resources.

[0028] S103: Obtain the current individual scores of each performance indicator of the current trading task, calculate the current comprehensive score of the current trading task based on the current individual scores and the target weights of each performance indicator, and adjust the priority of the current trading task based on the current comprehensive score.

[0029] Optionally, the priority of the current trading task can be adjusted based on the current overall score, including: obtaining a priority reference list, wherein the priority reference list includes the correspondence between each priority level and the overall score range; querying the priority reference list based on the current overall score to obtain the target overall score range, and determining the current priority based on the target overall score range; obtaining the historical priority of the current trading task, and adjusting the historical priority based on the current priority.

[0030] Specifically, when a new trading task requires priority adjustment, the current individual scores of each performance indicator of the current trading task are obtained. The method for calculating the current individual scores is roughly the same as the calculation process for the individual scores of the historical trading tasks, so it will not be repeated in this embodiment. Since the target weights of each performance indicator have been calculated, the current comprehensive score of the current trading task can be directly calculated by referring to the above formula (1). In this embodiment, the priority of the current trading task can be adjusted according to the current comprehensive score.

[0031] In this embodiment, a priority reference list can be pre-constructed, as shown in Table 1 below: In this embodiment, after obtaining the current comprehensive score, the target comprehensive score range can be determined by consulting Table 1, i.e., determining which interval of Table 1 the current comprehensive score falls within. Based on the determined target comprehensive score range, the current priority is then determined. For example, when the current comprehensive score is 0.1, the current priority can be directly determined as high by consulting the table. Furthermore, this embodiment also obtains the historical priority of the current transaction task. For example, if it is low, the historical priority is adjusted based on the current priority, i.e., the priority of the current transaction task is adjusted from low to high. Therefore, different priorities determine the start time and allocated resources of the transaction task. Especially in a cluster environment with limited computing resources, such as a cloud server or big data platform, when many transaction tasks are waiting in the queue, priority can be used to determine which transaction task will be executed first and which requires more resources, thereby improving the overall execution efficiency of the system.

[0032] The technical solution of this invention constructs an objective function model based on information from historical trading tasks, and integrates chaotic mapping and particle swarm optimization into a genetic algorithm for solving the model. This yields the target weights for each performance indicator. Chaotic mapping increases ergodicity, while particle weights accelerate convergence, thereby improving the accuracy of the obtained target weights. Based on the target weights and the current comprehensive score determined by the current individual scores of each performance indicator, the priority is automatically adjusted, thus improving the efficiency and accuracy of priority adjustment for the current trading task.

[0033] Example 2 Figure 2 This is a flowchart of another method for adjusting the priority of a trading task according to an embodiment of the present invention. Based on the above embodiment, this embodiment, after adjusting the priority of the current trading task according to the current comprehensive score, further includes detecting the adjusted priority of the current trading task. Figure 2As shown, the method includes: S201, obtain the actual overall score of historical transaction tasks, as well as at least one performance indicator and its corresponding individual score.

[0034] Optionally, the actual comprehensive score of historical trading tasks, as well as at least one performance indicator and its corresponding individual score, are obtained, including: obtaining the actual comprehensive score marked for each historical trading task, and collecting the raw data of each performance indicator of each historical trading task; obtaining the pre-configured score calculation strategy for each performance indicator, and calculating the raw data according to the score calculation strategy for each performance indicator to obtain the individual score corresponding to the performance indicator.

[0035] S202. Based on the individual scores and actual comprehensive scores of each performance indicator of each historical transaction task, the objective function model is solved iteratively using the chaotic particle swarm genetic algorithm to determine the target weight of each performance indicator in the objective function model.

[0036] Optionally, before iteratively solving the objective function model using the chaotic particle swarm genetic algorithm based on the individual scores and actual comprehensive scores of each performance indicator for each historical trading task, the algorithm further includes: calculating the product of the individual scores of each performance indicator and the corresponding dynamic weights for each historical trading task, where the dynamic weights are variables; and summing the product results of each performance indicator to obtain the predicted comprehensive score of the historical trading task.

[0037] Optionally, a chaotic particle swarm optimization (PSO) genetic algorithm is used to iteratively solve the objective function model to determine the target weights of each performance index in the objective function model. This includes: initializing parameters for the objective function model, where the parameters include the maximum number of iterations and the population size; randomly generating a set number of chaotic variables using chaotic mapping and mapping the chaotic variables to the actual value range of the weights, where the set number is greater than the population size; calculating the fitness of each chaotic variable and sorting them in ascending order of fitness to obtain a sequence of chaotic variables, retaining the chaotic variables at the front of the sequence that match the population size as selected individuals, and constructing an initial population based on the selected individuals, where the individuals in the initial population represent the dynamic weight set corresponding to all performance indices; iteratively solving the objective solution using a genetic algorithm and a particle swarm optimization algorithm based on the initial population, and extracting the target weights of each performance index from the objective solution.

[0038] Optionally, the target weights are obtained by iteratively solving using a genetic algorithm and a particle swarm optimization algorithm based on the initial population. This includes: calculating the fitness of each individual in the current population for each iteration, and updating the current population elite set by comparing the current best with the historical best based on the fitness; obtaining an updated population using a genetic algorithm for the current population elite set, and performing hierarchical perturbation using a particle swarm optimization algorithm on the updated population, wherein the genetic operations include selection, crossover, and mutation; stopping the iteration to obtain the target solution when the maximum number of iterations is reached or the target solution no longer improves after multiple consecutive iterations.

[0039] Optionally, a particle swarm optimization algorithm is used to perform hierarchical perturbation on the updated population, including: dividing individuals in the updated population into three types based on fitness, namely elite individuals in the head region, tail individuals in the tail region, and intermediate individuals in the middle region; performing small perturbations on elite individuals to conduct fine-grained searches near the target solution; keeping intermediate individuals unchanged and directly entering the next generation; determining whether tail individuals are trapped in local traps, and if so, regenerating the perturbation point set for tail individuals; otherwise, updating the velocity and position of tail individuals using the particle swarm optimization algorithm.

[0040] S203: Obtain the current individual scores of each performance indicator of the current trading task, calculate the current comprehensive score of the current trading task based on the current individual scores and the target weights of each performance indicator, and adjust the priority of the current trading task based on the current comprehensive score.

[0041] Optionally, the priority of the current trading task can be adjusted based on the current overall score, including: obtaining a priority reference list, wherein the priority reference list includes the correspondence between each priority level and the overall score range; querying the priority reference list based on the current overall score to obtain the target overall score range, and determining the current priority based on the target overall score range; obtaining the historical priority of the current trading task, and adjusting the historical priority based on the current priority.

[0042] S204 checks the priority of the current transaction task after adjustment.

[0043] Specifically, in this embodiment, after adjusting the current trading task, the adjusted priority is checked. Specifically, it checks whether the adjusted priority matches the actual situation. For example, after adjusting the current trading task according to the above adjustment strategy and determining that it has been adjusted to a higher priority, the actual situation of the current trading task in the system is checked. If the detection determines that the current trading task is actually set to a dormant or suspended state in the system, it means that there is no need to allocate resources for the current trading task. Adjusting it to a higher priority is inconsistent with the actual state of the system. This indicates that the previous priority adjustment strategy was incorrect, so an alarm prompt is immediately generated to remind the management personnel to perform maintenance.

[0044] In addition, in this embodiment, after the maintenance is completed by the management personnel, the maintenance record for the current transaction task will be obtained, and the parameters of the chaotic particle swarm genetic algorithm will be adjusted or the hardware will be improved based on the maintenance record to ensure the accuracy of the subsequent transaction task priority adjustment. Of course, this embodiment is only an example and does not limit the specific role of the maintenance record. As long as it can improve the accuracy of the subsequent priority adjustment, it is within the scope of protection of this application.

[0045] The technical solution of this invention constructs an objective function model based on information from historical trading tasks, and integrates chaotic mapping and particle swarm optimization into a genetic algorithm for solving the model. This yields the target weights for each performance indicator. Chaotic mapping increases ergodicity, while particle weights accelerate convergence, thereby improving the accuracy of the obtained target weights. Based on the target weights and the current comprehensive score determined by the current individual scores of each performance indicator, the priority is automatically adjusted, thus improving the efficiency and accuracy of priority adjustment for the current trading task.

[0046] Example 3 Figure 3 This is a schematic diagram of a priority adjustment device for transaction tasks provided in an embodiment of the present invention. Figure 3 As shown, the device includes: a score acquisition module 310, a target weight calculation module 320, and a priority adjustment module 330.

[0047] The score acquisition module 310 is used to acquire the actual comprehensive score of historical transaction jobs, as well as at least one performance indicator and its corresponding individual score. The performance indicator includes at least one of the following: job importance guarantee level, job runtime, job computing resource consumption, and cluster remaining resources. The objective weight solution module 320 is used to iteratively solve the objective function model using a chaotic particle swarm genetic algorithm based on the individual scores and actual comprehensive scores of each performance index of each historical trading task, so as to determine the objective weight of each performance index in the objective function model. The objective function model is used to characterize the minimization error between the predicted comprehensive score and the actual comprehensive score of the historical trading task. The priority adjustment module 330 is used to obtain the current individual scores of each performance indicator of the current trading task, calculate the current comprehensive score of the current trading task based on the current individual scores and the target weights of each performance indicator, and adjust the priority of the current trading task based on the current comprehensive score.

[0048] Optionally, a score acquisition module is used to obtain the actual comprehensive score marked by each historical trading task and to collect the raw data of each performance indicator of each historical trading task. Obtain the pre-configured score calculation strategy for each performance metric. For each performance metric, calculate the original data according to the score calculation strategy to obtain the individual score corresponding to the performance metric.

[0049] Optionally, the device also includes a predictive comprehensive score acquisition module, used to calculate the product of the individual score of each performance indicator and the corresponding dynamic weight for each historical transaction task, wherein the dynamic weight is a variable; The sum of the products of each performance indicator is used to obtain the overall prediction score for historical trading tasks.

[0050] Optionally, the objective weight calculation module includes: an initialization setting unit, used to initialize the parameters of the objective function model, wherein the parameters include the maximum number of iterations and the population size; The chaotic mapping unit is used to randomly generate a set number of chaotic variables using chaotic mapping, and to map the chaotic variables to the actual value range of the weights, wherein the set number is greater than the population size. The initial population construction unit is used to calculate the fitness of each chaotic variable and sort them in order of fitness from small to large to obtain the chaotic variable sequence. The chaotic variables at the front of the chaotic variable sequence that match the population size are retained as screening individuals, and the initial population is constructed based on the screening individuals. The individuals in the initial population represent the dynamic weight set corresponding to all performance indicators. The target weight acquisition unit is used to obtain the target solution by iteratively solving the initial population using genetic algorithm and particle swarm algorithm, and to extract the target weights of each performance index from the target solution.

[0051] Optionally, a target weight acquisition unit is used to calculate the fitness of each individual in the current population for each iteration, and update the current population elite set based on the fitness by comparing the current best with the historical best. A genetic algorithm is used to obtain an updated population from the current elite set of the population, and a particle swarm optimization algorithm is used to perform hierarchical perturbation on the updated population. The genetic operations include selection, crossover and mutation. Stop iterating to obtain the target solution when the maximum number of iterations is reached or when the target solution no longer improves after multiple consecutive iterations.

[0052] Optionally, the target weight acquisition unit is also used to classify individuals in the updated population into three types based on fitness, wherein the types include elite individuals located in the head region, tail individuals located in the tail region, and intermediate individuals located in the middle region. Make small perturbations for elite individuals to conduct a fine search in the vicinity of the target solution; The intermediate individuals remain unchanged and directly enter the next generation; For each tail individual, determine whether it has fallen into a local trap. If so, regenerate a new set of perturbation points for the tail individual; otherwise, use the particle swarm optimization algorithm to update the velocity and position of the tail individual.

[0053] Optionally, a priority adjustment module is used to obtain a priority reference list, wherein the priority reference list includes the correspondence between the priority of each level and the comprehensive score range; Based on the current comprehensive score, query the priority list to obtain the target comprehensive score range, and determine the current priority based on the target comprehensive score range; Obtain the historical priority of the current transaction task and adjust the historical priority according to the current priority.

[0054] The priority adjustment device for transaction tasks provided in this embodiment of the invention can execute the priority adjustment method for transaction tasks provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0055] Example 4 Figure 4 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0056] The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the invention described and / or claimed herein.

[0057] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0058] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other electronic devices through computer networks such as the Internet and / or various telecommunications networks.

[0059] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the priority adjustment method for trading job tasks.

[0060] That is, obtain the actual comprehensive score of historical transaction tasks, as well as at least one performance indicator and its corresponding individual score. The performance indicator includes at least one of the following: job importance guarantee level, job runtime, job computing resource consumption, and cluster remaining resources. Based on the individual scores and actual comprehensive scores of each performance indicator of each historical trading task, the objective function model is iteratively solved using the chaotic particle swarm genetic algorithm to determine the target weights of each performance indicator in the objective function model. The objective function model is used to characterize the minimization error between the predicted comprehensive score and the actual comprehensive score of the historical trading task. Obtain the current individual scores of each performance metric for the current trading task, calculate the current overall score of the current trading task based on the current individual scores and the target weights of each performance metric, and adjust the priority of the current trading task based on the current overall score.

[0061] In some embodiments, the priority adjustment method for trading tasks can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the priority adjustment method for trading tasks described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the priority adjustment method for trading tasks by any other suitable means (e.g., by means of firmware).

[0062] Various embodiments of the apparatuses and techniques described above herein can be implemented in digital electronic circuit devices, integrated circuit devices, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), device-on-a-chip (SoC) devices, complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable device including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage device, at least one input device, and at least one output device, and transmitting data and instructions to the storage device, the at least one input device, and the at least one output device.

[0063] Computer programs used to implement the priority adjustment method for transaction tasks of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to general-purpose computers or special-purpose computers, such that when executed by a processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer programs can be executed entirely on the machine, partially on the machine, or as a standalone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0064] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution apparatus, device, or electronic device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage electronics, magnetic storage electronics, or any suitable combination thereof.

[0065] To provide interaction with a user, the devices and techniques described herein can be implemented on an electronic device having: a display device (e.g., a touchscreen) for displaying information to the user; and buttons through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0066] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0067] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for prioritizing transaction tasks, characterized in that, The method includes: Obtain the actual overall score of historical transaction jobs, as well as at least one performance indicator and its corresponding individual score, wherein the performance indicator includes at least one of the following: job importance guarantee level, job runtime, job computing resource consumption, and cluster remaining resources; Based on the individual scores of each performance indicator of each historical trading task and the actual comprehensive score, a chaotic particle swarm genetic algorithm is used to iteratively solve the objective function model to determine the target weight of each performance indicator in the objective function model. The objective function model is used to characterize the minimization error between the predicted comprehensive score and the actual comprehensive score of the historical trading task. Obtain the current individual scores of each performance metric for the current trading task, calculate the current overall score of the current trading task based on the current individual scores and the target weights of each performance metric, and adjust the priority of the current trading task based on the current overall score.

2. The method according to claim 1, characterized in that, The process of obtaining the actual overall score of historical transaction tasks, as well as at least one performance indicator and its corresponding individual score, includes: Obtain the actual comprehensive score marked by each historical transaction task, and collect the raw data of each performance indicator of each historical transaction task; Obtain a pre-configured score calculation strategy for each of the performance metrics, and calculate the original data according to the score calculation strategy for each performance metric to obtain the individual score corresponding to the performance metric.

3. The method according to claim 1, characterized in that, Before iteratively solving the objective function model using the chaotic particle swarm genetic algorithm based on the individual scores of each performance indicator of each historical transaction task and the actual comprehensive score, the method further includes: For each of the aforementioned historical transaction tasks, calculate the product of the individual score of each performance indicator and its corresponding dynamic weight, where the dynamic weight is a variable; The predicted comprehensive score of the historical transaction task is obtained by adding the product results of the various performance indicators.

4. The method according to claim 2, characterized in that, The step of iteratively solving the objective function model using a chaotic particle swarm genetic algorithm to determine the target weights of each performance index in the objective function model includes: The parameters of the objective function model are initialized, including the maximum number of iterations and the population size. A set number of chaotic variables are randomly generated using chaotic mapping, and the chaotic variables are mapped to the actual value range of the weights, wherein the set number is greater than the population size; Calculate the fitness of each chaotic variable and sort them in ascending order of fitness to obtain a chaotic variable sequence. Keep the chaotic variables at the front of the chaotic variable sequence that match the population size as screening individuals, and construct an initial population based on the screening individuals. The individuals in the initial population represent the dynamic weight set corresponding to all performance indicators. Based on the initial population, a genetic algorithm and a particle swarm optimization algorithm are used to iteratively solve for the target solution, and the target weights of each performance index are extracted from the target solution, wherein the target solution is a set of target weights of each performance index.

5. The method according to claim 4, characterized in that, The step of using a genetic algorithm and a particle swarm optimization algorithm to iteratively solve for the target solution based on the initial population includes: For each iteration, the fitness of each individual in the current population is calculated, and the current elite set is updated by comparing the current best with the historical best based on the fitness. A genetic algorithm is used to obtain an updated population for the current elite set of the population, and a particle swarm optimization algorithm is used to perform hierarchical perturbation on the updated population. The genetic operations include selection, crossover, and mutation. The iteration should stop when the maximum number of iterations is reached or when the target solution no longer improves after multiple consecutive iterations.

6. The method according to claim 5, characterized in that, The step of performing layered perturbation on the updated population using the particle swarm optimization algorithm includes: Individuals in the updated population are divided into three types based on fitness: elite individuals in the head region, tail individuals in the tail region, and intermediate individuals in the middle region. A small perturbation is applied to the elite individual to perform a fine search in the vicinity of the target solution; The intermediate individuals remain unchanged and directly enter the next generation; For each tail individual, it is determined whether it has fallen into a local trap. If so, a new set of perturbation points is generated for the tail individual; otherwise, the velocity and position of the tail individual are updated using the particle swarm optimization algorithm.

7. The method according to claim 1, characterized in that, The step of adjusting the priority of the current transaction task based on the current overall score includes: Obtain a priority reference list, wherein the priority reference list includes the correspondence between each level of priority and the comprehensive score range; Based on the current comprehensive score, query the priority reference list to obtain the target comprehensive score range, and determine the current priority based on the target comprehensive score range; Obtain the historical priority of the current transaction task, and adjust the historical priority according to the current priority.

8. A priority adjustment device for transaction operation tasks, characterized in that, The device includes: The score acquisition module is used to acquire the actual comprehensive score of historical transaction jobs, as well as at least one performance indicator and its corresponding individual score. The performance indicator includes at least one of the following: job importance guarantee level, job runtime, job computing resource consumption, and cluster remaining resources. The objective weight solution module is used to iteratively solve the objective function model using a chaotic particle swarm genetic algorithm based on the individual scores of each performance indicator of each historical trading task and the actual comprehensive score, so as to determine the objective weight of each performance indicator in the objective function model. The objective function model is used to characterize the minimized error between the predicted comprehensive score and the actual comprehensive score of the historical trading task. The priority adjustment module is used to obtain the current individual scores of each performance indicator of the current trading task, calculate the current comprehensive score of the current trading task based on the current individual scores and the target weights of each performance indicator, and adjust the priority of the current trading task based on the current comprehensive score.

9. An electronic device, characterized in that, The electronic device includes: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-7.

10. A storage medium for computer-executable instructions, wherein a computer program is stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-7.

Images