WO2020186787A1 - 智能任务调度方法、装置、设备及存储介质 - Google Patents

智能任务调度方法、装置、设备及存储介质 Download PDF

Info

Publication number
WO2020186787A1
WO2020186787A1 PCT/CN2019/118370 CN2019118370W WO2020186787A1 WO 2020186787 A1 WO2020186787 A1 WO 2020186787A1 CN 2019118370 W CN2019118370 W CN 2019118370W WO 2020186787 A1 WO2020186787 A1 WO 2020186787A1
Authority
WO
WIPO (PCT)
Prior art keywords
task
tasks
scheduled
preset
dependency
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2019/118370
Other languages
English (en)
French (fr)
Inventor
陈万慧
汪伟
简杰生
苏雪婷
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ping An Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Publication of WO2020186787A1 publication Critical patent/WO2020186787A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/48—Program initiating; Program switching, e.g. by interrupt
    • G06F9/4806—Task transfer initiation or dispatching
    • G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
    • G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues

Definitions

  • This application relates to the field of artificial intelligence, and in particular to an intelligent task scheduling method, device, equipment and storage medium.
  • the existing scheduling system will perform a large number of tasks, many of which are data output and related important information alarm output.
  • the existing scheduling method is to specify the highest priority task after setting the level. Recursively find upstream tasks and set their corresponding task priorities, and sort the task data sets, but there will be uncontrolled low recursion levels in the recursive process, resulting in data anomalies, which will affect the performance of the algorithm and user experience; existing scheduling The method needs to re-order the algorithmic calculation task priority every time it is scheduled, which will generate redundant calculations and cause low office efficiency.
  • the main purpose of this application is to provide an intelligent task scheduling method, device, equipment, and storage medium, aiming to solve the technical problems of low office efficiency and poor user experience due to repetitive calculation and uncontrolled recursive levels in the prior art.
  • the intelligent task scheduling method includes the following steps:
  • the tasks to be scheduled are scheduled according to the preset task execution sequence until all the tasks to be scheduled are completed.
  • FIG. 1 is a schematic diagram of the structure of an intelligent task scheduling device in a hardware operating environment involved in a solution of an embodiment of the application;
  • FIG. 3 is a schematic flowchart of a second embodiment of the intelligent task scheduling method of this application.
  • FIG. 4 is a schematic flowchart of a third embodiment of a smart task scheduling method according to this application.
  • Fig. 5 is a functional module diagram of the first embodiment of the intelligent task scheduling device of this application.
  • FIG. 1 is a schematic structural diagram of a smart task scheduling device in a hardware operating environment involved in a solution of an embodiment of the application.
  • the intelligent task scheduling device may include: a processor 1001, such as a central processing unit (Central Processing Unit, CPU), communication bus 1002, user interface 1003, network interface 1004, memory 1005.
  • the communication bus 1002 is used to implement connection and communication between these components.
  • the user interface 1003 may include a standard wired interface and a wireless interface.
  • the network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (WIreless-FIdelity, WI-FI) interface).
  • the memory 1005 may be a high-speed random access memory (Random Access Memory, RAM) memory, can also be a stable memory (Non-volatile Memory, NVM), such as disk storage.
  • the memory 1005 may also be a storage device independent of the foregoing processor 1001.
  • the structure of the smart task scheduling device shown in FIG. 1 does not constitute a limitation on the smart task scheduling device, and may include more or less components than shown in the figure, or combine certain components, or Different component arrangements.
  • the memory 1005 as a storage medium may include an operating device, a network communication module, a user-side interface module, and computer readable instructions.
  • the smart task scheduling device executes the application.
  • the smart task scheduling method provided by the embodiment.
  • FIG. 2 is a schematic flowchart of the first embodiment of the intelligent task scheduling method of this application.
  • the smart task scheduling method includes the following steps:
  • Step S101 Obtain multiple tasks to be scheduled that need to be scheduled, analyze each task to be scheduled, and obtain an analysis result;
  • Step S102 Determine, according to the analysis result, the number of dependencies each task to be scheduled depends on other tasks to be scheduled during execution;
  • Step S103 Find the association degree corresponding to each dependency quantity according to the preset association degree data table, and the preset association degree data table reflects the mapping relationship between the dependency quantity and the association degree.
  • the task to be scheduled is a pre-set task that needs to be executed or scheduled or a task that needs to be executed and scheduled to meet the scheduling adjustment needs, and the degree of association between the tasks to be scheduled is that each task to be scheduled is related to other tasks.
  • the degree of relevance of tasks that is, whether each task to be scheduled is related to other tasks during execution, and the degree of relevance is used to indicate the relationship between each task to be scheduled and other tasks to be scheduled during execution.
  • each to-be-scheduled task can be analyzed, and it can be determined whether any task depends on other tasks during execution or when it is executed by other tasks.
  • the number of other tasks and/or the number of other tasks on which the execution of other tasks depends determines the degree of association between each task to be scheduled and other tasks to be scheduled.
  • the preset association degree data table is a preset data table for obtaining association degree.
  • the preset relevance data table reflects the mapping relationship between the number of dependencies and relevance.
  • the preset relevance data table may be obtained through a large amount of experimental data training, or may be a data table determined by a technician based on daily experience. It may also be a data table determined in other ways, which is not limited in this embodiment.
  • Step S20 Determine, according to the correlation degree, a plurality of pre-tasks with pre-dependency among the tasks to be scheduled.
  • the predecessor task is a task with predetermination dependencies among the tasks to be scheduled
  • the predetermination dependency is the predecessor of another or other tasks to be scheduled among the tasks to be scheduled.
  • the dependency relationship of the tasks can be used to determine multiple pre-tasks with pre-dependence among the tasks to be scheduled through the correlation degree.
  • step S20 includes the following steps:
  • the remaining to-be-scheduled tasks in each to-be-scheduled task are regarded as multiple pre-tasks with pre-dependency.
  • Scheduling tasks are screened to filter out the tasks to be scheduled whose relevance is less than the preset relevance threshold, and the remaining tasks to be scheduled are pre-reliant tasks; the preset relevance threshold is The pre-set relevance threshold, by comparing the relevance of each task to be scheduled with the preset relevance threshold, it is possible to filter out the tasks to be scheduled with less relevance among the tasks to be scheduled, and the remaining tasks to be scheduled.
  • the task is treated as multiple pre-tasks with pre-dependency.
  • Step S30 Obtain the degree of dependency of each predecessor task, and find the predecessor task with the highest degree of dependency.
  • the predecessor tasks with the highest level of dependency can be obtained by sorting the dependency levels from high to low according to the degree of dependence, and of course it can also be based solely on the predecessor tasks.
  • the degree of relevance determines the degree of dependency of each predecessor task. After obtaining the degree of dependency of each predecessor task, the degree of dependency of each predecessor task can be sorted from high to low to obtain the predecessor with the highest degree of dependency; of course, it is also possible
  • the predecessor tasks in the highest level of dependency are obtained by other methods, which are not limited in this embodiment.
  • step S30 includes the following steps:
  • the historical execution data is the execution data of each predecessor task in a preset period.
  • the historical dependency frequency of each predecessor task is obtained, and the historical dependency frequency is The frequency that each predecessor task is dependent on, the dependence degree of each predecessor task can be determined by the historical dependence frequency and the degree of relevance, and the dependence degree can be sorted through the preset sort order, so as to determine the predecessor with the highest degree of dependence task.
  • Step S40 Obtain the priority levels of target tasks other than the predecessor tasks among the tasks to be scheduled, and construct a preset task execution sequence according to the degree of dependence and the priority level.
  • the target tasks other than the predecessor tasks are other tasks for which each task to be scheduled has no prerequisite dependencies.
  • the scheduling priority of each task in the target task can be determined. According to the degree of dependence and the priority level, each task to be scheduled can be determined, that is, the execution scheduling priority level of each task in the task to be scheduled is determined.
  • Step S50 Schedule each task to be scheduled according to the preset task execution sequence until all tasks to be scheduled are scheduled.
  • the preset task execution order is a preset execution order for executing tasks to be scheduled, and the preset task execution order can be constructed by the degree of dependency and the priority level, and of course it can also be Confirmed by other means, this embodiment does not impose restrictions on this; each task to be scheduled can be scheduled according to the preset task execution sequence until all tasks are scheduled.
  • multiple tasks to be scheduled that need to be scheduled are obtained, and the degree of association between each task to be scheduled is obtained; according to the degree of association, multiple pre-dependent tasks in each task to be scheduled are determined.
  • This can prevent other non-important tasks from competing with important tasks for resources, so that important tasks can get the most Excellent execution rights ensure the priority execution of important tasks, facilitate the development of follow-up tasks, improve work speed and efficiency, and enhance user experience.
  • FIG. 3 is a schematic flowchart of the second embodiment of the intelligent task scheduling method of this application.
  • the second embodiment of the intelligent task scheduling method of this application is proposed based on the first embodiment.
  • the step S40 specifically includes the following steps:
  • Step S41 Sort the target tasks other than the predecessor tasks among the tasks to be scheduled according to the preset insertion sorting algorithm, and obtain the sorting order of the target tasks.
  • the preset insertion sorting algorithm is a preset insertion algorithm. Generally, a piece of data is inserted into the sorted ordered data, so as to obtain a new ordered data with the number plus one. , Divide the array to be sorted into two parts: the first part contains all the elements of the array, except for the last element (let the array have one more space for insertion), and the second part contains only this element ( That is, the element to be inserted); after the sorting of the first part is completed, the last element is inserted into the sorted first part; the preset insertion sorting algorithm is used to sort all the tasks to be scheduled except the predecessor Target tasks are sorted; according to the preset insertion sorting algorithm, target tasks other than the predecessor tasks among the tasks to be scheduled can be sorted to obtain the sort order of the target tasks.
  • Step S42 Determine the priority level of the target task according to the sort order.
  • the order of execution of the target tasks can be obtained according to the sorting order of the target tasks, so as to determine the priority level of the target tasks.
  • Step S43 Construct a preset task execution order according to the degree of dependency and the priority level.
  • the execution order of the preset task is that the execution order of the predecessor task with the highest degree of dependency has priority over the target task with the highest level.
  • the execution order of the target tasks of a level has priority over the predecessor tasks of other levels, and the execution order of the predecessor tasks of the other levels has priority over the target tasks of other levels.
  • a preset task execution order is constructed according to the degree of dependency and the priority level, the preset task execution order is a preset order, and the preset task execution order is the predecessor with the highest degree of dependency.
  • the execution order of tasks takes precedence over the highest-level target task, the execution order of the highest-level target task takes precedence over the predecessor tasks of other levels, and the execution order of the predecessor tasks of the other levels takes precedence over the target tasks of other levels, That is, the priority of the predecessor task of the highest level of the goal is higher than the task of the highest level of the goal, the task of the highest level of the goal is higher than the predecessor tasks of other levels, and the predecessor tasks of the other levels are higher than the ordered tasks of other levels.
  • the priority level of general tasks other than the predecessor task in each task to be scheduled is obtained, that is, the original execution priority level of the target task, that is, other tasks to be scheduled except the predecessor task also have the original Execution order, according to the dependency of each pre-task and the priority of template tasks, the preset task execution order can be constructed, that is, the scheduling order of each task to be scheduled can be determined according to the preset task execution order, that is, the execution order can be set Because the execution order of the predecessor tasks with the highest degree of dependency takes precedence over the target task at the highest level, the execution order of the target task at the highest level takes precedence over the predecessor tasks at other levels, and the execution order of the predecessor tasks at the other levels Priority to other levels of target tasks.
  • the target tasks other than the predecessor tasks among the tasks to be scheduled are sorted by a preset insertion sort algorithm to obtain the sort order of the target tasks; the target tasks are determined according to the sort order.
  • the priority level of the task; a preset task execution order is constructed according to the dependency level and the priority level, and the preset task execution order is that the execution order of the predecessor task with the highest dependency level has priority over the target task at the highest level, so
  • the execution order of the highest-level target task takes precedence over the predecessor tasks of other levels, and the execution order of the predecessor tasks of the other levels takes precedence over the target tasks of other levels, which can further accurately obtain the priority of the execution tasks and avoid other tasks.
  • Non-important tasks compete with important tasks for resources, so that important tasks obtain optimal execution rights, ensure the priority execution of important tasks, facilitate the development of follow-up tasks, improve work speed and efficiency, and enhance user experience.
  • FIG. 4 is a schematic flowchart of the third embodiment of the intelligent task scheduling method of this application.
  • the third embodiment of the intelligent task scheduling method of this application is proposed based on the second embodiment.
  • the smart task scheduling method further includes the following steps:
  • Step S51 Obtain the number of tasks for each task to be scheduled, and determine the target thread pool according to the number of tasks.
  • each task to be scheduled corresponds to a corresponding number of tasks
  • the target thread pool is a thread pool that executes the task to be scheduled.
  • the required thread pool can be determined by the number of tasks, and the number of different tasks determines different totals. The thread pool.
  • step S51 specifically includes the following steps:
  • the idle degree and task processing speed of each thread pool in the preset thread pool can be obtained, and the idle degree can be determined according to the comparison result of the idle degree and the preset idle degree.
  • a thread pool, so that a thread pool matching the number of tasks can be selected from each idle thread pool as the target thread pool.
  • Step S52 Select a corresponding thread in the target thread pool to execute each task to be scheduled according to the preset task execution order, until all tasks to be scheduled are completed.
  • the number of tasks for each task to be scheduled is obtained, and each task to be scheduled is scheduled in the target thread pool according to the preset task execution order, until all tasks to be scheduled are scheduled, the target thread pool
  • the determination is based on the idleness and task processing capacity of each thread in the preset thread pool.
  • the idleness and task processing speed of each thread are obtained in real time, and the idle threads and processing speed Faster threads are used as a new CPU scheduling unit for processing tasks, and task scheduling is performed according to the preset task execution sequence until all tasks are scheduled.
  • the target thread pool is determined according to the number of tasks by obtaining the number of tasks for each task to be scheduled; corresponding threads are selected from the target thread pool to execute each task to be scheduled according to the preset task execution order Tasks, until all tasks to be scheduled are completed, it can prevent other non-important tasks from competing for resources with important tasks, so that important tasks can be optimally executed, ensuring the priority execution of important tasks, conducive to the development of subsequent tasks, and improving work Speed and efficiency improve the user experience.
  • the present application further provides a smart task scheduling device.
  • Fig. 5 is a functional module diagram of the first embodiment of the intelligent task scheduling device of this application.
  • the smart task scheduling device includes: an association degree acquisition 10, a pre-task determination module 20, a dependency degree acquisition module 30, a sequence determination module 40, and a scheduling module 50;
  • the degree of relevance obtaining module 10 is used to obtain multiple tasks to be scheduled that need to be scheduled, and to obtain the degree of relevance between the tasks to be scheduled.
  • the task to be scheduled is a pre-set task that needs to be executed or scheduled or a task that needs to be executed and scheduled to meet the scheduling adjustment needs, and the degree of association between the tasks to be scheduled is that each task to be scheduled is related to other tasks.
  • the degree of relevance of tasks that is, whether each task to be scheduled is related to other tasks during execution, and the degree of relevance is used to indicate the relationship between each task to be scheduled and other tasks to be scheduled during execution.
  • each to-be-scheduled task can be analyzed, and it can be determined whether any task depends on other tasks during execution or when it is executed by other tasks.
  • the number of other tasks and/or the number of other tasks on which the execution of other tasks depends determines the degree of association between each task to be scheduled and other tasks to be scheduled.
  • the correlation degree obtaining module 10 includes:
  • the analysis module is used to obtain multiple tasks to be scheduled that need to be scheduled, analyze each task to be scheduled, and obtain analysis results.
  • the dependent quantity determining module is used to determine the dependent quantity of each task to be scheduled depending on other tasks to be scheduled during execution according to the analysis result.
  • the correlation degree search module is used to find the correlation degree corresponding to each dependency quantity according to the preset correlation degree data table, and the preset correlation degree data table reflects the mapping relationship between the dependency quantity and the correlation degree.
  • the preset association degree data table is a preset data table for obtaining association degree.
  • the preset relevance data table reflects the mapping relationship between the number of dependencies and relevance.
  • the preset relevance data table may be obtained through a large amount of experimental data training, or may be a data table determined by a technician based on daily experience. It may also be a data table determined in other ways, which is not limited in this embodiment.
  • the pre-task determining module 20 is configured to determine multiple pre-tasks with pre-dependency among the tasks to be scheduled according to the degree of association.
  • the predecessor task is a task with predetermination dependencies among the tasks to be scheduled
  • the predetermination dependency is the predecessor of another or other tasks to be scheduled among the tasks to be scheduled.
  • the dependency relationship of the tasks can be used to determine multiple pre-tasks with pre-dependence among the tasks to be scheduled through the correlation degree.
  • the pre-task determination module 20 includes:
  • the screening module is used to screen each to-be-scheduled task, and screen out the to-be-scheduled tasks whose relevance degree is less than the preset relevance degree threshold.
  • the pre-task screening module is used to treat the remaining to-be-scheduled tasks among the to-be-scheduled tasks as multiple pre-tasks with pre-dependence.
  • Scheduling tasks are screened to filter out the tasks to be scheduled whose relevance is less than the preset relevance threshold, and the remaining tasks to be scheduled are pre-reliant tasks; the preset relevance threshold is The pre-set relevance threshold, by comparing the relevance of each task to be scheduled with the preset relevance threshold, it is possible to filter out the tasks to be scheduled with less relevance among the tasks to be scheduled, and the remaining tasks to be scheduled.
  • the task is treated as multiple pre-tasks with pre-dependency.
  • the dependency degree obtaining module 30 is used to obtain the dependency degree of each predecessor task, and find the predecessor task with the highest degree of dependency.
  • the predecessor tasks with the highest level of dependency can be obtained by sorting the dependency levels from high to low according to the degree of dependence, and of course it can also be based solely on the predecessor tasks.
  • the degree of relevance determines the degree of dependency of each predecessor task. After obtaining the degree of dependency of each predecessor task, the degree of dependency of each predecessor task can be sorted from high to low to obtain the predecessor with the highest degree of dependency; of course, it is also possible
  • the predecessor tasks in the highest level of dependency are obtained by other methods, which is not limited in this embodiment.
  • the dependence degree obtaining module 30 includes:
  • the dependent frequency acquisition module is used to acquire historical execution data of each pre-task, analyze the historical execution data, and obtain the historical dependence frequency of each pre-task, where the historical dependence frequency is the frequency at which each pre-task is dependent ;
  • the dependence degree determination module is used to determine the dependence degree of each predecessor task according to the historical dependence frequency and correlation degree of each predecessor task;
  • the dependency sorting module is used to sort the dependency levels according to the preset sort order to obtain the pre-task with the highest dependency level.
  • the historical execution data is the execution data of each predecessor task in a preset period.
  • the historical dependency frequency of each predecessor task is obtained, and the historical dependency frequency is The frequency that each predecessor task is dependent on, the dependence degree of each predecessor task can be determined by the historical dependence frequency and the degree of relevance, and the dependence degree can be sorted through the preset sort order, so as to determine the predecessor with the highest degree of dependence task.
  • the sequence determination module 40 is configured to obtain the priority levels of target tasks other than the predecessor tasks among the multiple to-be-scheduled tasks, and construct a preset task execution sequence according to the degree of dependence and the priority level.
  • the target tasks other than the predecessor tasks are other tasks for which each task to be scheduled has no prerequisite dependencies.
  • the scheduling priority of each task in the target task can be determined. According to the degree of dependence and the priority level, each task to be scheduled can be determined, that is, the execution scheduling priority level of each task in the task to be scheduled is determined.
  • sequence determining module 40 includes:
  • the target task sorting module is used to sort target tasks other than the predecessor tasks among the to-be-scheduled tasks according to a preset insertion sorting algorithm to obtain the sorting order of the target tasks.
  • the preset insertion sorting algorithm is a preset insertion algorithm. Generally, a piece of data is inserted into the sorted ordered data, so as to obtain a new ordered data with the number plus one. , Divide the array to be sorted into two parts: the first part contains all the elements of the array, except for the last element (let the array have one more space for insertion), and the second part contains only this element ( That is, the element to be inserted); after the sorting of the first part is completed, the last element is inserted into the sorted first part; the preset insertion sorting algorithm is used to sort all the tasks to be scheduled except the predecessor Target tasks are sorted; according to the preset insertion sorting algorithm, target tasks other than the predecessor tasks among the tasks to be scheduled can be sorted to obtain the sort order of the target tasks.
  • the target task priority acquiring module is configured to determine the priority of the target task according to the sorting sequence.
  • the order of execution of the target tasks can be obtained according to the sorting order of the target tasks, so as to determine the priority level of the target tasks.
  • the task execution order construction module is used to construct a preset task execution order according to the degree of dependency and the priority level.
  • the execution order of the preset task is that the execution order of the predecessor task with the highest degree of dependency takes precedence over the target with the highest level For tasks, the execution order of the highest-level target tasks has priority over other levels of predecessor tasks, and the execution order of the other levels of predecessor tasks has priority over other levels of target tasks.
  • a preset task execution order is constructed according to the degree of dependency and the priority level, the preset task execution order is a preset order, and the preset task execution order is the predecessor with the highest degree of dependency.
  • the execution order of tasks takes precedence over the highest-level target task, the execution order of the highest-level target task takes precedence over the predecessor tasks of other levels, and the execution order of the predecessor tasks of the other levels takes precedence over the target tasks of other levels, That is, the priority of the predecessor task of the highest level of the goal is higher than the task of the highest level of the goal, the task of the highest level of the goal is higher than the predecessor tasks of other levels, and the predecessor tasks of the other levels are higher than the ordered tasks of other levels.
  • the priority level of general tasks other than the predecessor task in each task to be scheduled is obtained, that is, the original execution priority level of the target task, that is, other tasks to be scheduled except the predecessor task also have the original Execution order, according to the dependency of each pre-task and the priority of template tasks, the preset task execution order can be constructed, that is, the scheduling order of each task to be scheduled can be determined according to the preset task execution order, that is, the execution order can be set Because the execution order of the predecessor tasks with the highest degree of dependency has priority over the target task of the highest level, the execution order of the target task of the highest level has priority over the predecessor tasks of other levels, and the execution order of the predecessor tasks of the other levels Prioritize the target tasks of other levels.
  • the scheduling module 50 is configured to schedule each task to be scheduled according to the preset task execution sequence until all tasks to be scheduled are completed.
  • the preset task execution order is a preset execution order for executing tasks to be scheduled, and the preset task execution order can be constructed by the degree of dependency and the priority level, and of course it can also be Confirmed by other means, this embodiment does not impose restrictions on this; each task to be scheduled can be scheduled by preset task execution sequence until all tasks are scheduled.
  • scheduling module 50 includes:
  • the thread pool determination module is used to obtain the number of tasks for each task to be scheduled, and determine the target thread pool according to the number of tasks.
  • each task to be scheduled corresponds to a corresponding number of tasks
  • the target thread pool is a thread pool that executes the task to be scheduled.
  • the required thread pool can be determined by the number of tasks, and the number of different tasks determines different totals. The thread pool.
  • the thread pool determining module includes:
  • the task quantity obtaining module is used to obtain the task quantity of each task to be scheduled.
  • the processing speed acquisition module is used to acquire the idleness and task processing speed of each thread pool in the preset thread pool.
  • the idle thread pool determining module is configured to use a thread pool whose idle degree is greater than a preset idle degree and whose task processing speed is greater than a preset processing speed as an idle thread pool.
  • the target thread pool matching module is used to select a thread pool matching the number of tasks from each idle thread pool as the target thread pool.
  • the idle degree and task processing speed of each thread pool in the preset thread pool can be obtained, and the idle degree can be determined according to the comparison result of the idle degree and the preset idle degree.
  • a thread pool, so that a thread pool matching the number of tasks can be selected from each idle thread pool as the target thread pool.
  • scheduling module 50 further includes:
  • the scheduling execution module is configured to select corresponding threads in the target thread pool to execute each task to be scheduled according to the preset task execution order, until all tasks to be scheduled are completed.
  • the number of tasks for each task to be scheduled is obtained, and each task to be scheduled is scheduled in the target thread pool according to the preset task execution order, until all tasks to be scheduled are scheduled, the target thread pool
  • the determination is based on the idleness and task processing capacity of each thread in the preset thread pool.
  • the idleness and task processing speed of each thread are obtained in real time, and the idle threads and processing speed Faster threads are used as a new CPU scheduling unit for processing tasks, and task scheduling is performed according to the preset task execution sequence until all tasks are scheduled.
  • multiple tasks to be scheduled that need to be scheduled are obtained, and the degree of association between each task to be scheduled is obtained; according to the degree of association, multiple pre-dependent tasks in each task to be scheduled are determined.
  • This can prevent other non-important tasks from competing with important tasks for resources, so that important tasks can get the most Excellent execution rights ensure the priority execution of important tasks, facilitate the development of follow-up tasks, improve work speed and efficiency, and enhance user experience.
  • an embodiment of the present application also proposes a storage medium.
  • the storage medium of the present application may be a non-volatile computer-readable storage medium.
  • the storage medium stores computer-readable instructions, and the computer-readable instructions are processed.
  • the smart task scheduling method described above is implemented when the device is executed.

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Debugging And Monitoring (AREA)

Abstract

一种智能任务调度方法、装置、设备及存储介质，获取需要进行调度的多个待调度任务，对各待调度任务进行分析，并获得分析结果（S101）；根据所述分析结果确定各待调度任务在执行时依赖于其他待调度任务的依赖数量（S102）；根据预设关联度数据表查找到与各依赖数量对应的关联度，所述预设关联度数据表反映依赖数量与关联度的映射关系（S103）；根据所述关联度确定各待调度任务中具有前置依赖的多个前置任务（S20）；获取各前置任务的依赖程度，找到依赖程度最高级别的前置任务（S30）；获取各待调度任务中除了前置任务之外的目标任务的优先级别，根据所述依赖程度和所述优先级别构建预设任务执行顺序（S40）；根据所述预设任务执行顺序对各待调度任务进行调度，直至调度完成所有待调度任务（S50）。该方法可以避免其他非重要的任务与重要任务竞争资源，使重要任务获得最优执行权，保证了重要任务的优先执行，有利于后续任务的开展，提高了工作速度和效率，提升了用户体验。

Description

智能任务调度方法、装置、设备及存储介质
本申请要求于2019年03月16日提交中国专利局、申请号为201910206130.3、发明名称为“智能任务调度方法、设备、存储介质及装置”的中国专利申请的优先权，其全部内容通过引用结合在申请中。
技术领域
本申请涉及人工智能领域，尤其涉及一种智能任务调度方法、装置、设备及存储介质。
背景技术
在日常办公过程中，现有调度系统会有大量任务进行执行，其中有不少任务都是数据产出和相关重要信息告警输出的，现有的调度方式是对指定最优先任务设置级别后，递归找到上游任务并对其进行相应任务优先级设置，并进行任务数据集排序，但是在递归过程中会存在未控制低递归层级导致数据异常从而影响算法的性能和用户使用体验；现有的调度方式每次调度都需要重新进行算法计算任务优先级排序，会产生多余的计算量，造成办公效率较低。
发明内容
本申请的主要目的在于提供一种智能任务调度方法、装置、设备及存储介质，旨在解决现有技术中由于重复计算和递归层级未控制导致办公效率低下，用户使用体验差的技术问题。
为实现上述目的，本申请提供一种智能任务调度方法，所述智能任务调度方法包括以下步骤：
获取需要进行调度的多个待调度任务，并获得各待调度任务之间的关联度；
根据所述关联度确定各待调度任务中具有前置依赖的多个前置任务；
获取各前置任务的依赖程度，找到依赖程度最高级别的前置任务；
获取各待调度任务中除了前置任务之外的目标任务的优先级别，根据所述依赖程度和所述优先级别构建预设任务执行顺序；
根据所述预设任务执行顺序对各待调度任务进行调度，直至调度完成所有待调度任务。
本申请的一个或多个实施例的细节在下面的附图和描述中提出，本申请的其它特征和优点将从说明书、附图以及权利要求书变得明显 。
附图说明
图1为本申请实施例方案涉及的硬件运行环境的智能任务调度设备结构示意图；
图2为本申请智能任务调度方法第一实施例的流程示意图；
图3为本申请智能任务调度方法第二实施例的流程示意图；
图4为本申请智能任务调度方法第三实施例的流程示意图；
图5为本申请智能任务调度装置第一实施例的功能模块图。
本申请目的实现、功能特点及优点将结合实施例，参照附图做进一步说明。
具体实施方式
应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
参照图1，图1为本申请实施例方案涉及的硬件运行环境的智能任务调度设备结构示意图。
如图1所示，该智能任务调度设备可以包括：处理器1001，例如中央处理器（Central Processing Unit，CPU），通信总线1002、用户接口1003，网络接口1004，存储器1005。其中，通信总线1002用于实现这些组件之间的连接通信。用户接口1003可以包括标准的有线接口、无线接口。网络接口1004可选的可以包括标准的有线接口、无线接口（如无线保真（WIreless-FIdelity，WI-FI）接口）。存储器1005可以是高速的随机存取存储器（Random Access Memory，RAM）存储器，也可以是稳定的存储器（Non-volatile Memory，NVM），例如磁盘存储器。存储器1005可选的还可以是独立于前述处理器1001的存储装置。
本领域技术人员可以理解，图1中示出的智能任务调度设备结构并不构成对该智能任务调度设备的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。
如图1所示，作为一种存储介质的存储器1005中可以包括操作装置、网络通信模块、用户端接口模块以及计算机可读指令，所述智能任务调度设备，所述智能任务调度设备执行本申请实施例提供的智能任务调度方法。
基于上述硬件结构，提出本申请智能任务调度方法实施例。
参照图2，图2为本申请智能任务调度方法第一实施例的流程示意图。
在第一实施例中，所述智能任务调度方法包括以下步骤：
步骤S101、获取需要进行调度的多个待调度任务，对各待调度任务进行分析，并获得分析结果；
步骤S102、根据所述分析结果确定各待调度任务在执行时依赖于其他待调度任务的依赖数量；
步骤S103、根据预设关联度数据表查找到与各依赖数量对应的关联度，所述预设关联度数据表反映依赖数量与关联度的映射关系。
需要说明的是，所述待调度任务为预先设置的需要进行执行调度的任务或者满足了调度调节需要进行执行调度的任务，所述各待调度任务之间的关联度为各待调度任务与其他任务的关联程度，即各待调度任务在执行时是否与其他任务有关联，所述关联度用来表示各待调度任务在执行时与其他待调度任务的关系。
在具体实现中，对于需要进行调度的多个待调度任务，可以对各待调度任务进行分析，可以确定任一任务是否在执行时依赖于其他任务或被其他任务执行时所依赖，根据依赖于其他任务的数量，和/或被其他任务执行所依赖的其他任务的数量确定各待调度任务与其他待调度任务的关联程度。
可以理解的是，根据所述分析结果可以获得各待调度任务在执行时依赖于其他待调度任务的依赖数量，所述预设关联度数据表为预先设置的用于获取关联度的数据表，所述预设关联度数据表反映依赖数量与关联度的映射关系，所述预设关联度数据表可以是通过大量实验数据训练获得，也可以是通过技术人员根据日常经验确定的数据表，当然也可以是其他方式确定的数据表，本实施例对此不加以限制。
步骤S20、根据所述关联度确定各待调度任务中具有前置依赖的多个前置任务。
可以理解的是，所述前置任务为各待调度任务中的具有前置依赖的任务，所述前置依赖为各待调度任务中待调度任务为另一个或另一些待调度任务的前置任务的依赖关系，通过所述关联度能够确定各待调度任务中多个具有前置依赖的前置任务。
进一步地，所述步骤S20包括以下步骤：
对各待调度任务进行筛选，筛除各待调度任务中关联度小于预设关联度阈值的待调度任务；
将各待调度任务中剩下的待调度任务作为具有前置依赖的多个前置任务。
应当理解的是，对于一些与其他任务有点关联，但是实际任务执行时又一般不会用到其他任务执行后所产生的数据，此时需要对这一类不太相关任务以及完全独立执行的待调度任务进行筛选，筛除各待调度任务中关联度小于预设关联度阈值的待调度任务，剩下的待调度任务即为具有前置依赖的前置任务；所述预设关联度阈值为预先设置的关联度阈值，通过将各待调度任务中关联度与所述预设关联度阈值进行比较，可以筛选出各待调度任务中关联度较小的待调度任务，将剩下的待调度任务作为具有前置依赖的多个前置任务。
步骤S30、获取各前置任务的依赖程度，找到依赖程度最高级别的前置任务。
应当理解的是，可以通过所述依赖程度，通过对所述依赖程度按照从高到低的顺序进行排序，可以获得依赖程度最高级别的前置任务，当然也可以是仅仅根据各前置任务的关联度确定各前置任务的依赖程度，获得各前置任务的依赖程度后，可以对各前置任务的依赖程度按照从高到低进行排序，获取依赖程度最高的前置任务；当然还可以通过其他方式获取依赖程度最高级别中的前置任务，本实施例对此不加以限制。
进一步地，所述步骤S30包括以下步骤：
获取各前置任务的历史执行数据，对所述历史执行数据进行分析，获得各前置任务的历史依赖频率，所述历史依赖频率为各前置任务被依赖的频率；
根据各前置任务的历史依赖频率与关联度确定各前置任务的依赖程度；
根据预设排序顺序对各依赖程度进行排序，获得依赖程度最高级别的前置任务。
应当理解的是，所述历史执行数据为各前置任务在预设周期内的执行数据，通过对所述历史执行数据进行分析，获得各前置任务的历史依赖频率，所述历史依赖频率为各前置任务被依赖的频率，通过所述历史依赖频率和关联度可以确定各前置任务的依赖程度，通过预设排序顺序可以对各依赖程度进行排序，从而确定依赖程度最高级别的前置任务。
步骤S40、获取各待调度任务中除了前置任务之外的目标任务的优先级别，根据所述依赖程度和所述优先级别构建预设任务执行顺序。
可以理解的是，所述除了前置任务之外的目标任务为各待调度任务没有前置依赖的其他任务，通过获取目标任务的优先级别，可以确定目标任务之中各任务的调度优先级，根据所述依赖程度和所述优先级别可以确定各待调度任务，即确定所述各待调度任务中各任务的执行调度优先级别。
步骤S50、根据所述预设任务执行顺序对各待调度任务进行调度，直至调度完成所有待调度任务。
应当理解的是，所述预设任务执行顺序为预先设置的用于执行待调度任务的执行顺序，所述预设任务执行顺序可以通过所述依赖程度和所述优先级别构建，当然也可以是通过其他方式确认，本实施例对此不加以限制；可以通过所述预设任务执行顺序对各待调度任务进行调度直到所有任务调度完毕。
本实施例通过上述方案，通过获取需要进行调度的多个待调度任务，并获得各待调度任务之间的关联度；根据所述关联度确定各待调度任务中具有前置依赖的多个前置任务；获取各前置任务的依赖程度，找到依赖程度最高级别的前置任务；获取各待调度任务中除了前置任务之外的目标任务的优先级别，根据所述依赖程度和所述优先级别构建预设任务执行顺序；根据所述预设任务执行顺序对各待调度任务进行调度，直至调度完成所有待调度任务，可以避免其他非重要的任务与重要任务竞争资源，使重要任务获得最优执行权，保证了重要任务的优先执行，有利于后续任务的开展，提高了工作速度和效率，提升了用户体验。
进一步地，图3为本申请智能任务调度方法第二实施例的流程示意图，如图3所示，基于第一实施例提出本申请智能任务调度方法第二实施例，在本实施例中，所述步骤S40，具体包括以下步骤：
步骤S41、根据预设插入排序算法将各待调度任务中除了前置任务之外的目标任务进行排序，获得所述目标任务的排序顺序。
需要说明的是，所述预设插入排序算法为预先设置的插入算法，一般的将一个数据插入到已经排好序的有序数据中，从而得到一个新的、个数加一的有序数据，把要排序的数组分成两部分：第一部分包含了这个数组的所有元素，但将最后一个元素除外（让数组多一个空间才有插入的位置），而第二部分就只包含这一个元素（即待插入元素）；在第一部分排序完成后，再将这个最后元素插入到已排好序的第一部分中；通过所述预设插入排序算法将各待调度任务中除了前置任务之外的目标任务进行排序；根据所述预设插入排序算法可以将各待调度任务中除了前置任务之外的目标任务进行排序，获得所述目标任务的排序顺序。
步骤S42、根据所述排序顺序对确定所述目标任务的优先级别。
可以理解的是，根据所述目标任务的排序顺序可以获得所述目标任务的执行顺序，进而确定所述目标任务的优先级别。
步骤S43、根据所述依赖程度和所述优先级别构建预设任务执行顺序，所述预设任务执行顺序为依赖程度最高级别的前置任务的执行顺序优先于最高级别的目标任务，所述最高级别的目标任务的执行顺序优先于其他级别的前置任务，所述其他级别的前置任务的执行顺序优先于其他级别的目标任务。
应该理解的是，根据所述依赖程度和所述优先级别构建预设任务执行顺序，所述预设任务执行顺序为预先设置的顺序，所述预设任务执行顺序为依赖程度最高级别的前置任务的执行顺序优先于最高级别的目标任务，所述最高级别的目标任务的执行顺序优先于其他级别的前置任务，所述其他级别的前置任务的执行顺序优先于其他级别的目标任务，即目标最高级别前置任务的优先级别高于目标最高级别的任务，所述目标最高级别的任务高于其他级别的前置任务，所述其他级别的前置任务高于其他级别的有序任务。
可以理解的是，获取各待调度任务中除了前置任务之外的一般任务的优先级别，即目标任务的原本执行的优先级别，即除前置任务之外的其他待调度任务也具有原本的执行顺序，根据各前置任务的依赖程度和模板任务的优先级别可以构建预设任务执行顺序，即可以根据所述预设任务执行顺序确定各待调度任务的调度顺序，即执行顺序，可以设置为依赖程度最高级别的前置任务的执行顺序优先于最高级别的目标任务，所述最高级别的目标任务的执行顺序优先于其他级别的前置任务，所述其他级别的前置任务的执行顺序优先于其他级别的目标任务。
本实施例通过上述方案，通过预设插入排序算法将各待调度任务中除了前置任务之外的目标任务进行排序，获得所述目标任务的排序顺序；根据所述排序顺序对确定所述目标任务的优先级别；根据所述依赖程度和所述优先级别构建预设任务执行顺序，所述预设任务执行顺序为依赖程度最高级别的前置任务的执行顺序优先于最高级别的目标任务，所述最高级别的目标任务的执行顺序优先于其他级别的前置任务，所述其他级别的前置任务的执行顺序优先于其他级别的目标任务，能够进一步准确获取执行任务的优先级，可以避免其他非重要的任务与重要任务竞争资源，使重要任务获得最优执行权，保证了重要任务的优先执行，有利于后续任务的开展，提高了工作速度和效率，提升了用户体验。
进一步地，图4为本申请智能任务调度方法第三实施例的流程示意图，如图4所示，基于第二实施例提出本申请智能任务调度方法第三实施例，在本实施例中，所述步骤S50之前，所述智能任务调度方法还包括以下步骤：
步骤S51、获取各待调度任务的任务数量，根据所述任务数量确定目标线程池。
需要说明的是，各待调度任务对应着相应的任务数量，所述目标线程池为执行待调度任务的线程池，通过所述任务数量可以确定需要的线程池，不同的任务数量确定不同的所述线程池。
进一步地，所述步骤S51具体包括以下步骤：
获取各待调度任务的任务数量；
获取预设线程池中各线程池的空闲程度和任务处理速度；
将所述空闲程度大于预设空闲程度并且所述任务处理速度大于预设处理速度的线程池作为空闲线程池；
从各空闲线程池中选取与所述任务数量匹配的线程池作为目标线程池。
可以理解的是，获取各待调度任务的任务数量后，可以获得预设线程池中各线程池的空闲程度和任务处理速度，在根据所述空闲程度与预设空闲程度的比较结果可以确定空闲线程池，从而可以从各空闲线程池中选取与所述任务数量匹配的线程池作为目标线程池。
步骤S52、根据所述预设任务执行顺序在所述目标线程池中选取对应的线程执行各待调度任务，直至调度完成所有待调度任务。
应当理解的是，获取各待调度任务的任务数量，根据所述预设任务执行顺序在所述目标线程池中对各待调度任务进行调度，直至调度完所有待调度任务，所述目标线程池的确定是根据所述预设线程池中各线程的空闲程度和任务处理能力确定，通过线程池中各线程执行任务，实时获得各线程的空闲度和任务处理速度，动态的将空闲线程和处理速度较快线程作为处理任务的新的CPU调度单元，根据预设任务执行顺序进行任务调度直至所有任务都调度完毕。
本实施例通过上述方案，通过获取各待调度任务的任务数量，根据所述任务数量确定目标线程池；根据所述预设任务执行顺序在所述目标线程池中选取对应的线程执行各待调度任务，直至调度完成所有待调度任务，能够避免其他非重要的任务与重要任务竞争资源，使重要任务获得最优执行权，保证了重要任务的优先执行，有利于后续任务的开展，提高了工作速度和效率，提升了用户体验。
基于上述智能任务调度方法的实施例，本申请进一步提供一种智能任务调度装置。
参照图5，图5为本申请智能任务调度装置第一实施例的功能模块图。
本申请智能任务调度装置第一实施例中，该智能任务调度装置包括：关联度获取10、前置任务确定模块20、依赖程度获取模块30、顺序确定模块40和调度模块50；
其中，所述关联度获取模块10，用于获取需要进行调度的多个待调度任务，并获得各待调度任务之间的关联度。
需要说明的是，所述待调度任务为预先设置的需要进行执行调度的任务或者满足了调度调节需要进行执行调度的任务，所述各待调度任务之间的关联度为各待调度任务与其他任务的关联程度，即各待调度任务在执行时是否与其他任务有关联，所述关联度用来表示各待调度任务在执行时与其他待调度任务的关系。
在具体实现中，对于需要进行调度的多个待调度任务，可以对各待调度任务进行分析，可以确定任一任务是否在执行时依赖于其他任务或被其他任务执行时所依赖，根据依赖于其他任务的数量，和/或被其他任务执行所依赖的其他任务的数量确定各待调度任务与其他待调度任务的关联程度。
进一步地，所述关联度获取模块10包括：
分析模块，用于获取需要进行调度的多个待调度任务，对各待调度任务进行分析，并获得分析结果。
依赖数量确定模块，用于根据所述分析结果确定各待调度任务在执行时依赖于其他待调度任务的依赖数量。
关联度查找模块，用于根据预设关联度数据表查找到与各依赖数量对应的关联度，所述预设关联度数据表反映依赖数量与关联度的映射关系。
可以理解的是，根据所述分析结果可以获得各待调度任务在执行时依赖于其他待调度任务的依赖数量，所述预设关联度数据表为预先设置的用于获取关联度的数据表，所述预设关联度数据表反映依赖数量与关联度的映射关系，所述预设关联度数据表可以是通过大量实验数据训练获得，也可以是通过技术人员根据日常经验确定的数据表，当然也可以是其他方式确定的数据表，本实施例对此不加以限制。
所述前置任务确定模块20，用于根据所述关联度确定各待调度任务中具有前置依赖的多个前置任务。
可以理解的是，所述前置任务为各待调度任务中的具有前置依赖的任务，所述前置依赖为各待调度任务中待调度任务为另一个或另一些待调度任务的前置任务的依赖关系，通过所述关联度能够确定各待调度任务中多个具有前置依赖的前置任务。
进一步地，所述前置任务确定模块20包括：
筛选模块，用于对各待调度任务进行筛选，筛除各待调度任务中关联度小于预设关联度阈值的待调度任务。
前置任务筛选模块，用于将各待调度任务中剩下的待调度任务作为具有前置依赖的多个前置任务。
应当理解的是，对于一些与其他任务有点关联，但是实际任务执行时又一般不会用到其他任务执行后所产生的数据，此时需要对这一类不太相关任务以及完全独立执行的待调度任务进行筛选，筛除各待调度任务中关联度小于预设关联度阈值的待调度任务，剩下的待调度任务即为具有前置依赖的前置任务；所述预设关联度阈值为预先设置的关联度阈值，通过将各待调度任务中关联度与所述预设关联度阈值进行比较，可以筛选出各待调度任务中关联度较小的待调度任务，将剩下的待调度任务作为具有前置依赖的多个前置任务。
所述依赖程度获取模块30，用于获取各前置任务的依赖程度，找到依赖程度最高级别的前置任务。
应当理解的是，可以通过所述依赖程度，通过对所述依赖程度按照从高到低的顺序进行排序，可以获得依赖程度最高级别的前置任务，当然也可以是仅仅根据各前置任务的关联度确定各前置任务的依赖程度，获得各前置任务的依赖程度后，可以对各前置任务的依赖程度按照从高到低进行排序，获取依赖程度最高的前置任务；当然还可以通过其他方式获取依赖程度最高级别中的前置任务，本实施例对此不加以限制。
进一步地，所述依赖程度获取模块30包括：
依赖频率获取模块，用于获取各前置任务的历史执行数据，对所述历史执行数据进行分析，获得各前置任务的历史依赖频率，所述历史依赖频率为各前置任务被依赖的频率；
依赖程度确定模块，用于根据各前置任务的历史依赖频率与关联度确定各前置任务的依赖程度；
依赖排序模块，用于根据预设排序顺序对各依赖程度进行排序，获得依赖程度最高级别的前置任务。
应当理解的是，所述历史执行数据为各前置任务在预设周期内的执行数据，通过对所述历史执行数据进行分析，获得各前置任务的历史依赖频率，所述历史依赖频率为各前置任务被依赖的频率，通过所述历史依赖频率和关联度可以确定各前置任务的依赖程度，通过预设排序顺序可以对各依赖程度进行排序，从而确定依赖程度最高级别的前置任务。
所述顺序确定模块40，用于获取多个待调度任务中除了前置任务之外的目标任务的优先级别，根据所述依赖程度和所述优先级别构建预设任务执行顺序。
可以理解的是，所述除了前置任务之外的目标任务为各待调度任务没有前置依赖的其他任务，通过获取目标任务的优先级别，可以确定目标任务之中各任务的调度优先级，根据所述依赖程度和所述优先级别可以确定各待调度任务，即确定所述各待调度任务中各任务的执行调度优先级别。
进一步地，所述顺序确定模块40包括：
目标任务排序模块，用于根据预设插入排序算法将各待调度任务中除了前置任务之外的目标任务进行排序，获得所述目标任务的排序顺序。
需要说明的是，所述预设插入排序算法为预先设置的插入算法，一般的将一个数据插入到已经排好序的有序数据中，从而得到一个新的、个数加一的有序数据，把要排序的数组分成两部分：第一部分包含了这个数组的所有元素，但将最后一个元素除外（让数组多一个空间才有插入的位置），而第二部分就只包含这一个元素（即待插入元素）；在第一部分排序完成后，再将这个最后元素插入到已排好序的第一部分中；通过所述预设插入排序算法将各待调度任务中除了前置任务之外的目标任务进行排序；根据所述预设插入排序算法可以将各待调度任务中除了前置任务之外的目标任务进行排序，获得所述目标任务的排序顺序。
目标任务优先级获取模块，用于根据所述排序顺序对确定所述目标任务的优先级别。
可以理解的是，根据所述目标任务的排序顺序可以获得所述目标任务的执行顺序，进而确定所述目标任务的优先级别。
任务执行顺序构建模块，用于根据所述依赖程度和所述优先级别构建预设任务执行顺序，所述预设任务执行顺序为依赖程度最高级别的前置任务的执行顺序优先于最高级别的目标任务，所述最高级别的目标任务的执行顺序优先于其他级别的前置任务，所述其他级别的前置任务的执行顺序优先于其他级别的目标任务。
应该理解的是，根据所述依赖程度和所述优先级别构建预设任务执行顺序，所述预设任务执行顺序为预先设置的顺序，所述预设任务执行顺序为依赖程度最高级别的前置任务的执行顺序优先于最高级别的目标任务，所述最高级别的目标任务的执行顺序优先于其他级别的前置任务，所述其他级别的前置任务的执行顺序优先于其他级别的目标任务，即目标最高级别前置任务的优先级别高于目标最高级别的任务，所述目标最高级别的任务高于其他级别的前置任务，所述其他级别的前置任务高于其他级别的有序任务。
可以理解的是，获取各待调度任务中除了前置任务之外的一般任务的优先级别，即目标任务的原本执行的优先级别，即除前置任务之外的其他待调度任务也具有原本的执行顺序，根据各前置任务的依赖程度和模板任务的优先级别可以构建预设任务执行顺序，即可以根据所述预设任务执行顺序确定各待调度任务的调度顺序，即执行顺序，可以设置为依赖程度最高级别的前置任务的执行顺序优先于最高级别的目标任务，所述最高级别的目标任务的执行顺序优先于其他级别的前置任务，所述其他级别的前置任务的执行顺序优先于其他级别的目标任务。
所述调度模块50，用于根据所述预设任务执行顺序对各待调度任务进行调度，直至调度完成所有待调度任务。
应当理解的是，所述预设任务执行顺序为预先设置的用于执行待调度任务的执行顺序，所述预设任务执行顺序可以通过所述依赖程度和所述优先级别构建，当然也可以是通过其他方式确认，本实施例对此不加以限制；可以通过预设任务执行顺序对各待调度任务进行调度直到所有任务调度完毕。
进一步地，所述调度模块50包括：
线程池确定模块，用于获取各待调度任务的任务数量，根据所述任务数量确定目标线程池。
需要说明的是，各待调度任务对应着相应的任务数量，所述目标线程池为执行待调度任务的线程池，通过所述任务数量可以确定需要的线程池，不同的任务数量确定不同的所述线程池。
进一步地，所述线程池确定模块包括：
任务数量获取模块，用于获取各待调度任务的任务数量。
处理速度获取模块，用于获取预设线程池中各线程池的空闲程度和任务处理速度。
空闲线程池确定模块，用于将所述空闲程度大于预设空闲程度并且所述任务处理速度大于预设处理速度的线程池作为空闲线程池。
目标线程池匹配模块，用于从各空闲线程池中选取与所述任务数量匹配的线程池作为目标线程池。
可以理解的是，获取各待调度任务的任务数量后，可以获得预设线程池中各线程池的空闲程度和任务处理速度，在根据所述空闲程度与预设空闲程度的比较结果可以确定空闲线程池，从而可以从各空闲线程池中选取与所述任务数量匹配的线程池作为目标线程池。
进一步地，所述调度模块50还包括：
调度执行模块，用于根据所述预设任务执行顺序在所述目标线程池中选取对应的线程执行各待调度任务，直至调度完成所有待调度任务。
应当理解的是，获取各待调度任务的任务数量，根据所述预设任务执行顺序在所述目标线程池中对各待调度任务进行调度，直至调度完所有待调度任务，所述目标线程池的确定是根据所述预设线程池中各线程的空闲程度和任务处理能力确定，通过线程池中各线程执行任务，实时获得各线程的空闲度和任务处理速度，动态的将空闲线程和处理速度较快线程作为处理任务的新的CPU调度单元，根据预设任务执行顺序进行任务调度直至所有任务都调度完毕。
本实施例通过上述方案，通过获取需要进行调度的多个待调度任务，并获得各待调度任务之间的关联度；根据所述关联度确定各待调度任务中具有前置依赖的多个前置任务；获取各前置任务的依赖程度，找到依赖程度最高级别的前置任务；获取各待调度任务中除了前置任务之外的目标任务的优先级别，根据所述依赖程度和所述优先级别构建预设任务执行顺序；根据所述预设任务执行顺序对各待调度任务进行调度，直至调度完成所有待调度任务，可以避免其他非重要的任务与重要任务竞争资源，使重要任务获得最优执行权，保证了重要任务的优先执行，有利于后续任务的开展，提高了工作速度和效率，提升了用户体验。
此外，本申请实施例还提出一种存储介质，本申请的存储介质可以为非易失性计算机可读存储介质，所述存储介质上存储有计算机可读指令，所述计算机可读指令被处理器执行时实现如上文所述的智能任务调度方法。

Claims (20)

  1. 一种智能任务调度方法，其特征在于，所述方法包括：
    获取需要进行调度的多个待调度任务，对各待调度任务进行分析，并获得分析结果；
    根据所述分析结果确定各待调度任务在执行时依赖于其他待调度任务的依赖数量；
    根据预设关联度数据表查找到与各依赖数量对应的关联度，所述预设关联度数据表反映依赖数量与关联度的映射关系；
    根据所述关联度确定各待调度任务中具有前置依赖的多个前置任务；
    获取各前置任务的依赖程度，找到依赖程度最高级别的前置任务；
    获取各待调度任务中除了前置任务之外的目标任务的优先级别，根据所述依赖程度和所述优先级别构建预设任务执行顺序；
    根据所述预设任务执行顺序对各待调度任务进行调度，直至调度完成所有待调度任务。
  2. 如权利要求1所述的方法，其特征在于，所述根据所述关联度确定各待调度任务中具有前置依赖的多个前置任务的步骤，包括：
    对各待调度任务进行筛选，筛除各待调度任务中关联度小于预设关联度阈值的待调度任务；
    将各待调度任务中剩下的待调度任务作为具有前置依赖的多个前置任务。
  3. 如权利要求2所述的方法，其特征在于，所述获取各前置任务的依赖程度，找到依赖程度最高级别的前置任务的步骤，包括：
    获取各前置任务的历史执行数据，对所述历史执行数据进行分析，获得各前置任务的历史依赖频率，所述历史依赖频率为各前置任务被依赖的频率；
    根据各前置任务的历史依赖频率与关联度确定各前置任务的依赖程度；
    根据预设排序顺序对各依赖程度进行排序，获得依赖程度最高级别的前置任务。
  4. 如权利要求3所述的方法，其特征在于，所述获取各待调度任务中除了前置任务之外的目标任务的优先级别，根据所述依赖程度和所述优先级别构建预设任务执行顺序的步骤，包括：
    根据预设插入排序算法将各待调度任务中除了前置任务之外的目标任务进行排序，获得所述目标任务的排序顺序；
    根据所述排序顺序对确定所述目标任务的优先级别；
    根据所述依赖程度和所述优先级别构建预设任务执行顺序，所述预设任务执行顺序为依赖程度最高级别的前置任务的执行顺序优先于最高级别的目标任务，所述最高级别的目标任务的执行顺序优先于其他级别的前置任务，所述其他级别的前置任务的执行顺序优先于其他级别的目标任务。
  5. 如权利要求1所述的方法，其特征在于，所述根据所述预设任务执行顺序对各待调度任务进行调度，直至调度完成所有待调度任务的步骤，包括：
    获取各待调度任务的任务数量，根据所述任务数量确定目标线程池；
    根据所述预设任务执行顺序在所述目标线程池中选取对应的线程执行各待调度任务，直至调度完成所有待调度任务。
  6. 如权利要求5所述的方法，其特征在于，所述获取各待调度任务的任务数量，根据所述任务数量确定目标线程池的步骤，包括：
    获取各待调度任务的任务数量；
    获取预设线程池中各线程池的空闲程度和任务处理速度；
    将所述空闲程度大于预设空闲程度并且所述任务处理速度大于预设处理速度的线程池作为空闲线程池；
    从各空闲线程池中选取与所述任务数量匹配的线程池作为目标线程池。
  7. 一种智能任务调度装置，其特征在于，所述装置包括：、前置任务确定模块、依赖程度获取模块、顺序确定模块和调度模块；
    其中，所述；
    所述前置任务确定模块，用于根据所述关联度确定各待调度任务中具有前置依赖的多个前置任务；
    所述依赖程度获取模块，用于获取各前置任务的依赖程度，找到依赖程度最高级别的前置任务；
    所述顺序确定模块，用于获取各待调度任务中除了前置任务之外的目标任务的优先级别，根据所述依赖程度和所述优先级别构建预设任务执行顺序；
    所述调度模块，用于根据所述预设任务执行顺序对各待调度任务进行调度，直至调度完成所有待调度任务。
  8. 如权利要求7所述的智能任务调度装置，其特征在于，所述前置任务确定模块包括：
    筛选模块，用于对各待调度任务进行筛选，筛除各待调度任务中关联度小于预设关联度阈值的待调度任务；
    前置任务筛选模块，用于将各待调度任务中剩下的待调度任务作为具有前置依赖的多个前置任务。
  9. 如权利要求8所述的智能任务调度装置，其特征在于，所述依赖程度获取模块包括：
    依赖频率获取模块，用于获取各前置任务的历史执行数据，对所述历史执行数据进行分析，获得各前置任务的历史依赖频率，所述历史依赖频率为各前置任务被依赖的频率；
    依赖程度确定模块，用于根据各前置任务的历史依赖频率与关联度确定各前置任务的依赖程度；
    依赖排序模块，用于根据预设排序顺序对各依赖程度进行排序，获得依赖程度最高级别的前置任务。
  10. 如权利要求9所述的智能任务调度装置，其特征在于，所述顺序确定模块包括：
    目标任务排序模块，用于根据预设插入排序算法将各待调度任务中除了前置任务之外的目标任务进行排序，获得所述目标任务的排序顺序；
    目标任务优先级获取模块，用于根据所述排序顺序对确定所述目标任务的优先级别；
    任务执行顺序构建模块，用于根据所述依赖程度和所述优先级别构建预设任务执行顺序，所述预设任务执行顺序为依赖程度最高级别的前置任务的执行顺序优先于最高级别的目标任务，所述最高级别的目标任务的执行顺序优先于其他级别的前置任务，所述其他级别的前置任务的执行顺序优先于其他级别的目标任务。
  11. 如权利要求10所述的智能任务调度装置，其特征在于，所述调度模块包括：
    线程池确定模块，用于获取各待调度任务的任务数量，根据所述任务数量确定目标线程池；
    调度执行模块，用于根据所述预设任务执行顺序在所述目标线程池中选取对应的线程执行各待调度任务，直至调度完成所有待调度任务。
  12. 如权利要求11所述的智能任务调度装置，其特征在于，所述线程池确定模块包括：
    任务数量获取模块，用于获取各待调度任务的任务数量。
    处理速度获取模块，用于获取预设线程池中各线程池的空闲程度和任务处理速度；
    空闲线程池确定模块，用于将所述空闲程度大于预设空闲程度并且所述任务处理速度大于预设处理速度的线程池作为空闲线程池；
    目标线程池匹配模块，用于从各空闲线程池中选取与所述任务数量匹配的线程池作为目标线程池。
  13. 一种智能任务调度设备，其特征在于，所述智能任务调度设备包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机可读指令，所述计算机可读指令配置为实现如下步骤：
    获取需要进行调度的多个待调度任务，对各待调度任务进行分析，并获得分析结果；
    根据所述分析结果确定各待调度任务在执行时依赖于其他待调度任务的依赖数量；
    根据预设关联度数据表查找到与各依赖数量对应的关联度，所述预设关联度数据表反映依赖数量与关联度的映射关系；
    根据所述关联度确定各待调度任务中具有前置依赖的多个前置任务；
    获取各前置任务的依赖程度，找到依赖程度最高级别的前置任务；
    获取各待调度任务中除了前置任务之外的目标任务的优先级别，根据所述依赖程度和所述优先级别构建预设任务执行顺序；
    根据所述预设任务执行顺序对各待调度任务进行调度，直至调度完成所有待调度任务。
  14. 如权利要求13所述的智能任务调度设备，其特征在于，所述根据所述关联度确定各待调度任务中具有前置依赖的多个前置任务的步骤包括：
    对各待调度任务进行筛选，筛除各待调度任务中关联度小于预设关联度阈值的待调度任务；
    将各待调度任务中剩下的待调度任务作为具有前置依赖的多个前置任务。
  15. 如权利要求14所述的智能任务调度设备，其特征在于，所述获取各前置任务的依赖程度，找到依赖程度最高级别的前置任务的步骤包括：
    获取各前置任务的历史执行数据，对所述历史执行数据进行分析，获得各前置任务的历史依赖频率，所述历史依赖频率为各前置任务被依赖的频率；
    根据各前置任务的历史依赖频率与关联度确定各前置任务的依赖程度；
    根据预设排序顺序对各依赖程度进行排序，获得依赖程度最高级别的前置任务。
  16. 如权利要求15所述的智能任务调度设备，其特征在于，所述获取各待调度任务中除了前置任务之外的目标任务的优先级别，根据所述依赖程度和所述优先级别构建预设任务执行顺序的步骤包括：
    根据预设插入排序算法将各待调度任务中除了前置任务之外的目标任务进行排序，获得所述目标任务的排序顺序；
    根据所述排序顺序对确定所述目标任务的优先级别；
    根据所述依赖程度和所述优先级别构建预设任务执行顺序，所述预设任务执行顺序为依赖程度最高级别的前置任务的执行顺序优先于最高级别的目标任务，所述最高级别的目标任务的执行顺序优先于其他级别的前置任务，所述其他级别的前置任务的执行顺序优先于其他级别的目标任务。
  17. 如权利要求16所述的智能任务调度设备，其特征在于，所述根据所述预设任务执行顺序对各待调度任务进行调度，直至调度完成所有待调度任务的步骤包括：
    获取各待调度任务的任务数量，根据所述任务数量确定目标线程池；
    根据所述预设任务执行顺序在所述目标线程池中选取对应的线程执行各待调度任务，直至调度完成所有待调度任务。
  18. 如权利要求17所述的智能任务调度设备，其特征在于，所述获取各待调度任务的任务数量，根据所述任务数量确定目标线程池的步骤包括：
    获取各待调度任务的任务数量；
    获取预设线程池中各线程池的空闲程度和任务处理速度；
    将所述空闲程度大于预设空闲程度并且所述任务处理速度大于预设处理速度的线程池作为空闲线程池；
    从各空闲线程池中选取与所述任务数量匹配的线程池作为目标线程池。
  19. 一种存储介质，其特征在于，所述存储介质上存储有计算机可读指令，所述计算机可读指令被处理器执行时实现如下步骤：
    获取需要进行调度的多个待调度任务，对各待调度任务进行分析，并获得分析结果；
    根据所述分析结果确定各待调度任务在执行时依赖于其他待调度任务的依赖数量；
    根据预设关联度数据表查找到与各依赖数量对应的关联度，所述预设关联度数据表反映依赖数量与关联度的映射关系；
    根据所述关联度确定各待调度任务中具有前置依赖的多个前置任务；
    获取各前置任务的依赖程度，找到依赖程度最高级别的前置任务；
    获取各待调度任务中除了前置任务之外的目标任务的优先级别，根据所述依赖程度和所述优先级别构建预设任务执行顺序；
    根据所述预设任务执行顺序对各待调度任务进行调度，直至调度完成所有待调度任务。
  20. 如权利要求19所述的存储介质，其特征在于，所述根据所述关联度确定各待调度任务中具有前置依赖的多个前置任务的步骤包括：
    对各待调度任务进行筛选，筛除各待调度任务中关联度小于预设关联度阈值的待调度任务；
    将各待调度任务中剩下的待调度任务作为具有前置依赖的多个前置任务。
PCT/CN2019/118370 2019-03-16 2019-11-14 智能任务调度方法、装置、设备及存储介质 Ceased WO2020186787A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910206130.3 2019-03-16
CN201910206130.3A CN110096345B (zh) 2019-03-16 2019-03-16 智能任务调度方法、装置、设备及存储介质

Publications (1)

Publication Number Publication Date
WO2020186787A1 true WO2020186787A1 (zh) 2020-09-24

Family

ID=67443343

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/118370 Ceased WO2020186787A1 (zh) 2019-03-16 2019-11-14 智能任务调度方法、装置、设备及存储介质

Country Status (2)

Country Link
CN (1) CN110096345B (zh)
WO (1) WO2020186787A1 (zh)

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112306471A (zh) * 2020-10-19 2021-02-02 北京金山云网络技术有限公司 一种任务的调度方法和装置
CN112363819A (zh) * 2020-12-02 2021-02-12 深圳市房多多网络科技有限公司 大数据任务动态编排调度方法、装置及计算设备
CN113449991A (zh) * 2021-06-29 2021-09-28 中国工商银行股份有限公司 动态批量作业调度方法及装置
CN114020431A (zh) * 2021-11-04 2022-02-08 杭州东方通信软件技术有限公司 一种任务调度方法及系统
CN114091777A (zh) * 2021-11-29 2022-02-25 上海有个机器人有限公司 智能柜闲时调仓方法、系统、电子设备、介质及程序产品
CN114661431A (zh) * 2020-12-23 2022-06-24 深圳Tcl新技术有限公司 一种任务调度方法、存储介质及终端设备
CN114896095A (zh) * 2022-05-10 2022-08-12 中国平安财产保险股份有限公司 异常影响分析方法、装置、计算机设备及存储介质
CN115114002A (zh) * 2022-06-29 2022-09-27 苏州浪潮智能科技有限公司 一种进程调度方法、装置、设备及计算机可读存储介质
CN118467143A (zh) * 2024-07-12 2024-08-09 四川融科智联科技有限公司 一种银行自动化切换任务调度系统
CN118708362A (zh) * 2024-08-29 2024-09-27 净化控股集团股份有限公司 一种数据的处理方法、装置及设备
CN119065858A (zh) * 2024-11-05 2024-12-03 安徽习承科技有限公司 一种基于模块化设计的计量与采集设备检测方法
CN119225932A (zh) * 2024-09-23 2024-12-31 厦门熵基科技有限公司 多任务编排处理方法、装置、存储介质及计算机设备
CN119441054A (zh) * 2024-11-06 2025-02-14 康键信息技术（深圳）有限公司 针对金融业务应用程序的测试方法、电子设备、存储介质
CN119806786A (zh) * 2025-03-12 2025-04-11 北京大数据先进技术研究院 一种基于数据任务网络的任务节点执行控制方法
CN120234124A (zh) * 2025-05-30 2025-07-01 深圳市捷易科技有限公司 算力任务执行方法、算力调度设备以及存储介质
CN120335969A (zh) * 2025-06-18 2025-07-18 宁波银行股份有限公司 数据处理方法、装置、电子设备及存储介质
CN121094489A (zh) * 2025-11-10 2025-12-09 邯郸市友发钢管有限公司 一种钢管柔性输送调度方法及系统

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110096345B (zh) * 2019-03-16 2024-04-12 平安科技（深圳）有限公司 智能任务调度方法、装置、设备及存储介质
CN110554909A (zh) * 2019-09-06 2019-12-10 腾讯科技（深圳）有限公司 任务的调度处理方法、装置及计算机设备
CN110852593A (zh) * 2019-11-06 2020-02-28 重庆大学 任务处理方法、设备、存储介质及装置
CN113282383B (zh) * 2020-02-19 2024-03-12 中科寒武纪科技股份有限公司 任务调度方法、任务处理方法及相关产品
CN113534750B (zh) * 2020-04-15 2022-12-02 北京旷视机器人技术有限公司 密集存储下的作业调度方法、装置、系统、设备及介质
CN111858013A (zh) * 2020-06-19 2020-10-30 中科院计算所西部高等技术研究院 工作流作业调度控制方法
CN112306661B (zh) * 2020-11-06 2023-09-15 平安科技（深圳）有限公司 任务调度方法、装置、电子设备及可读存储介质
CN112395088B (zh) * 2020-11-17 2024-08-27 中国平安人寿保险股份有限公司 任务调度方法、装置、设备及存储介质
CN112817721B (zh) * 2021-01-29 2022-08-09 中国平安财产保险股份有限公司 基于人工智能的任务调度方法、装置、计算机设备和介质
CN113326114B (zh) * 2021-06-11 2024-10-11 深圳前海微众银行股份有限公司 一种批量任务处理方法及装置
CN113485810A (zh) * 2021-07-23 2021-10-08 平安科技（深圳）有限公司 一种任务的调度执行方法、装置、设备及存储介质
CN114185665A (zh) * 2021-12-16 2022-03-15 北京少广科技集团有限公司 一种云计算操作系统高效任务调度技术方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140136255A1 (en) * 2012-11-14 2014-05-15 Wal-Mart Stores, Inc. Dynamic Task Management
CN104536811A (zh) * 2014-12-26 2015-04-22 广州华多网络科技有限公司 基于hive任务的任务调度方法及装置
CN107688488A (zh) * 2016-08-03 2018-02-13 中国移动通信集团湖北有限公司 一种基于元数据的任务调度的优化方法及装置
CN109086986A (zh) * 2018-07-20 2018-12-25 中国邮政储蓄银行股份有限公司 作业调度方法及装置
CN110096345A (zh) * 2019-03-16 2019-08-06 平安科技（深圳）有限公司 智能任务调度方法、装置、设备及存储介质

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102981904B (zh) * 2011-09-02 2016-08-03 阿里巴巴集团控股有限公司 一种任务调度方法及系统
CN103885839B (zh) * 2014-04-06 2017-02-22 孙凌宇 基于多水平划分法和赋权有向超图的云计算任务调度方法
US9529643B2 (en) * 2015-01-26 2016-12-27 Qualcomm Incorporated Method and system for accelerating task control flow
CN104915260B (zh) * 2015-06-19 2018-05-25 北京搜狐新媒体信息技术有限公司 一种Hadoop集群管理任务的分发方法和系统
CN108268319A (zh) * 2016-12-31 2018-07-10 中国移动通信集团河北有限公司 任务调度方法、装置及系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140136255A1 (en) * 2012-11-14 2014-05-15 Wal-Mart Stores, Inc. Dynamic Task Management
CN104536811A (zh) * 2014-12-26 2015-04-22 广州华多网络科技有限公司 基于hive任务的任务调度方法及装置
CN107688488A (zh) * 2016-08-03 2018-02-13 中国移动通信集团湖北有限公司 一种基于元数据的任务调度的优化方法及装置
CN109086986A (zh) * 2018-07-20 2018-12-25 中国邮政储蓄银行股份有限公司 作业调度方法及装置
CN110096345A (zh) * 2019-03-16 2019-08-06 平安科技（深圳）有限公司 智能任务调度方法、装置、设备及存储介质

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112306471A (zh) * 2020-10-19 2021-02-02 北京金山云网络技术有限公司 一种任务的调度方法和装置
CN112306471B (zh) * 2020-10-19 2024-09-13 北京金山云网络技术有限公司 一种任务的调度方法和装置
CN112363819A (zh) * 2020-12-02 2021-02-12 深圳市房多多网络科技有限公司 大数据任务动态编排调度方法、装置及计算设备
CN114661431A (zh) * 2020-12-23 2022-06-24 深圳Tcl新技术有限公司 一种任务调度方法、存储介质及终端设备
CN113449991A (zh) * 2021-06-29 2021-09-28 中国工商银行股份有限公司 动态批量作业调度方法及装置
CN114020431A (zh) * 2021-11-04 2022-02-08 杭州东方通信软件技术有限公司 一种任务调度方法及系统
CN114091777B (zh) * 2021-11-29 2025-04-04 上海有个机器人有限公司 智能柜闲时调仓方法、系统、电子设备、介质及程序产品
CN114091777A (zh) * 2021-11-29 2022-02-25 上海有个机器人有限公司 智能柜闲时调仓方法、系统、电子设备、介质及程序产品
CN114896095A (zh) * 2022-05-10 2022-08-12 中国平安财产保险股份有限公司 异常影响分析方法、装置、计算机设备及存储介质
CN115114002A (zh) * 2022-06-29 2022-09-27 苏州浪潮智能科技有限公司 一种进程调度方法、装置、设备及计算机可读存储介质
CN118467143A (zh) * 2024-07-12 2024-08-09 四川融科智联科技有限公司 一种银行自动化切换任务调度系统
CN118708362A (zh) * 2024-08-29 2024-09-27 净化控股集团股份有限公司 一种数据的处理方法、装置及设备
CN119225932A (zh) * 2024-09-23 2024-12-31 厦门熵基科技有限公司 多任务编排处理方法、装置、存储介质及计算机设备
CN119065858A (zh) * 2024-11-05 2024-12-03 安徽习承科技有限公司 一种基于模块化设计的计量与采集设备检测方法
CN119441054A (zh) * 2024-11-06 2025-02-14 康键信息技术（深圳）有限公司 针对金融业务应用程序的测试方法、电子设备、存储介质
CN119806786A (zh) * 2025-03-12 2025-04-11 北京大数据先进技术研究院 一种基于数据任务网络的任务节点执行控制方法
CN120234124A (zh) * 2025-05-30 2025-07-01 深圳市捷易科技有限公司 算力任务执行方法、算力调度设备以及存储介质
CN120335969A (zh) * 2025-06-18 2025-07-18 宁波银行股份有限公司 数据处理方法、装置、电子设备及存储介质
CN121094489A (zh) * 2025-11-10 2025-12-09 邯郸市友发钢管有限公司 一种钢管柔性输送调度方法及系统

Also Published As

Publication number Publication date
CN110096345B (zh) 2024-04-12
CN110096345A (zh) 2019-08-06

Similar Documents

Publication Publication Date Title
WO2020186787A1 (zh) 智能任务调度方法、装置、设备及存储介质
WO2020253116A1 (zh) 数据跑批方法、装置、存储介质及集群中的成员主机
WO2020253135A1 (zh) 自动化分析方法、用户设备、存储介质及装置
WO2020015067A1 (zh) 数据采集方法、装置、设备及存储介质
WO2018036167A1 (zh) 测试任务执行机分配方法、装置、服务器和存储介质
WO2018068533A1 (zh) 一种人脸检测的方法及装置
WO2020215681A1 (zh) 指示信息生成方法、装置、终端及存储介质
WO2020233089A1 (zh) 测试用例生成方法、装置、终端及计算机可读存储介质
WO2018076800A1 (zh) 一种数据异步更新方法及其系统
WO2020119115A1 (zh) 数据审核方法、装置、设备及存储介质
WO2014044136A1 (zh) 基于分布式数据的并发处理方法、系统和计算机存储介质
WO2019132299A1 (ko) 클라우드 시스템에서의 우선 순위 기반 자원 스케일링 시스템, 장치 및 방법
WO2020073494A1 (zh) 网页后门检测方法、设备、存储介质及装置
WO2020015060A1 (zh) 用电量异常评估方法、装置、设备和计算机存储介质
WO2020119069A1 (zh) 基于自编码神经网络的文本生成方法、装置、终端及介质
WO2021012508A1 (zh) Ai影像识别方法、装置、设备及存储介质
WO2020052247A1 (zh) 网络资源批量加载方法、智能电视、存储介质及装置
EP3942481A1 (en) Method of performing, by electronic device, convolution operation at certain layer in neural network, and electronic device therefor
WO2020107591A1 (zh) 重复投保限制方法、装置、设备及可读存储介质
WO2018018820A1 (zh) 处理器工作频率和/或工作核数的调整方法、装置和终端
WO2017206885A1 (zh) 应用程序的关闭方法、装置、存储介质及电子设备
WO2019168265A1 (ko) 전자 장치, 전자 장치의 태스크 처리 방법 및 컴퓨터 판독 가능 매체
WO2018236141A1 (ko) 크래시 리포트 그룹핑 방법, 서버 및 컴퓨터 프로그램
WO2019221377A1 (ko) 블록체인 기반 트랜잭션 관리 시스템 및 그 동작 방법
WO2021107326A1 (ko) 실전 작업을 기반으로 한 검증용 크라우드소싱 작업을 제공하는 방법 및 장치

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19919880

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19919880

Country of ref document: EP

Kind code of ref document: A1