Task processing method, edge computing device, computer equipment, medium

Abstract

The present disclosure proposes a task processing method, edge computing device, computer device and computer-readable medium, the method comprising: receiving a task sent by a network device; determining the allocation type of the task, and dividing the task into task units according to the allocation type, and placing the task units into a task pool; determining the dispatch priority of each task unit in the task pool according to the type of each task unit in the task pool, the type of the task unit being the allocation type of the task to which the task unit belongs; determining the task unit to be processed according to the dispatch priority; if the current processing capacity of the edge computing device does not meet the preset first requirement, determining the first network element device, and sending the task unit to be processed to the first network element device. When the computing power resources of the edge computing device are insufficient, the first network element device is selected from the surrounding network element group to perform task transfer calculation, which improves the computing power of the edge computing device and reduces the task calculation delay.

Description

Technical Field

[0001] The present disclosure relates to the field of edge computing technology, and in particular to a task processing method, an edge computing device, a computer device, and a computer-readable medium. Background Art

[0002] MEC (Multi-Access Edge Computing / Mobile Edge Computing), a carrier-specific edge computing technology deployed at the edge of the network, has attracted considerable attention and is widely considered one of the two key enabling technologies for 5G, particularly for its computing capabilities. With the continued expansion of various operations and maintenance services, and considering the potential future development and changes in operations and maintenance networks, the requirements for real-time and intelligent performance are becoming increasingly stringent. There is a desire to increase the computing power of MEC and enable intelligent MEC operations and maintenance.

[0003] However, in the case of short-range edge cloud, although the short-range edge cloud computing capabilities of MEC are used, the computing power of MEC is limited, especially when facing some urgent tasks that need to be solved in real time, resulting in the inability to process tasks in a timely manner or facing high latency. Summary of the Invention

[0004] In response to the above-mentioned deficiencies in the prior art, the present disclosure provides a task processing method, a network management system, an independent computing node, a computer device, and a computer-readable medium.

[0005] In a first aspect, an embodiment of the present disclosure provides a task processing method, including:

[0006] Receive tasks sent by network devices;

[0007] Determining an allocation type of the task, dividing the task into task units according to the allocation type, and placing the task units into a task pool;

[0008] Determining the dispatch priority of each task unit in the task pool according to the type of each task unit in the task pool, where the type of the task unit is the allocation type of the task to which the task unit belongs;

[0009] Determine the task unit to be processed according to the priority of the dispatch;

[0010] If the current processing capability of the edge computing device does not meet the preset first requirement, a first network element device is determined, and the task unit to be processed is sent to the first network element device so that the first network element device processes the task unit to be processed.

[0011] In some embodiments, the assignment types include tasks, subtasks, iterative tasks, and iterative subtasks.

[0012] In some embodiments, the task includes a task type, and before determining the assignment type of the task, the method further includes: configuring an algorithm for the task according to the task type of the task;

[0013] Determining the assignment type of the task includes:

[0014] If it is determined according to the task type that the task cannot be divided into subtasks, and it is determined according to the algorithm that the task cannot be divided into iterative tasks, then determining the assignment type of the task to be task;

[0015] If it is determined according to the task type that the task cannot be divided into subtasks, and it is determined according to the algorithm that the task can be divided into an iterative task, then determining the assignment type of the task to be an iterative task;

[0016] If it is determined according to the task type that the task can be divided into subtasks, and it is determined according to the algorithm that the subtask cannot be divided into iterative subtasks, then determining the assignment type of the task to be a subtask;

[0017] If it is determined according to the task type that the task can be divided into subtasks, and if it is determined according to the algorithm that the subtask can be divided into iterative subtasks, then the allocation type of the task is determined to be an iterative subtask.

[0018] In some embodiments, determining the dispatch priority of each task unit in the task pool according to the type of each task unit in the task pool includes:

[0019] If the type of each task unit in the task pool includes a task, determining the dispatch priority of each task unit in the task pool according to the configured weight, scheduling frequency and task execution deadline of the task;

[0020] If the types of each task unit in the task pool include tasks and subtasks, then determining the tasks to which each task unit in the task pool belongs, and determining the dispatching priority of each task unit in the task pool according to the configured weight, scheduling frequency, task execution deadline and subtask sequence number of the task;

[0021] If the types of each task unit in the task pool include tasks and iterative tasks, then determining the tasks to which each task unit in the task pool belongs, and determining the dispatching priority of each task unit in the task pool according to the configured weight, scheduling frequency, task execution deadline and iteration order of the task;

[0022] If the types of each task unit in the task pool include tasks, subtasks and iterative tasks, or the types of each task unit in the task pool include tasks, subtasks and iterative subtasks, or the types of each task unit in the task pool include tasks, subtasks, iterative tasks and iterative subtasks, then determine the tasks to which each task unit in the task pool belongs, and determine the issuance priority of each task unit in the task pool based on the configured weight, scheduling frequency, task execution deadline, iteration order and subtask sequence number of the task.

[0023] In some embodiments, determining the dispatch priority of each task unit in the task pool according to the type of each task unit in the task pool includes:

[0024] If the task units in the task pool include multiple types, determining the issuing priority of each type of task unit respectively, and determining the issuing priority of each task unit in the task pool according to the issuing priority of each type of task unit;

[0025] Among them, the sending priority of the task unit of type task is determined according to the configured weight, scheduling frequency and task execution deadline of the task to which the task unit belongs; the sending priority of the task unit of type subtask is determined according to the subtask sequence number; the sending priority of the task unit of type iterative task is determined according to the iteration order; the sending priority of the task unit of type iterative subtask is determined according to the iteration order and subtask sequence number.

[0026] In some embodiments, the method further comprises:

[0027] If the type of the task unit in the task pool is an iterative task or an iterative subtask, and it is determined according to the algorithm that the number of iterations is not fixed, then when the processing result of the previous iteration returned by the first network element device is received, if it is determined according to the processing result that the iteration has not ended, the next iterative task or the next iterative subtask is generated, and the next iterative task or the next iterative subtask is placed in the task pool.

[0028] In some embodiments, the method further comprises:

[0029] If the allocation type is a subtask or an iterative subtask, receiving the processing results returned by each first network element device;

[0030] If it is determined according to the processing result that the corresponding task has been completed and the current processing capability of the edge computing device meets the preset second requirement, iteratively optimizing the processing result to generate a model;

[0031] If it is determined based on the processing result that the corresponding task has been completed and the current processing capability of the edge computing device does not meet the preset second requirement, a convergence task is generated, a second network element device is determined, and the convergence task is sent to the second network element device.

[0032] In some embodiments, determining the first network element device includes:

[0033] Acquire network element data of each network element device in the network element data pool, wherein the network element data includes distance parameters and performance parameters;

[0034] Determining a network element device whose distance parameter and performance parameter meet preset conditions;

[0035] The first network element device is determined according to the performance parameter.

[0036] In some embodiments, the performance parameters include: a CPU free space ratio, a MEM free space ratio, a DISK free space ratio, and a cache free space ratio;

[0037] The distance parameter and the performance parameter satisfying the preset conditions include: the distance parameter is less than or equal to a first threshold, the CPU remaining space ratio is greater than a second threshold, the MEM remaining space ratio is greater than a third threshold, the DISK remaining space ratio is greater than a fourth threshold, and the CACHE remaining space ratio is greater than a fifth threshold;

[0038] The determining the first network element device according to the performance parameter includes:

[0039] Sort the network element devices according to the CPU remaining space ratio, MEM remaining space ratio, DISK remaining space ratio and CACHE remaining space ratio;

[0040] The first network element device is determined according to the ranking.

[0041] In some embodiments, after sending the task unit to be processed to the first network element device, the method further includes:

[0042] If a task interruption message returned by the first network element device is received and the current processing capability of the edge computing device does not meet the preset third requirement, a third network element device is determined so that the third network element device processes the task unit corresponding to the task interruption message.

[0043] In some embodiments, after sending the task unit to be processed to the first network element device, the method further includes:

[0044] If the processing result returned by the first network element device is not received within the preset period, other first network element devices are determined, and the task units to be processed are sent to the other first network element devices.

[0045] In a second aspect, an embodiment of the present disclosure provides an edge computing device, including:

[0046] A receiving module, used for receiving tasks sent by network devices;

[0047] A first determining module, configured to determine the assignment type of the task;

[0048] a partitioning module, configured to partition the task into task units according to the allocation type, and place the task units into a task pool;

[0049] A second determining module is configured to determine the issuing priority of each task unit in the task pool according to the type of each task unit in the task pool, where the type of the task unit is the allocation type of the task to which the task unit belongs;

[0050] A third determining module is used to determine the task unit to be processed according to the sending priority;

[0051] A fourth determining module is configured to determine a first network element device if the current processing capability of the edge computing device does not meet the preset first requirement;

[0052] The sending module is used to send the task unit to be processed to the first network element device, so that the first network element device processes the task unit to be processed.

[0053] In a third aspect, an embodiment of the present disclosure provides a computer device, including:

[0054] one or more processors;

[0055] a storage device having one or more programs stored thereon;

[0056] When the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement the task processing method as described above.

[0057] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable medium having a computer program stored thereon, wherein the program implements the task processing method as described above when executed.

[0058] In the embodiment of the present disclosure, the edge computing device determines the allocation type of the task, and divides the task into task units according to the allocation type, and puts the task units into the task pool. According to the type of each task unit in the task pool, the sending priority of each task unit in the task pool is determined, and the task unit to be processed is determined according to the sending priority. If the current processing capacity of the edge computing device does not meet the preset first requirement, the first network element device is determined, and the task unit to be processed is sent to the first network element device so that the first network element device processes the task unit to be processed. When the computing power resources of the edge computing device are insufficient, the embodiment of the present disclosure, based on the idea of ​​federated learning, splits the current task into multiple task units, selects the first network element device from the surrounding network element group to process the task unit, realizes the transfer calculation of the task, shares the computing power of the first network element device, improves the computing power of the edge computing device, and reduces the task calculation delay. The embodiment of the present disclosure is particularly suitable for urgent real-time tasks, and can quickly transfer urgent real-time tasks to other network element devices, reducing the task calculation delay. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is a flowchart of a task processing method provided by an embodiment of the present disclosure;

[0060] Figure 2 is another flow chart of a task processing method provided by an embodiment of the present disclosure;

[0061] Figure 3 is another flow chart of a task processing method provided by an embodiment of the present disclosure;

[0062] Figure 4 is another flow chart of a task processing method provided by an embodiment of the present disclosure;

[0063] Figure 5 is another flow chart of a task processing method provided by an embodiment of the present disclosure;

[0064] Figure 6 It is a structural diagram of an edge computing device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0065] To enable those skilled in the art to better understand the technical solution of the present disclosure, the message transmission method, access gateway, and computer-readable medium provided by the present disclosure are described in detail below with reference to the accompanying drawings.

[0066] Example embodiments will be described more fully hereinafter with reference to the accompanying drawings, but the example embodiments may be embodied in different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete and will fully convey the scope of this disclosure to those skilled in the art.

[0067] In the absence of conflict, the various embodiments of the present disclosure and the various features therein may be combined with each other.

[0068] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0069] The terms used herein are used only to describe specific embodiments and are not intended to limit the present disclosure. As used herein, the singular forms "a," "an," and "the" are also intended to include the plural forms, unless the context clearly indicates otherwise. It will also be understood that when the terms "comprising" and / or "made of" are used in this specification, the presence of the features, wholes, steps, operations, elements, and / or components is specified, but the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or groups thereof is not excluded.

[0070] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and the present disclosure, and will not be interpreted as having an idealized or overly formal meaning unless expressly defined as such herein.

[0071] MEC can provide various resource or service scheduling optimization algorithms. However, MEC is not an intelligent node and is limited by hardware resources, making it difficult to support complex model training and optimization. MEC is connected to multiple network elements. To increase MEC's ​​computing power, process tasks in a timely manner, or reduce high latency, this disclosure proposes a task processing method, edge computing device, computer device, and computer-readable medium based on federated learning.

[0072] Figure 1 The flowchart of the task processing method provided by the embodiment of the present disclosure is shown. The task processing method can be applied to edge computing devices, namely MEC. Figure 1 As shown, the task processing method provided by this embodiment includes the following steps.

[0073] Step 101: Receive a task sent by a network device.

[0074] In this step, the MEC receives a task sent by the network device. The task can be a periodic task or an urgent real-time task. The task sent by the network device to the MEC includes the task type and the task content corresponding to the task type. The task content may include the operation object, operation time, task intensity, the complete or partial task goal, the computing network structure used for task calculation, various hyperparameters used for task calculation, and other parameters. Task types may include: KPI (Key Performance Indicator, Key Performance Indicator) anomaly detection model training, RF (Radio Frequency, Radio Frequency) fingerprint library model training, etc.

[0075] Step 102: Determine the assignment type of the task.

[0076] In this step, the MEC determines the allocation type for each task. Allocation types include tasks, subtasks, iterative tasks, and iterative subtasks. Each task received by the MEC can be allocated by task, subtask, iterative task, or iterative subtask. The MEC must determine the allocation type that matches the received task.

[0077] Step 103: Divide the task into task units according to the allocation type, and put the task units into a task pool.

[0078] In this step, MEC divides the task into task units according to the allocation type, and the type of the task unit is the allocation type of the task to which the task unit belongs. That is, when the allocation type is task, the type of the task unit is task. When the allocation type is subtask, the task is divided into subtasks, and the type of the task unit is subtask. The number of subtasks is pre-configured by MEC and divided according to the configured fixed number. When the allocation type is iterative task, the task is divided into iterative tasks, and the type of the task unit is iterative task. The number of iterations can be fixed or not fixed. When the number of iterations is fixed, the iterative task is configured according to the fixed number; when the number of iterative tasks is not fixed, only one iterative task is divided. When the allocation type is iterative subtask, the task is divided into iterative subtasks, and the type of the task unit is iterative subtask. The number of iterations can be fixed or not fixed. When the number of iterations is fixed, the iterative subtask is configured according to the fixed number; when the number of iterative tasks is not fixed, only one iterative subtask is divided. For each task, the divided task units are placed in the task pool.

[0079] Step 104 : Determine the dispatch priority of each task unit in the task pool according to the type of each task unit in the task pool.

[0080] In this step, each task unit in the task pool must determine its corresponding dispatch priority. All task units in the task pool are sorted, with the task units with the highest dispatch priority having the highest dispatch priority and the task units with the lowest dispatch priority having the lowest dispatch priority. When determining the dispatch priority of each task unit in the task pool, the type of each task unit must be considered.

[0081] It should be noted that the task units in the task pool are updated in real time, and the dispatch priority is also updated in real time. When a task unit is added to the task pool, all task units in the task pool must be re-sorted and the dispatch priority of each task unit in the task pool must be re-determined.

[0082] Step 105: Determine the task unit to be processed according to the priority level.

[0083] In this step, MEC finds the task unit with the highest priority in the task pool and uses it as the task unit to be processed.

[0084] Step 106: If the current processing capability of the edge computing device does not meet the preset first requirement, determine the first network element device.

[0085] In this step, if the current processing capacity of MEC does not meet the preset first requirement, that is, the task consumption resources of the task unit to be processed exceed the current processing capacity of MEC, and the idle computing resources of MEC are insufficient, then the task unit to be processed needs to perform task transfer and determine the corresponding first network element device for task transfer.

[0086] If the current processing capability of the MEC meets the preset first requirement, that is, the MEC can process the task to be processed, then the MEC processes the task unit to be processed. The first network element device here can be a pre-specified one or an arbitrary network element device.

[0087] Step 107: Send the task unit to be processed to the first network element device, so that the first network element device processes the task unit to be processed.

[0088] In this step, the MEC sends the task unit to be processed to the first network element device. The task unit sent includes parameters such as the task type, operation object, operation time, task intensity, and various hyperparameters used for task calculation corresponding to the task unit. The first network element device can find the appropriate algorithm and iterative function based on the parameters carried in the task unit, and use the data in the local detection sample database to process the task to be processed. After the first network element device processes the task unit to be processed, it will return the processing result to the MEC. If the MEC receives the processing result, it determines that the task unit to be processed is completed. If the MEC does not receive the processing result, it determines that the task unit to be processed is not completed.

[0089] For example, when the type of the task unit to be processed is a subtask, the first network element device can execute the subtask based on the Gaussian detection algorithm. When the type of the task unit to be processed is an iterative task and iterative subtask, the first network element device can execute the iterative task and iterative subtask based on the Holt-Winters algorithm.

[0090] It should be noted that the MEC is the central network element, and the first network element device is the computing network element. By default, the first network element device can apply all algorithms, that is, by default, the first network element device pre-stores all algorithms that will be used. The MEC can encrypt the task unit to be processed and send it, and the first network element device decrypts it using a pre-agreed key.

[0091] When edge computing devices are short on computing power, this disclosed embodiment, based on federated learning, splits the current task into multiple task units. A first network element device is selected from a cluster of surrounding network elements to process the task units, enabling task transfer and computing. This allows the first network element device to share computing power, thereby increasing the computing power of the edge computing device and reducing task computation latency. This disclosed embodiment is particularly suitable for urgent real-time tasks, enabling rapid transfer of urgent real-time tasks to other network element devices, reducing task computation latency.

[0092] Furthermore, the task includes a task type, and before determining the assignment type of the task, the step further includes: configuring an algorithm for the task according to the task type.

[0093] In this step, the MEC configures the algorithm based on the task type for each task. In scenarios where iteration is required, the iteration function must also be configured based on the algorithm. It should be noted that the configured algorithm and iteration function do not need to be sent to the network element. The network element can find the algorithm and iteration function that match the received task unit in the locally stored algorithm.

[0094] Figure 2 Another flow chart of the task processing method provided by the embodiment of the present disclosure is shown. Figure 2 As shown, the determining of the assignment type of the task (i.e., step 102) includes:

[0095] Step 201: If it is determined according to the task type that the task cannot be divided into subtasks, and it is determined according to the algorithm that the task cannot be divided into iterative tasks, then the allocation type of the task is determined to be task.

[0096] Step 202: If it is determined according to the task type that the task cannot be divided into subtasks, and it is determined according to the algorithm that the task can be divided into an iterative task, the assignment type of the task is determined to be an iterative task.

[0097] Step 203: If it is determined according to the task type that the task can be divided into subtasks, and it is determined according to the algorithm that the subtask cannot be divided into iterative subtasks, then the assignment type of the task is determined to be a subtask.

[0098] Step 204: If it is determined according to the task type that the task can be divided into subtasks, and it is determined according to the algorithm that the subtask can be divided into iterative subtasks, then the assignment type of the task is determined to be an iterative subtask.

[0099] In steps 201-204, the MEC determines whether the task can be split into subtasks based on the task type. Specifically, it determines whether the task can be split into subtasks based on the task type and parameters such as the operation object and operation time corresponding to the task type. The MEC determines whether the task can be split into iterative tasks based on a preconfigured algorithm that matches the task, and also determines whether the subtask can be split into iterative subtasks based on the algorithm that matches the task. This results in four types of task allocation: tasks, subtasks, iterative tasks, and iterative subtasks.

[0100] For example, if the task type is KPI anomaly detection model training, the task can be split into subtasks based on the task type, and the subtasks can also be split into iterative subtasks. KPI anomaly detection model training requires training the anomaly detection model for 1000 cells. This can be split into subtasks of 100 cells, and each subtask can be unpacked into iterative subtasks for iterative gradients and hyperparameters. In this case, the iterative processing results of each first network element device are gradients, hyperparameters, and model output data.

[0101] Further, Figure 3 Another flow chart of the task processing method provided by the embodiment of the present disclosure is shown. Figure 3 As shown, the step of determining the dispatch priority of each task unit in the task pool according to the type of each task unit in the task pool (i.e., step 104) includes:

[0102] Step 301: If the type of each task unit in the task pool includes tasks, the issuing priority of each task unit in the task pool is determined according to the configured weight, scheduling frequency and task execution deadline of the task.

[0103] The MEC pre-configures a weight, scheduling frequency, and execution deadline for each task, based on the task type. The scheduling frequency is the number of times a task is scheduled per cycle. The execution deadline is the validity period of the task, within which each task must be completed. Generally, the task execution deadline is shorter than the scheduling period. Tasks with higher scheduling frequencies, higher weights, and shorter execution deadlines are prioritized.

[0104] When the allocation type is iterative task and iterative subtask, determine in advance whether the number of iterations for the iterative task and iterative subtask is fixed. If the number of iterations is fixed, configure the fixed value. When each task is pre-divided into task units, the iterative task and iterative subtask are divided according to the fixed number of iterations. If the number of iterations is not fixed, when each task is pre-divided into task units, each task is divided into only one iterative task or iterative subtask. When the allocation type is iterative subtask and subtask, configure the number of subtasks in advance. When each task is pre-divided into task units, the iterative subtasks and subtasks are divided according to the fixed number of subtasks.

[0105] In this step, if the types of each task unit in the task pool only include tasks, first sort each task in descending order of weight. Then sort tasks with the same weight in descending order of scheduling frequency. Then sort tasks with the same weight and scheduling frequency in ascending order of task execution deadline. Tasks with higher rankings have higher dispatch priority, while tasks with lower rankings have lower dispatch priority. When sorting tasks, weight is given priority, followed by scheduling frequency, and finally task execution deadline. This sorting method is defined as the task sorting method.

[0106] For example, there are four tasks in the task pool. Task 1 has a weight, scheduling frequency, and deadline of 5, 2, and 3, respectively; Task 2 has a weight, scheduling frequency, and deadline of 5, 6, and 3, respectively; Task 3 has a weight, scheduling frequency, and deadline of 5, 2, and 4, respectively; and Task 4 has a weight, scheduling frequency, and deadline of 10, 2, and 4, respectively. The final ranking result is Task 4, Task 2, Task 1, and Task 3.

[0107] Step 302: If the types of each task unit in the task pool include tasks and subtasks, determine the tasks to which each task unit in the task pool belongs, and determine the dispatch priority of each task unit in the task pool based on the configured weight, scheduling frequency, task execution deadline and subtask sequence number of the task.

[0108] In this step, if the types of each task unit in the task pool include tasks and subtasks, the tasks are first sorted according to the task sorting method, and then the subtasks under each task branch are sorted in ascending order of subtask numbers. The task units with higher sorting have higher dispatch priority, while the task units with lower sorting have lower dispatch priority.

[0109] Step 303: If the types of each task unit in the task pool include tasks and iterative tasks, determine the tasks to which each task unit in the task pool belongs, and determine the dispatch priority of each task unit in the task pool according to the configured weight, scheduling frequency, task execution deadline and iteration order of the task.

[0110] In this step, if the types of task units in the task pool include tasks and iterative tasks, the tasks are first sorted according to the task sorting method. Then, the iterative tasks under each task branch are sorted in descending order of iteration. Task units with higher rankings have higher dispatch priority, while task units with lower rankings have lower dispatch priority.

[0111] Step 304: If the types of each task unit in the task pool include tasks, subtasks and iterative tasks, or the types of each task unit in the task pool include tasks, subtasks and iterative subtasks, or the types of each task unit in the task pool include tasks, subtasks, iterative tasks and iterative subtasks, then determine the tasks to which each task unit in the task pool belongs, and determine the dispatch priority of each task unit in the task pool based on the configured weight, scheduling frequency, task execution deadline, iteration order and subtask sequence number of the task.

[0112] In this step, if the types of task units in the task pool include tasks, subtasks, and iterative tasks, the tasks are first sorted according to the task sorting method. Then, the iterative tasks under each task branch are sorted in descending order of iteration order. Finally, the subtasks under each task branch are sorted in ascending order of subtask numbers. Task units with higher rankings have higher dispatch priority, while task units with lower rankings have lower dispatch priority.

[0113] In this step, if the types of task units in the task pool include tasks, subtasks, and iterative subtasks, the tasks are first sorted according to the task sorting method. Then, the subtasks under each task branch are sorted in ascending order of subtask sequence numbers. Then, the iterative subtasks under each task branch are sorted in descending order of iteration order and descending order of subtask sequence numbers. Task units with higher rankings have higher dispatch priority, while task units with lower rankings have lower dispatch priority.

[0114] For example, if a task is divided into two subtasks, and one subtask is divided into two iterative subtasks, the final sorting result is 2-iteration subtask 1, 2-iteration subtask 2, 1-iteration subtask 1, and 1-iteration subtask 1. The dispatching priority is descending in descending order.

[0115] In some embodiments, the method of determining the dispatching priority of each task unit in the task pool according to the type of each task unit in the task pool includes: if there are multiple types of task units in the task pool, then determining the dispatching priority of each type of task unit separately, and determining the dispatching priority of each task unit in the task pool according to the dispatching priority of each type of task unit.

[0116] Among them, the dispatching priority of a task unit of type task is determined according to the configured weight, scheduling frequency and task execution deadline of the task to which the task unit belongs; the dispatching priority of a task unit of type subtask is determined according to the subtask sequence number; the dispatching priority of a task unit of type iterative task is determined according to the iteration order; the dispatching priority of a task unit of type iterative subtask is determined according to the iteration order and subtask sequence number of the iterative subtask.

[0117] In this step, the dispatch priority of each task unit in the task pool is determined based on the dispatch priority of each type of task unit. That is, the dispatch priority of each type of task unit is summarized to obtain the dispatch priority of each task unit in the task pool. The specific summarization process is described in steps 301 to 304 above. When determining the dispatch priority of a task unit, the configured weight, scheduling frequency, and task execution deadline of the task to which the task unit belongs are given priority, followed by the number of iterations and then the subtask sequence number.

[0118] In some embodiments, the method further comprises:

[0119] If the type of the task unit in the task pool is an iterative task or an iterative subtask, and it is determined according to the algorithm that the number of iterations is not fixed, then when the processing result of the previous iteration returned by the first network element device is received, if it is determined according to the processing result that the iteration has not ended, the next iterative task or the next iterative subtask is generated, and the next iterative task or the next iterative subtask is placed in the task pool.

[0120] In this step, if the task unit type in the task pool is an iterative task or iterative subtask, and the algorithm determines that the number of iterations is not fixed, all iterative tasks or iterative subtasks other than the first iterative task or the first iterative subtask need to be newly generated. The newly generated iterative task or iterative subtask is placed in the task pool, and the dispatch priority of each task unit in the task pool is recalculated. When the dispatch priority of the newly generated iterative task or iterative subtask is optimized, it is sent to the first network element device. Until the MEC receives a processing result returned by the first network element device and determines that the iteration is complete based on the processing result, no new iterative tasks or iterative subtasks will be generated.

[0121] Further, Figure 4 Another flow chart of the task processing method provided by the embodiment of the present disclosure is shown. Figure 4 As shown, the method further includes:

[0122] Step 401: If the allocation type is a subtask or an iterative subtask, then the processing results returned by each first network element device are received.

[0123] In step 402, if it is determined based on the processing result that the corresponding task has been completed and the current processing capability of the edge computing device meets the preset second requirement, the processing result is iteratively optimized to generate a model.

[0124] In step 403, if it is determined based on the processing result that the corresponding task has been completed and the current processing capability of the edge computing device does not meet the preset second requirement, a convergence task is generated, a second network element device is determined, and the convergence task is sent to the second network element device.

[0125] In steps 401-403, if the assignment type is a subtask or iterative subtask, the processing results returned by each first network element device need to be aggregated. If the MEC's ​​current processing capacity meets the preset second requirement (i.e., its own calculation instances are sufficient to process the aggregated task), the processing results returned by each first network element device are iteratively optimized to generate a model. If the MEC's ​​current processing capacity does not meet the preset second requirement (i.e., its own calculation instances are insufficient to process the aggregated task), an aggregated task is generated, a second network element device is determined, and the aggregated task is sent to the second network element device.

[0126] It should be noted that the aggregation task here does not need to be split. The second network element device can be specified or any network element device. The aggregation process can involve various algorithms, which are not limited here.

[0127] For example, the iterative processing results of each first network element device are gradients, hyperparameters, and model output data. Then, the output results of the aggregation task completion are gradients and hyperparameters.

[0128] Further, Figure 5 Another flow chart of the task processing method provided by the embodiment of the present disclosure is shown. Figure 5 As shown, the determining of the first network element device (i.e., step 106) includes:

[0129] Step 501: Acquire network element data of each network element device in a network element data pool.

[0130] The network element data includes distance parameters and performance parameters. Performance parameters are parameters indicating the load capacity of the network element.

[0131] A network element data pool is pre-established in the MEC. The MEC periodically collects network element data from each connected network element and stores it in the network element data pool. The data in the network element data pool is updated in real time. It should be noted that the network element data collection period is inconsistent with the delivery period of the task units to be processed. The network element data collection process and the delivery process of the task units to be processed are independent of each other.

[0132] Step 502: Determine a network element device whose distance parameters and performance parameters meet preset conditions.

[0133] In this step, a network element set is first determined based on distance parameters and performance parameters.

[0134] Step 503: Determine the first network element device according to the performance parameters.

[0135] In this step, a first network element device is determined from network element devices that meet preset conditions based on performance parameters. The first network element device is the network element device with the best computing power determined by the MEC in the current cycle.

[0136] In some embodiments, the performance parameters include: the remaining space ratio of the central processing unit CPU, the remaining space ratio of the memory MEM, the remaining space ratio of the disk DISK and the remaining space ratio of the cache memory CACHE; the distance parameters and performance parameters meet the preset conditions including: the distance parameter is less than or equal to the first threshold, and the remaining space ratio of the CPU is greater than the second threshold, and the remaining space ratio of the MEM is greater than the third threshold, and the remaining space ratio of the DISK is greater than the fourth threshold, and the remaining space ratio of the CACHE is greater than the fifth threshold.

[0137] It should be noted that when designating the first network element to receive the pending task unit, consideration should be given to network elements that are closer to the MEC and have higher CPU free space ratios, higher MEM free space ratios, higher disk free space ratios, and higher cache free space ratios. Specifically, consideration should be given to network elements that are closer to the MEC and have higher computing power. The selected network elements must also be online.

[0138] Determining the first network element device according to the performance parameters includes: sorting the network element devices according to the CPU remaining space ratio, MEM remaining space ratio, DISK remaining space ratio and CACHE remaining space ratio; and determining the first network element device according to the sorting.

[0139] In the above steps, first sort the network elements in descending order of CPU remaining space ratio. Then sort the network elements with the same CPU remaining space ratio in descending order of MEM remaining space ratio. Then sort the network elements with the same CPU remaining space ratio and MEM remaining space ratio in descending order of DISK remaining space ratio. Then sort the network elements with the same CPU remaining space ratio, MEM remaining space ratio, and DISK remaining space ratio in descending order of CACHE remaining space ratio. The network elements with higher rankings have higher sending priorities, while the network elements with lower rankings have lower sending priorities. The first network element is the network element that ranks first in the sequence, that is, the first network element has the strongest computing power in the sequence.

[0140] In the embodiment of the present disclosure, when MEC computing power resources are insufficient, based on the idea of ​​federated learning, a first network element device with sufficient computing power is selected from the surrounding network element group to perform task transfer calculations, sharing the computing power of the network element devices, thereby improving the edge computing capabilities of MEC and reducing task calculation delays.

[0141] In some embodiments, after sending the task unit to be processed to the first network element device (step 107), the method further includes:

[0142] If a task interruption message returned by the first network element device is received and the current processing capability of the edge computing device does not meet the preset third requirement, a third network element device is determined so that the third network element device processes the task unit corresponding to the task interruption message.

[0143] In this step, if the MEC receives a task interruption message returned by the first network element device, and the current processing capability of the MEC does not meet the preset third requirement, that is, the computing power of the MEC is insufficient to process the task unit corresponding to the task interruption message, the third network element device is re-determined so that the third network element device processes the task unit corresponding to the task interruption message.

[0144] In a preferred implementation, the third network element device may also receive and process the unfinished portion of the task in the task unit corresponding to the task interruption message.

[0145] In the embodiment of the present disclosure, when task processing is interrupted in a first network element device, the task can be quickly transferred to another first network element device, and the task calculation is finally completed after a limited number of task transfers.

[0146] In some embodiments, after sending the task unit to be processed to the first network element device (step 107), the method further includes:

[0147] If the processing result returned by the first network element device is not received within the preset period, other first network element devices are determined, and the task units to be processed are sent to the other first network element devices.

[0148] After determining the assignment type of each task and dividing each task into task units based on the assignment type, a preset deadline, i.e., an execution deadline for the task unit, is assigned to each task unit. If no processing result is received from the first network element device within the preset deadline, indicating that the first network element device has not completed the task within the preset deadline, another first network element device is identified, and the task unit to be processed is sent to the other first network element device.

[0149] In the disclosed embodiment, especially when the task is an urgent real-time task, when the MEC computing power resources are insufficient, based on the idea of ​​federated learning, another first network element device with sufficient computing power is quickly selected from the surrounding network element group to perform task transfer calculations, thereby improving the edge computing capabilities of MEC and reducing task calculation delays.

[0150] Figure 6 The schematic diagram of the structure of the edge computing device provided by the embodiment of the present disclosure is shown. Figure 1 The corresponding embodiments have the same technical concept, such as Figure 6 As shown, the edge computing device provided by the embodiment of the present disclosure includes the following modules.

[0151] The receiving module 11 is used to receive tasks sent by a network device.

[0152] The first determining module 12 is configured to determine the assignment type of the task.

[0153] The division module 13 is configured to divide the task into task units according to the allocation type and place the task units into a task pool.

[0154] The second determining module 14 is configured to determine the issuing priority of each task unit in the task pool according to the type of each task unit in the task pool, where the type of the task unit is the allocation type of the task to which the task unit belongs.

[0155] The third determining module 15 is configured to determine the task unit to be processed according to the sending priority.

[0156] The fourth determination module 16 is configured to determine a first network element device if the current processing capability of the edge computing device does not meet the preset first requirement.

[0157] The sending module 17 is configured to send the task unit to be processed to the first network element device, so that the first network element device processes the task unit to be processed.

[0158] In some embodiments, the assignment types include tasks, subtasks, iterative tasks, and iterative subtasks.

[0159] In some embodiments, the task includes a task type, and before determining the assignment type of the task, the method further includes: configuring an algorithm for the task according to the task type of the task;

[0160] The first determining module 12 is specifically configured to:

[0161] If it is determined according to the task type that the task cannot be divided into subtasks, and it is determined according to the algorithm that the task cannot be divided into iterative tasks, then the allocation type of the task is determined to be task.

[0162] If it is determined according to the task type that the task cannot be divided into subtasks, and it is determined according to the algorithm that the task can be divided into an iterative task, then the assignment type of the task is determined to be an iterative task.

[0163] If it is determined according to the task type that the task can be divided into subtasks, and it is determined according to the algorithm that the subtask cannot be divided into iterative subtasks, then the allocation type of the task is determined to be a subtask.

[0164] If it is determined according to the task type that the task can be divided into subtasks, and if it is determined according to the algorithm that the subtask can be divided into iterative subtasks, then the allocation type of the task is determined to be an iterative subtask.

[0165] In some embodiments, the second determining module 14 is specifically configured to:

[0166] If the type of each task unit in the task pool includes a task, the issuing priority of each task unit in the task pool is determined according to the configured weight, scheduling frequency and task execution deadline of the task.

[0167] If the types of each task unit in the task pool include tasks and subtasks, then determine the tasks to which each task unit in the task pool belongs, and determine the dispatching priority of each task unit in the task pool based on the configured weight, scheduling frequency, task execution deadline and subtask sequence number of the task.

[0168] If the types of each task unit in the task pool include tasks and iterative tasks, then determine the tasks to which each task unit in the task pool belongs, and determine the dispatching priority of each task unit in the task pool based on the configured weight, scheduling frequency, task execution deadline and iteration order of the task.

[0169] If the types of each task unit in the task pool include tasks, subtasks and iterative tasks, or the types of each task unit in the task pool include tasks, subtasks and iterative subtasks, or the types of each task unit in the task pool include tasks, subtasks, iterative tasks and iterative subtasks, then determine the tasks to which each task unit in the task pool belongs, and determine the issuance priority of each task unit in the task pool according to the configured weight, scheduling frequency, task execution deadline, iteration order and subtask sequence number of the task.

[0170] In some embodiments, the second determining module 14 is specifically configured to:

[0171] If the task units in the task pool include multiple types, the issuing priority of each type of task unit is determined respectively, and the issuing priority of each task unit in the task pool is determined according to the issuing priority of each type of task unit.

[0172] Among them, the sending priority of the task unit of type task is determined according to the configured weight, scheduling frequency and task execution deadline of the task to which the task unit belongs; the sending priority of the task unit of type subtask is determined according to the subtask sequence number; the sending priority of the task unit of type iterative task is determined according to the iteration order; the sending priority of the task unit of type iterative subtask is determined according to the iteration order and subtask sequence number.

[0173] In some embodiments, the edge computing device further includes:

[0174] The first generation module is used to generate the next iterative task or the next iterative subtask if the type of the task unit in the task pool is an iterative task or an iterative subtask, and it is determined according to the algorithm that the number of iterations is not fixed. When the processing result of the previous iteration returned by the first network element device is received, if it is determined according to the processing result that the iteration has not ended, the next iterative task or the next iterative subtask is generated.

[0175] A storage module is used to put the next iteration task or the next iteration subtask into the task pool.

[0176] In some embodiments, the edge computing device further includes:

[0177] The receiving module 11 is further configured to receive processing results returned by each first network element device if the allocation type is a subtask or an iterative subtask.

[0178] The second generation module is used to iteratively optimize the processing result and generate a model if it is determined that the corresponding task has been completed based on the processing result and the current processing capability of the edge computing device meets the preset second requirement.

[0179] The sending module 17 is also used to generate a convergence task, determine a second network element device, and send the convergence task to the second network element device if it is determined based on the processing result that the corresponding task has been completed and the current processing capacity of the edge computing device does not meet the preset second requirement.

[0180] In some embodiments, the fourth determining module 16 is specifically configured to:

[0181] The network element data of each network element device in the network element data pool is obtained, where the network element data includes a distance parameter and a performance parameter.

[0182] Determine the network element device whose distance parameter and performance parameter meet the preset conditions.

[0183] The first network element device is determined according to the performance parameter.

[0184] In some embodiments, the performance parameters include: a remaining space ratio of a central processing unit (CPU), a remaining space ratio of a memory (MEM), a remaining space ratio of a disk (DISK), and a remaining space ratio of a cache memory (CACHE).

[0185] The distance parameter and the performance parameter meet the preset conditions, including: the distance parameter is less than or equal to the first threshold, the CPU remaining space ratio is greater than the second threshold, the MEM remaining space ratio is greater than the third threshold, the DISK remaining space ratio is greater than the fourth threshold, and the CACHE remaining space ratio is greater than the fifth threshold.

[0186] The fourth determining module 16 is specifically configured to:

[0187] The network element devices are sorted according to the CPU remaining space ratio, the MEM remaining space ratio, the DISK remaining space ratio, and the CACHE remaining space ratio.

[0188] The first network element device is determined according to the ranking.

[0189] In some embodiments, the edge computing device further includes:

[0190] The fifth determination module is used to determine the third network element device if a task interruption message returned by the first network element device is received and the current processing capability of the edge computing device does not meet the preset third requirement, so that the third network element device processes the task unit corresponding to the task interruption message.

[0191] In some embodiments, the edge computing device further includes:

[0192] The sixth determining module is configured to determine other first network element devices if the processing result returned by the first network element device is not received within a preset period, and send the task unit to be processed to the other first network element devices.

[0193] An embodiment of the present disclosure also provides a computer device, which includes: one or more processors and a storage device; wherein one or more programs are stored on the storage device, and when the one or more programs are executed by the one or more processors, the one or more processors implement the task processing method provided in the aforementioned embodiments.

[0194] An embodiment of the present disclosure further provides a computer-readable medium having a computer program stored thereon, wherein the computer program, when executed, implements the task processing method provided in the aforementioned embodiments.

[0195] It will be appreciated by those skilled in the art that all or some of the steps in the method disclosed above, and the functional modules / units in the device can be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware implementation, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As is well known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media generally embodies computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0196] Example embodiments have been disclosed herein, and although specific terms are employed, they are used and should be interpreted only in a general illustrative sense and not for purposes of limitation. In some instances, it will be apparent to those skilled in the art that, unless otherwise expressly indicated, features, characteristics, and / or elements described in conjunction with a particular embodiment may be used alone or in combination with features, characteristics, and / or elements described in conjunction with other embodiments. Therefore, it will be understood by those skilled in the art that various changes in form and detail may be made without departing from the scope of the invention as set forth in the appended claims.

Claims

1. A task processing method, comprising: receiving a task sent by a network device, wherein the task includes a task type; configuring an algorithm for the task according to the task type of the task; Determine the assignment type of the task, divide the task into task units according to the assignment type, and place the task units into a task pool; the assignment types include tasks, subtasks, iterative tasks, and iterative subtasks; Determining the dispatch priority of each task unit in the task pool according to the type of each task unit in the task pool, where the type of the task unit is the allocation type of the task to which the task unit belongs; Determine the task unit to be processed according to the priority of the dispatch; If the current processing capability of the edge computing device does not meet the preset first requirement, determining a first network element device and sending the task unit to be processed to the first network element device so that the first network element device processes the task unit to be processed; The method further comprises: If the type of the task unit in the task pool is an iterative task or an iterative subtask, and it is determined according to the algorithm that the number of iterations is not fixed, then when the processing result of the previous iteration returned by the first network element device is received, if it is determined according to the processing result that the iteration has not ended, the next iterative task or the next iterative subtask is generated, the next iterative task or the next iterative subtask is placed in the task pool, and the issuance priority of each task unit in the task pool is re-determined.

2. The method according to claim 1, wherein Determining the assignment type of the task includes: If it is determined according to the task type that the task cannot be divided into subtasks, and it is determined according to the algorithm that the task cannot be divided into iterative tasks, then determining the assignment type of the task to be task; If it is determined according to the task type that the task cannot be divided into subtasks, and it is determined according to the algorithm that the task can be divided into an iterative task, then determining the assignment type of the task to be an iterative task; If it is determined according to the task type that the task can be divided into subtasks, and it is determined according to the algorithm that the subtask cannot be divided into iterative subtasks, then determining the assignment type of the task to be a subtask; If it is determined according to the task type that the task can be divided into subtasks, and if it is determined according to the algorithm that the subtask can be divided into iterative subtasks, then the allocation type of the task is determined to be an iterative subtask.

3. The method according to claim 1, wherein Determining the issuing priority of each task unit in the task pool according to the type of each task unit in the task pool includes: If the type of each task unit in the task pool includes a task, determining the dispatch priority of each task unit in the task pool according to the configured weight, scheduling frequency and task execution deadline of the task; If the types of each task unit in the task pool include tasks and subtasks, then determining the tasks to which each task unit in the task pool belongs, and determining the dispatching priority of each task unit in the task pool according to the configured weight, scheduling frequency, task execution deadline and subtask sequence number of the task; If the types of each task unit in the task pool include tasks and iterative tasks, then determining the tasks to which each task unit in the task pool belongs, and determining the dispatching priority of each task unit in the task pool according to the configured weight, scheduling frequency, task execution deadline and iteration order of the task; If the types of each task unit in the task pool include tasks, subtasks and iterative tasks, or the types of each task unit in the task pool include tasks, subtasks and iterative subtasks, or the types of each task unit in the task pool include tasks, subtasks, iterative tasks and iterative subtasks, then determine the tasks to which each task unit in the task pool belongs, and determine the issuance priority of each task unit in the task pool according to the configured weight, scheduling frequency, task execution deadline, iteration order and subtask sequence number of the task.

4. The method according to claim 3, wherein: Determining the issuing priority of each task unit in the task pool according to the type of each task unit in the task pool includes: If the task units in the task pool include multiple types, determining the issuing priority of each type of task unit respectively, and determining the issuing priority of each task unit in the task pool according to the issuing priority of each type of task unit; Among them, the sending priority of the task unit of type task is determined according to the configured weight, scheduling frequency and task execution deadline of the task to which the task unit belongs; the sending priority of the task unit of type subtask is determined according to the subtask sequence number; the sending priority of the task unit of type iterative task is determined according to the iteration order; the sending priority of the task unit of type iterative subtask is determined according to the iteration order and subtask sequence number.

5. The method according to claim 1, wherein The method further comprises: If the allocation type is a subtask or an iterative subtask, receiving the processing results returned by each first network element device; If it is determined according to the processing result that the corresponding task has been completed and the current processing capability of the edge computing device meets the preset second requirement, iteratively optimizing the processing result to generate a model; If it is determined based on the processing result that the corresponding task has been completed and the current processing capability of the edge computing device does not meet the preset second requirement, a convergence task is generated, a second network element device is determined, and the convergence task is sent to the second network element device.

6. The method of claim 1, wherein: The determining of the first network element device includes: Acquire network element data of each network element device in the network element data pool, wherein the network element data includes distance parameters and performance parameters; Determining a network element device whose distance parameter and performance parameter meet preset conditions; The first network element device is determined according to the performance parameter.

7. The method according to claim 6, wherein: The performance parameters include: the remaining space ratio of the central processing unit (CPU), the remaining space ratio of the memory (MEM), the remaining space ratio of the disk (DISK) and the remaining space ratio of the cache memory (CACHE); The distance parameter and the performance parameter satisfying the preset conditions include: the distance parameter is less than or equal to a first threshold, the CPU remaining space ratio is greater than a second threshold, the MEM remaining space ratio is greater than a third threshold, the DISK remaining space ratio is greater than a fourth threshold, and the CACHE remaining space ratio is greater than a fifth threshold; The determining the first network element device according to the performance parameter includes: Sort the network element devices according to the CPU remaining space ratio, MEM remaining space ratio, DISK remaining space ratio and CACHE remaining space ratio; The first network element device is determined according to the ranking.

8. The method according to any one of claims 1 to 7, wherein: After sending the task unit to be processed to the first network element device, the method further includes: If a task interruption message returned by the first network element device is received and the current processing capability of the edge computing device does not meet the preset third requirement, a third network element device is determined so that the third network element device processes the task unit corresponding to the task interruption message.

9. The method of claim 1, wherein: After sending the task unit to be processed to the first network element device, the method further includes: If the processing result returned by the first network element device is not received within the preset period, other first network element devices are determined, and the task units to be processed are sent to the other first network element devices.

10. An edge computing device, comprising: A receiving module, configured to receive a task sent by a network device, wherein the task includes a task type; A first determining module is configured to configure an algorithm for the task according to the task type of the task and determine an allocation type of the task, where the allocation type includes a task, a subtask, an iterative task, and an iterative subtask; a partitioning module, configured to partition the task into task units according to the allocation type, and place the task units into a task pool; A second determining module is configured to determine the issuing priority of each task unit in the task pool according to the type of each task unit in the task pool, where the type of the task unit is the allocation type of the task to which the task unit belongs; A third determining module is used to determine the task unit to be processed according to the sending priority; A fourth determining module is configured to determine a first network element device if the current processing capability of the edge computing device does not meet the preset first requirement; A sending module, configured to send the task unit to be processed to the first network element device, so that the first network element device processes the task unit to be processed; A first generating module is configured to, if the type of the task unit in the task pool is an iterative task or an iterative subtask, and if it is determined according to the algorithm that the number of iterations is not fixed, upon receiving a processing result of a previous iteration returned by the first network element device, generate a next iterative task or a next iterative subtask if it is determined according to the processing result that the iteration is not completed; A storage module, configured to place the next iteration task or the next iteration subtask into the task pool; The second determination module is further configured to, after the storage module places the next iteration task or the next iteration subtask into the task pool, re-determine the sending priority of each task unit in the task pool.

11. A computer device comprising: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement the task processing method according to any one of claims 1 to 9.

12. A computer-readable medium having a computer program stored thereon, wherein: When the program is executed, the task processing method according to any one of claims 1 to 9 is implemented.

Images