Task unloading method, device, equipment and product

Abstract

The invention provides a task unloading method and device, equipment and a product, and the method comprises the steps: obtaining a first task with the highest priority in a plurality of to-be-unloaded tasks sorted according to the priority; wherein the priority is obtained according to the task characteristics of the to-be-unloaded task, the user behavior mode and the network condition; determining the channel with the minimum bidding density in the plurality of channels as a first channel; acquiring a first unloading probability of the first channel; wherein the first unloading probability is a probability that the first task is unloaded in a first channel in the plurality of channels; and under the condition that the first unloading probability is greater than a first threshold value, determining that the first task is unloaded through the first channel. The first task with the highest priority is determined according to the key index of the to-be-unloaded task, so that more emergency tasks can be processed preferentially according to the key index, the message priority can be clarified, the real-time communication processing capability can be enhanced, a channel with lower bidding density can be selected for unloading according to the current dynamic environment, and the adaptability of the dynamic environment is improved.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a task offloading method, device, equipment and product. Background Art

[0002] In the existing technology, task offloading technology allows computing tasks on mobile devices to be migrated to the cloud or edge nodes to improve efficiency; and resource optimization algorithms are designed to efficiently manage and allocate computing resources to ensure service quality while minimizing energy consumption and latency. However, current task offloading and resource optimization technologies have problems such as insufficient real-time communication processing, insufficient adaptability to dynamic environments, and unclear message priority processing. Summary of the invention

[0003] The purpose of the present invention is to provide a task offloading method, device, equipment and product for solving the problems of insufficient real-time communication processing, insufficient adaptability to dynamic environments and unclear message priority processing in the task offloading and resource optimization technologies in the prior art.

[0004] To achieve the above object, an embodiment of the present invention provides a task offloading method, which includes:

[0005] Acquire a first task with the highest priority among a plurality of tasks to be uninstalled that are sorted by priority; wherein the priority is acquired according to the task characteristics, user behavior patterns, and network conditions of the tasks to be uninstalled;

[0006] Determine a channel with the smallest bidding density among the multiple channels as a first channel;

[0007] Obtaining a first unloading probability of the first channel; wherein the first unloading probability is a probability of unloading the first task on the first channel among the multiple channels;

[0008] When the first offloading probability is greater than a first threshold, it is determined that the first task is offloaded through the first channel.

[0009] Optionally, in the method, before obtaining the first task with the highest priority among the plurality of tasks to be offloaded sorted by priority, the method further comprises:

[0010] Receiving multiple instant messages from multiple users;

[0011] Classifying and marking the multiple instant messages to obtain the multiple tasks to be offloaded;

[0012] Inputting the task characteristics of the multiple tasks to be offloaded, the user behavior pattern and the network status into a priority prediction model to obtain a prediction result; wherein the priority prediction model is obtained by training based on historical task offloading data;

[0013] The priorities of the multiple tasks to be offloaded are determined according to the prediction results.

[0014] Optionally, in the method, before determining the channel with the smallest bidding density among the multiple channels as the first channel, the method further includes:

[0015] When it is determined that the threshold value of one of the multiple channels is less than or equal to the second threshold value, the corresponding channel is removed from the multiple channels.

[0016] Optionally, in the method, after determining that the first task is to be unloaded through the first channel, the method further comprises:

[0017] Updating the historical load; the historical load is a record list for recording the unloading of the first task through the first channel;

[0018] The threshold value of the first channel is adjusted according to the historical load.

[0019] Optionally, in the method, after determining that the first task is to be unloaded through the first channel, the method further comprises:

[0020] Execute the first task on the server where the first task is currently located;

[0021] The first task is removed from the multiple tasks to be uninstalled.

[0022] Optionally, the method, before determining that the first task is to be unloaded through the first channel, further comprises:

[0023] determining a security assessment value of the first channel;

[0024] Determine whether the security assessment value is greater than or equal to a third threshold value, and obtain a second determination result;

[0025] When the second determination result is yes, it is determined whether the first uninstallation probability is greater than the first threshold.

[0026] Optionally, the method further comprises:

[0027] When the second judgment result is no, the first channel is removed from the plurality of channels.

[0028] In order to achieve the above object, an embodiment of the present invention further provides a task offloading device, which includes:

[0029] A first acquisition module, used to acquire a first task with the highest priority among a plurality of tasks to be uninstalled sorted by priority; wherein the priority is acquired according to the task characteristics, user behavior patterns and network conditions of the tasks to be uninstalled;

[0030] A first determining module, configured to determine a channel with the smallest bidding density among the multiple channels as a first channel;

[0031] A second acquisition module is used to acquire a first unloading probability of the first channel; wherein the first unloading probability is a probability of unloading the first task on the first channel among the multiple channels;

[0032] The second determining module is configured to determine that the first task is to be offloaded through the first channel when the first offloading probability is greater than a first threshold.

[0033] In order to achieve the above-mentioned purpose, an embodiment of the present invention also provides a network device, including: a transceiver, a processor, a memory, and a program or instruction stored in the memory and executable on the processor; wherein, when the processor executes the program or instruction, the task offloading method as described above is implemented.

[0034] In order to achieve the above-mentioned purpose, an embodiment of the present invention further provides a readable storage medium on which a program or instruction is stored, wherein the program or instruction, when executed by a processor, implements the steps in the task offloading method as described above.

[0035] In order to achieve the above object, an embodiment of the present invention further provides a computer program product, which includes computer instructions, and when the computer instructions are executed by a processor, the steps of the task offloading method described above are implemented.

[0036] The beneficial effects of the above technical solution of the present invention are as follows:

[0037] The embodiment of the present invention determines the first task with the highest priority according to the key indicators of the tasks to be offloaded, traverses each channel of multiple channels for the first task to obtain the corresponding offloading probability, and when it is determined that the offloading probability of the first channel with the smallest bidding density is greater than a first threshold, the first task is offloaded using the first channel. In this way, more urgent tasks can be processed preferentially according to key indicators, message priorities can be clarified and real-time communication processing capabilities can be enhanced, and channels with smaller bidding density can be selected for offloading according to the current dynamic environment, thereby improving adaptability to dynamic environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A schematic diagram of a task offloading method according to an embodiment of the present invention;

[0039] Figure 2A structural diagram of a task offloading method according to an embodiment of the present invention;

[0040] Figure 3 A schematic diagram of an intelligent adaptive dynamic resource allocation algorithm for a task offloading method according to an embodiment of the present invention;

[0041] Figure 4 It is a schematic diagram of a task offloading device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0042] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0043] It should be understood that the references to "one embodiment" or "an embodiment" throughout the specification mean that the specific features, structures, or characteristics associated with the embodiment are included in at least one embodiment of the present invention. Therefore, the references to "in one embodiment" or "in an embodiment" appearing throughout the specification do not necessarily refer to the same embodiment. In addition, these specific features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0044] In various embodiments of the present invention, it should be understood that the size of the serial numbers of the following processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0045] Additionally, the terms "system" and "network" are often used interchangeably herein.

[0046] In the embodiments provided in this application, it should be understood that "B corresponding to A" means that B is associated with A, and B can be determined according to A. However, it should also be understood that determining B according to A does not mean determining B only according to A, but B can also be determined according to A and / or other information.

[0047] For ease of understanding, some contents involved in the embodiments of the present invention are described below:

[0048] like Figure 1 As shown, a task offloading method according to an embodiment of the present invention includes:

[0049] S10, obtaining a first task with the highest priority among a plurality of tasks to be offloaded sorted by priority; wherein the priority is obtained according to the task characteristics of the tasks to be offloaded, the user behavior pattern and the network status;

[0050] It should be noted that if Figure 3As shown, the wireless access point AP uses machine learning technology to analyze historical operation data, establish a prediction model, predict the task priority according to the task execution requirements, user behavior patterns and network conditions, and obtain the first task with the highest priority. The wireless access point AP uses historical operation data information to train a regression prediction model of task priority based on a support vector machine (SVM) offline:

[0051] P i (t) = f(x i ,y i , z i θ i )

[0052] Among them, P i (t) is the priority prediction of the i-th task at time t, (x i ,y i , z i ) represent the feature vectors of the task characteristics, user behavior patterns and network conditions of the i-th task, respectively, and θ i is the model parameter of the i-th task. Using the prediction results, the corresponding priorities are determined for various users and different tasks, and it is decided whether to offload the task to the edge cloud server through the wireless channel and to which edge cloud server to offload it, so as to determine whether the task ensures that the critical tasks with high real-time requirements are processed in time, while optimizing the overall resource utilization efficiency. The embodiment of the present invention receives the priorities of different tasks of the user, as well as the task characteristics and the network device status, and uses the machine learning model to predict future task requirements and network changes, so as to make more accurate offloading decisions in advance.

[0053] S20, determining a channel with the smallest bidding density among the multiple channels as a first channel;

[0054] It should be noted that the first task is unloaded through the first channel with the smallest bidding density, so as to unload the first task with maximum efficiency.

[0055] S30, obtaining a first unloading probability of the first channel; wherein the first unloading probability is a probability of unloading the first task on the first channel among the multiple channels;

[0056] It should be noted that the first task traverses all the channels in the multiple channels, and a prediction model is used to predict the probability of the first task passing through each of the multiple channels, so as to obtain the first unloading probability of the first task being unloaded in the first channel.

[0057] S40: When the first offloading probability is greater than a first threshold, determine that the first task is to be offloaded through the first channel.

[0058] In this embodiment, the first task with the highest priority is determined based on the key indicators of the tasks to be unloaded, and the first task is traversed through each of the multiple channels to obtain a corresponding unloading probability. When it is determined that the unloading probability of the first channel with the smallest bidding density is greater than the first threshold, the first task is unloaded using the first channel. This allows more urgent tasks to be processed preferentially based on key indicators, clarifies message priorities and enhances real-time communication processing capabilities, and allows channels with smaller bidding density to be selected for unloading based on the current dynamic environment, thereby improving efficiency and enhancing adaptability to dynamic environments.

[0059] Optionally, in the method, before obtaining the first task with the highest priority among the plurality of tasks to be offloaded sorted by priority, the method further comprises:

[0060] Receiving multiple instant messages from multiple users;

[0061] Classifying and marking the multiple instant messages to obtain the multiple tasks to be offloaded;

[0062] Inputting the task characteristics of the multiple tasks to be offloaded, the user behavior pattern and the network status into a priority prediction model to obtain a prediction result; wherein the priority prediction model is obtained by training based on historical task offloading data;

[0063] The priorities of the multiple tasks to be offloaded are determined according to the prediction results.

[0064] In this embodiment, if Figure 2As shown, the instant messaging (before prioritization) module is responsible for receiving and processing instant messaging requests from users, including messages in various forms such as text, voice, and video. At this stage, the messages have not yet been prioritized, and the system needs to perform preliminary classification and labeling to form the multiple tasks to be uninstalled. The real-time monitoring module monitors the operating status of the system in real time, including key indicators such as network status, device performance, and resource usage. The real-time monitoring module can ensure that the system has a clear understanding of the current operating environment and provide necessary data support for the resource allocation algorithm. In particular, real-time task identification can be performed to help the system identify real-time tasks that have strict requirements on delay and response time in instant messaging. The task characteristics, user behavior patterns, and network conditions of the multiple tasks to be uninstalled are input into the priority prediction model in the intelligent adaptive dynamic resource allocation module, and different priorities are set for different users and the multiple tasks to be uninstalled, so as to make smarter decisions. After all instant messages are reorganized and sorted according to their priorities to form the multiple tasks to be uninstalled through the instant messaging (after prioritization) module, the high-priority tasks to be uninstalled are prioritized according to the urgency of the task and the resource allocation results to ensure the timeliness and effectiveness of key communications.

[0065] Optionally, in the method, before determining the channel with the smallest bidding density among the multiple channels as the first channel, the method further includes:

[0066] When it is determined that the threshold value of one of the multiple channels is less than or equal to the second threshold value, the corresponding channel is removed from the multiple channels.

[0067] In this embodiment, the channel that cannot offload tasks among the multiple channels is determined by using the threshold value, and is removed from the multiple channels.

[0068] Optionally, in the method, after determining that the first task is to be unloaded through the first channel, the method further comprises:

[0069] Updating the historical load; the historical load is a record list for recording the unloading of the first task through the first channel;

[0070] The threshold value of the first channel is adjusted according to the historical load.

[0071] In this embodiment, after determining that the first channel is to be unloaded with the first task, the historical load is updated, and the threshold value of the first channel is adjusted according to the historical load, so that the usage of the first channel can be determined at any time, the threshold value of the channel can be adjusted dynamically, and the unloading decision can be updated at any time. Through real-time feedback adjustment and dynamic threshold value adjustment, the efficiency of task unloading and the overall performance of the system are improved.

[0072] Optionally, in the method, after determining that the first task is to be unloaded through the first channel, the method further comprises:

[0073] Execute the first task on the server where the first task is currently located;

[0074] The first task is removed from the multiple tasks to be uninstalled.

[0075] In this embodiment, after the first task is uninstalled, the multiple tasks to be uninstalled are removed in time, and resource allocation decisions are dynamically adjusted according to real-time network conditions and changes in device mobility to enhance the robustness and stability of the system.

[0076] Optionally, the method, before determining that the first task is to be unloaded through the first channel, further comprises:

[0077] determining a security assessment value of the first channel;

[0078] Determine whether the security assessment value is greater than or equal to a third threshold value, and obtain a second determination result;

[0079] When the second determination result is yes, it is determined whether the first uninstallation probability is greater than the first threshold.

[0080] Optionally, the method further comprises:

[0081] When the second judgment result is no, the first channel is removed from the plurality of channels.

[0082] In this embodiment, before unloading the first task, it is necessary to confirm whether the first channel meets the security requirements, and compare the security assessment value of the first channel with the third threshold. The first task can be unloaded only if the third threshold is reached. If the third threshold cannot be reached, the first channel is removed from the multiple channels and the first channel is re-determined.

[0083] It should be noted that the embodiment of the present invention adopts an adaptive multi-objective optimization task offloading algorithm, and the specific pseudo code is as follows:

[0084] enter:

[0085] U1, Uo: classified task set (Ul represents the task set suitable for local execution, Uo represents the task set suitable for offloading)

[0086] eta: a set of priorities for different tasks

[0087] ML_model: A trained machine learning model used to predict the probability and performance indicators of task offloading.

[0088] security_threshold: security threshold

[0089] Objectives: multi-objective optimization objective set (time delay and energy consumption)

[0090] Output:

[0091] psi: offloading decision solutions for different tasks, psi_{i,j} indicates whether task i is offloaded through channel j

[0092] initialization:

[0093] psi_{i,j}=0, for all tasks i and channels j, the initial offloading decision is not to offload

[0094] threshold = security threshold, used for comparative security assessment

[0095] Load_history = Empty load history list, used to record the historical load of each channel

[0096] Set a temporary set: U'o = Uo

[0097] Cycle process:

[0098] while U'o! = empty:

[0099] i=argmax{eta_i},i∈U'o / / Select the task with the highest priority

[0100] for j=1toM: / / traverse all channels

[0101] Use ML_model to predict P_{i,j} / / predict the probability of task i being unloaded through channel j

[0102] Update R_{i,j} (the rate at which task i selects wireless channel j to upload data) and E_{off,i,j} (the energy consumption of task i offloading to the edge server) based on real-time monitoring data

[0103] Calculate the value of each target, such as local energy consumption E_{l_i} and transmission energy consumption E_{tr,i,j}, local delay D_{l_i} and transmission delay D_{tr,i,j}, etc.

[0104] If the threshold value π_j of channel j>0:

[0105] Calculate the bidding density b_{d_j} and the utility U_i of task i

[0106] end if

[0107] end for

[0108] / / Multi-objective optimization, select the channel j* that maximizes utility and satisfies security

[0109] j*=argmax_{j}U_i, where U_i is the comprehensive utility function considering objectives

[0110] if P_{i,j*}>threshold and security_{i,j*}>=threshold:

[0111] psi_{i,j*}=1 / / User i updates Load_history by unloading tasks through channel j*

[0112] Dynamically adjust π_{j*} according to Load_history / / Dynamically adjust the threshold value of channel j* according to historical load

[0113] else:

[0114] psi_{i,j*}=0 / / User i performs tasks locally

[0115] end if

[0116] Remove user i from U'o

[0117] end while

[0118] Return to psi

[0119] The embodiment of the present invention integrates the two parts of the task priority prediction model based on machine learning and the adaptive multi-objective optimization task offloading algorithm. The intelligent adaptive dynamic resource allocation algorithm aims to provide an efficient and intelligent task processing mechanism for the instant messaging system in the edge cloud environment. By integrating the prediction model based on machine learning, the algorithm can deeply understand and predict the characteristics and user behavior of instant messaging tasks, set reasonable priorities for different tasks, and ensure that critical and urgent tasks are responded to quickly and reliably. At the same time, the algorithm integrates the adaptive multi-objective optimization task offloading algorithm, which not only guarantees the fairness and efficiency of the task offloading process, but also can dynamically adjust the resource allocation decision according to the real-time network status and changes in device mobility, and enhance the robustness and stability of the system. In addition, by implementing the message priority processing strategy, the algorithm effectively manages different types of messages in instant messaging, ensures the priority transmission and processing of urgent and important messages, and thus provides users with more efficient, secure and reliable instant messaging services in the mobile edge computing environment. In general, the intelligent adaptive dynamic resource allocation algorithm solves the technical shortcomings of the existing technical solutions and meets the high standards of users for instant messaging services.

[0120] like Figure 4 As shown, in order to achieve the above-mentioned purpose, an embodiment of the present invention further provides a task offloading device, which includes:

[0121] A first acquisition module, used to acquire a first task with the highest priority among a plurality of tasks to be uninstalled sorted by priority; wherein the priority is acquired according to the task characteristics, user behavior patterns and network conditions of the tasks to be uninstalled;

[0122] A first determining module, configured to determine a channel with the smallest bidding density among the multiple channels as a first channel;

[0123] A second acquisition module is used to acquire a first unloading probability of the first channel; wherein the first unloading probability is a probability of unloading the first task on the first channel among the multiple channels;

[0124] The second determining module is configured to determine that the first task is to be offloaded through the first channel when the first offloading probability is greater than a first threshold.

[0125] Optionally, the device further comprises:

[0126] A first receiving module, used for receiving multiple instant messages from multiple users;

[0127] A third acquisition module is used to classify and mark the multiple instant messages to obtain the multiple tasks to be uninstalled;

[0128] A fourth acquisition module is used to input the task characteristics of the multiple tasks to be offloaded, the user behavior pattern and the network status into a priority prediction model to obtain a prediction result; wherein the priority prediction model is obtained by training based on historical task offloading data;

[0129] The third determination module is used to determine the priorities of the multiple tasks to be offloaded according to the prediction result.

[0130] Optionally, the device further comprises:

[0131] The first processing module is configured to, when it is determined that a threshold value of one of the multiple channels is less than or equal to a second threshold value, remove the corresponding channel from the multiple channels.

[0132] Optionally, the device further comprises:

[0133] A second processing module is used to update the historical load; the historical load is a record list used to record the unloading of the first task through the first channel;

[0134] The third processing module is configured to adjust the threshold value of the first channel according to the historical load.

[0135] Optionally, the device further comprises:

[0136] A fourth processing module, configured to execute the first task on the server where the first task is currently located;

[0137] A fifth processing module is configured to remove the first task from the multiple tasks to be uninstalled.

[0138] Optionally, the device further comprises:

[0139] a fourth determining module, configured to determine a security assessment value of the first channel;

[0140] A fifth acquisition module, used to determine whether the security assessment value is greater than or equal to a third threshold, and obtain a second determination result;

[0141] The sixth processing module is used to determine whether the first uninstallation probability is greater than the first threshold when the second judgment result is yes.

[0142] Optionally, the device further comprises:

[0143] A seventh processing module is configured to remove the first channel from the multiple channels when the second judgment result is no.

[0144] It should be noted here that the above-mentioned device provided in the embodiment of the present invention can implement all the method steps implemented in the above-mentioned method embodiment, and can achieve the same technical effect. The parts and beneficial effects that are the same as the method embodiment in this embodiment will not be described in detail here.

[0145] In order to achieve the above-mentioned purpose, an embodiment of the present invention also provides a network device, including: a transceiver, a processor, a memory, and a program or instruction stored in the memory and executable on the processor; wherein, when the processor executes the program or instruction, the task offloading method as described above is implemented.

[0146] In order to achieve the above-mentioned purpose, an embodiment of the present invention further provides a readable storage medium on which a program or instruction is stored, wherein the program or instruction, when executed by a processor, implements the steps in the task offloading method as described above.

[0147] It should be further explained that the terminals described in this specification include but are not limited to smart phones, tablet computers, etc., and many of the functional components described are called modules in order to more particularly emphasize the independence of their implementation methods.

[0148] In the embodiment of the present invention, module can be implemented with software so that it can be executed by various types of processors. For example, an executable code module of an identification can include one or more physical or logical blocks of computer instructions, for example, it can be constructed as an object, process or function. Nevertheless, the executable code of the identified module does not need to be physically located together, but can include different instructions stored in different positions, and when these instructions are logically combined together, it constitutes a module and realizes the specified purpose of the module.

[0149] In fact, executable code module can be a single instruction or many instructions, and can even be distributed on a plurality of different code segments, distributed among different programs, and distributed across a plurality of memory devices. Similarly, operating data can be identified in the module, and can be implemented and organized in the data structure of any appropriate type according to any appropriate form. The operating data can be collected as a single data set, or can be distributed in different locations (including on different storage devices), and can only be present on a system or network as an electronic signal at least in part.

[0150] When a module can be implemented by software, considering the level of existing hardware technology, a person skilled in the art can build a corresponding hardware circuit to implement the corresponding function of the module that can be implemented by software without considering the cost. The hardware circuit includes a conventional very large scale integration (VLSI) circuit or gate array and existing semiconductors such as logic chips, transistors, or other discrete components. The module can also be implemented by a programmable hardware device, such as a field programmable gate array, a programmable array logic, a programmable logic device, etc.

[0151] In order to achieve the above object, an embodiment of the present invention further provides a computer program product, which includes computer instructions, and when the computer instructions are executed by a processor, the steps of the task offloading method described above are implemented.

[0152] The above exemplary embodiments are described with reference to the accompanying drawings, and many different forms and embodiments are feasible without departing from the spirit and teachings of the present invention. Therefore, the present invention should not be constructed as a limitation of the exemplary embodiments proposed herein. More specifically, these exemplary embodiments are provided so that the present invention will be perfect and complete, and the scope of the present invention will be conveyed to those who are familiar with the technology. In these figures, the component sizes and relative sizes may be exaggerated for clarity. The terms used here are only based on the purpose of describing specific exemplary embodiments and are not intended to be limiting. As used herein, unless the text clearly indicates otherwise, the singular forms "one", "an" and "the" are intended to include these multiple forms. It will be further understood that the terms "including" and / or "comprising" when used in this specification indicate the presence of the features, integers, steps, operations, components and / or components, but do not exclude the presence or increase of one or more other features, integers, steps, operations, components, components and / or their groups. Unless otherwise indicated, when stated, a range of values ​​includes the upper and lower limits of that range and any subranges therebetween.

[0153] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A task offloading method, characterized in that: include: Acquire a first task with the highest priority among a plurality of tasks to be uninstalled that are sorted by priority; wherein the priority is acquired according to the task characteristics, user behavior patterns, and network conditions of the tasks to be uninstalled; Determine a channel with the smallest bidding density among the multiple channels as a first channel; Obtaining a first unloading probability of the first channel; wherein the first unloading probability is a probability of unloading the first task on the first channel among the multiple channels; When the first offloading probability is greater than a first threshold, it is determined that the first task is offloaded through the first channel.

2. The method according to claim 1, characterized in that Before obtaining a first task with the highest priority among the plurality of tasks to be offloaded sorted by priority, the method further includes: receiving multiple instant messages from multiple users; Classifying and marking the multiple instant messages to obtain the multiple tasks to be offloaded; Inputting the task characteristics of the multiple tasks to be offloaded, the user behavior pattern and the network status into a priority prediction model to obtain a prediction result; wherein the priority prediction model is obtained by training based on historical task offloading data; The priorities of the multiple tasks to be offloaded are determined according to the prediction results.

3. The method according to claim 1, characterized in that Before determining the channel with the smallest bidding density among the multiple channels as the first channel, the method further includes: When it is determined that the threshold value of one of the multiple channels is less than or equal to the second threshold value, the corresponding channel is removed from the multiple channels.

4. The method according to claim 3, characterized in that After determining that the first task is to be unloaded through the first channel, the method further includes: Updating the historical load; the historical load is a record list for recording the unloading of the first task through the first channel; The threshold value of the first channel is adjusted according to the historical load.

5. The method according to claim 1, characterized in that After determining that the first task is to be unloaded through the first channel, the method further includes: Execute the first task on the server where the first task is currently located; The first task is removed from the multiple tasks to be uninstalled.

6. The method according to claim 1, characterized in that Before determining that the first task is to be unloaded through the first channel, the method further includes: determining a security assessment value of the first channel; Determine whether the security assessment value is greater than or equal to a third threshold value, and obtain a second determination result; When the second determination result is yes, it is determined whether the first uninstallation probability is greater than the first threshold.

7. The method according to claim 6, characterized in that The method further comprises: When the second judgment result is no, the first channel is removed from the plurality of channels.

8. A task offloading device, characterized in that: include: A first acquisition module, used to acquire a first task with the highest priority among a plurality of tasks to be uninstalled sorted by priority; wherein the priority is acquired according to the task characteristics, user behavior patterns and network conditions of the tasks to be uninstalled; A first determining module, configured to determine a channel with the smallest bidding density among the multiple channels as a first channel; A second acquisition module is used to acquire a first unloading probability of the first channel; wherein the first unloading probability is a probability of unloading the first task on the first channel among the multiple channels; The second determining module is configured to determine that the first task is to be offloaded through the first channel when the first offloading probability is greater than a first threshold.

9. A network device comprising: A transceiver, a processor, a memory, and a program or instruction stored in the memory and executable on the processor; characterized in that when the processor executes the program or instruction, the task offloading method according to any one of claims 1 to 7 is implemented.

10. A readable storage medium having a program or instruction stored thereon, characterized in that: When the program or instruction is executed by the processor, the steps in the task offloading method according to any one of claims 1 to 7 are implemented.

11. A computer program product, characterized in that The method comprises computer instructions, which, when executed by a processor, implement the steps of the task offloading method according to any one of claims 1 to 7.

Images