A swarm intelligent control method for multimedia edge-cloud system

Abstract

The present invention discloses a swarm intelligent control method for a multimedia end-edge-cloud system, the method comprising: obtaining attribute information of task requests in a heterogeneous network; prioritizing tasks according to the attribute information of task requests to obtain a plurality of priority levels; calculating the total energy consumption value of tasks and the total response time value of tasks; wherein one priority level corresponds to a plurality of tasks; optimizing the deployment of a combination of terminals, edge servers and cloud servers in a multimedia end-edge-cloud system according to the attribute information of task requests, the total energy consumption value and the total response time value, and obtaining a target deployment scheme of the multimedia end-edge-cloud system, so that the total energy consumption value and the total response time value corresponding to the target deployment scheme are minimized. The embodiment of the present invention deploys a combination scheme of each node in a multimedia end-edge-cloud system so that each node can allocate efficient energy consumption requirements and lower response delay requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of network resource allocation, and in particular to a group intelligent control method for a multimedia end-edge-cloud system. Background Art

[0002] With the continuous development of smart cities, mobile video, mobile social network and 3D video have brought highly complex traffic patterns to urban governance. In particular, the rapid growth of multimedia system data has brought great challenges to data transmission and processing. It requires powerful computing resources and network resources to ensure the normal operation of the system. In recent years, with the continuous analysis of end-edge cloud, complex multimedia end-edge cloud system has realized the intelligent deployment and dynamic adjustment of smart city nodes. Through real-time monitoring and management of users, vehicles, mobile phones, camera layout location, damage and other data characteristics in the system, the multimedia end-edge cloud system can extract features from terminal data information and generate edge node deployment plans based on this information. At the same time, the generated layout strategy is adaptable to changes in terminal users, realizing intelligent deployment. Compared with the original manual deployment, efficient management of large-scale heterogeneous dynamic access terminal devices can be achieved on the edge nodes, especially for the situation where the dynamic increase of terminal devices leads to a large amount of burst network traffic, dynamic management of edge nodes can be performed to achieve reasonable allocation of resources. However, the existing group intelligent control method cannot ensure that all terminals, edge servers and cloud servers can allocate efficient energy consumption requirements and low response delay requirements.

[0003] Therefore, the existing technology still needs to be improved and developed. Summary of the invention

[0004] The technical problem to be solved by the present invention is that, in response to the above-mentioned defects of the prior art, a swarm intelligent control method for a multimedia terminal-edge-cloud system is provided, aiming to solve the problem that the swarm intelligent control method in the prior art cannot ensure that all terminals, edge servers and cloud servers can allocate efficient energy consumption requirements and lower response delay requirements.

[0005] The technical solution adopted by the present invention to solve the problem is as follows:

[0006] In a first aspect, an embodiment of the present invention provides a swarm intelligent control method of a multimedia end-edge-cloud system, characterized in that the method includes:

[0007] Acquire attribute information of a task request in a heterogeneous network; wherein the task request is used to represent a request instruction for a terminal to send a task to an edge server;

[0008] According to the attribute information of the task request, the tasks are prioritized to obtain several priority levels, and the total energy consumption value of the tasks and the total response time value of the tasks are calculated; wherein one priority level corresponds to multiple tasks;

[0009] According to the attribute information of the task request, the total energy consumption value and the total response time value, the combination of the terminal, edge server and cloud server in the multimedia end-edge-cloud system is optimized and deployed to obtain the target deployment scheme of the multimedia end-edge-cloud system so that the total energy consumption value and the total response time value corresponding to the target deployment scheme are minimized.

[0010] In one implementation, the number of tasks is several, each task includes several subtasks; the number of terminals is several; and the calculating of the total energy consumption value of the task and the total response time value of the task includes:

[0011] For each subtask in each terminal, calculating a first energy consumption value of each subtask in each terminal and a first response time value of each subtask;

[0012] According to the first energy consumption value of each subtask and the first response time value of each subtask in each terminal, the total energy consumption value of the task and the total response time value of the task are obtained.

[0013] In one implementation, for each subtask in each terminal, calculating the first energy consumption value of each subtask in each terminal and the first response time value of each subtask in each terminal includes:

[0014] For each terminal, calculating a first initial energy consumption value and a first initial response time value when each terminal executes each subtask;

[0015] Calculating a second initial energy consumption value and a second initial response time value when executing each subtask on a path from each terminal to the edge server;

[0016] Calculate a third initial energy consumption value and a third initial response time value when executing each subtask on a path from each terminal via the edge server to the cloud server;

[0017] Adding the first initial energy consumption value, the second initial energy consumption value, and the third initial energy consumption value to obtain a first energy consumption value for each subtask in each terminal;

[0018] The maximum value among the first initial response time value, the second initial response time value and the third initial response time value is obtained to obtain the first response time value of each subtask in each terminal.

[0019] In one implementation, obtaining the total energy consumption value and the total response time value of the task according to the first energy consumption value and the first response time value of each subtask in each terminal includes:

[0020] Adding the first energy consumption values ​​of all subtasks in each terminal to obtain the second energy consumption value of each task in each terminal;

[0021] Obtain the maximum value of the first response time of all subtasks in each terminal, and obtain the second response time value of each task in each terminal;

[0022] According to the second energy consumption value of each task in each terminal and the second response time value of each task in each terminal, the total energy consumption value of the task and the total response time value of the task are obtained.

[0023] In one implementation, obtaining the total energy consumption value of the task and the total response time value of the task according to the second energy consumption value of each task in each terminal and the second response time value of each task in each terminal includes:

[0024] Adding the second energy consumption values ​​of all tasks in each terminal to obtain a third energy consumption value in each terminal;

[0025] Add the third energy consumption values ​​of all terminals to obtain the total energy consumption value of the multimedia end-edge-cloud system;

[0026] Obtaining the maximum value of the second response time of each task in all terminals, and obtaining the third response time value of each task in all terminals;

[0027] Add the third response time value of each task in all terminals to obtain the total response time value of the multimedia end-edge-cloud system.

[0028] In one implementation, the combination of the terminal, the edge server, and the cloud server in the multimedia end-edge-cloud system is optimized and deployed according to the attribute information of the task request, the total energy consumption value, and the total response time value, and the target deployment scheme of the multimedia end-edge-cloud system is obtained, including:

[0029] Based on a preset first rule, and according to the attribute information of the task request, a node is allocated to each task to obtain a plurality of nodes corresponding to each task; wherein the node is used to represent a connection point of a terminal, an edge server or a cloud server;

[0030] When the node is an edge server or a cloud server, the computing capacity of the node is obtained, wherein the computing capacity is used to characterize the computing processing capacity of the server; based on a preset second rule, each node is grouped according to the computing capacity of the node to obtain two subset groups; according to the two subset groups, a number of the nodes are updated to obtain a number of updated nodes;

[0031] According to several priority levels, several updated nodes, the total energy consumption value and the total response time value, the combination of terminals, edge servers and cloud servers in the multimedia end-edge-cloud system is optimized and deployed to obtain a target deployment plan for the multimedia end-edge-cloud system.

[0032] In one implementation, the combination of terminals, edge servers, and cloud servers in the multimedia end-edge-cloud system is optimized and deployed according to the priority levels, the updated nodes, the total energy consumption value, and the total response time value, and the target deployment scheme of the multimedia end-edge-cloud system is obtained, including:

[0033] Get the transmission mode of several updated nodes;

[0034] Obtaining a total energy consumption value corresponding to the transmission mode and a total response time value corresponding to the transmission mode;

[0035] For each priority level, according to the node, the transmission mode corresponding to the node, the total energy consumption value corresponding to the transmission mode and the total response time value corresponding to the transmission mode, the combination of terminals, edge servers and cloud servers in the multimedia end-edge-cloud system is optimized and deployed to obtain the target deployment plan of the multimedia end-edge-cloud system.

[0036] In one implementation, the combination of the terminal, the edge server, and the cloud server in the multimedia end-edge-cloud system is optimized and deployed according to the node, the transmission mode corresponding to the node, the total energy consumption value corresponding to the transmission mode, and the total response time value corresponding to the transmission mode, and the target deployment scheme of the multimedia end-edge-cloud system is obtained, including:

[0037] Classify the transmission mode whose total energy consumption value is greater than the preset energy consumption threshold into the disposable solution set, and record the total energy consumption value in the disposable solution set; classify the transmission mode whose total energy consumption value is less than or equal to the preset energy consumption threshold into the non-disposable solution set; wherein the non-disposable solution set is the optimal solution set; the disposable solution set is the set other than the optimal solution set; classify the transmission mode whose total response time value is less than the preset response time threshold into the non-disposable solution set, and record the total response time value in the non-disposable solution set; classify the transmission mode whose total response time value is greater than or equal to the preset response time threshold into the disposable solution set; take the nodes in the disposable solution set and the transmission mode corresponding to the nodes as the first initial solution; take the nodes in the non-disposable solution set and the transmission mode corresponding to the nodes as the second initial solution;

[0038] sorting the first initial solutions in the set of dominable solutions;

[0039] Based on the congestion strategy, the first initial solution in the set of disposable solutions is weighted so that the difference between the weight of the first initial solution ranked first and the weight of the first initial solution ranked last is greater than a preset threshold;

[0040] Based on the optimal learning strategy and the suboptimal learning strategy, the weighted first initial solution is updated and ranked by performance to obtain an updated first initial solution;

[0041] Based on a preset elimination strategy, the updated first initial plan is eliminated to obtain an intermediate deployment plan;

[0042] Copy the intermediate deployment plan to obtain the target deployment plan;

[0043] Rank the second initial solutions in the non-dominant solution set by performance, and obtain the second initial solution ranked first;

[0044] When the total energy consumption value in the first-ranked second initial solution is smaller than the total energy consumption value in the target deployment solution and the total response time value in the first-ranked second initial solution is smaller than the total response time value in the target deployment solution, the first-ranked second initial solution is used as the target deployment solution;

[0045] Repeat the steps of weighting the first initial solution in the set of available solutions based on the congestion strategy so that the difference between the weight of the first initial solution ranked first and the weight of the first initial solution ranked last is greater than a preset threshold until the optimal target deployment solution is obtained.

[0046] In one implementation, the method further includes:

[0047] When the attribute information of the task request is updated, the priority sorting of the task is re-executed according to the updated attribute information of the task request to obtain several priority levels; the total energy consumption value of the task and the total response time value of the task are calculated; according to the attribute information of the task request, the total energy consumption value and the total response time value, the combination of the terminal, the edge server and the cloud server in the multimedia end-edge-cloud system is optimized and deployed to obtain the target deployment scheme of the multimedia end-edge-cloud system, so as to minimize the total energy consumption value and the total response time value corresponding to the target deployment scheme.

[0048] In a second aspect, an embodiment of the present invention further provides a swarm intelligent control device for a multimedia end-edge-cloud system, wherein the device comprises: a task request attribute information acquisition module, used to acquire attribute information of a task request in a heterogeneous network; wherein the task request is used to represent a request instruction for a terminal to send a task to an edge server;

[0049] A total energy consumption value and total response time value calculation module, used to prioritize the tasks according to the attribute information of the task request, obtain a number of priority levels, and calculate the total energy consumption value of the tasks and the total response time value of the tasks; wherein one priority level corresponds to a plurality of tasks;

[0050] The scheme optimization deployment module is used to optimize the deployment of the combination of terminals, edge servers and cloud servers in the multimedia end-edge-cloud system according to the attribute information of the task request, the total energy consumption value and the total response time value, and obtain the target deployment scheme of the multimedia end-edge-cloud system, so as to minimize the total energy consumption value and the total response time value corresponding to the target deployment scheme.

[0051] In the third aspect, an embodiment of the present invention also provides an intelligent terminal, comprising a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors, and the one or more programs include a group intelligent control method for executing a multimedia end-edge cloud system as described in any one of the above items.

[0052] In a fourth aspect, an embodiment of the present invention also provides a non-temporary computer-readable storage medium. When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device can execute a swarm intelligent control method of a multimedia end-edge cloud system as described in any one of the above.

[0053] Beneficial effects of the present invention: The embodiment of the present invention first obtains the attribute information of the task request in the heterogeneous network; wherein the task request is used to characterize the request instruction of the terminal to send the task to the edge server; then, according to the attribute information of the task request, the task is prioritized to obtain several priority levels, and the total energy consumption value of the task and the total response time value of the task are calculated; wherein one priority level corresponds to multiple tasks; finally, according to the attribute information of the task request, the total energy consumption value and the total response time value, the combination scheme of the terminal, edge server and cloud server in the multimedia end-edge cloud system is optimized and deployed to obtain the target deployment scheme of the multimedia end-edge cloud system, so that the total energy consumption value and the total response time value corresponding to the target deployment scheme are minimized; it can be seen that in the embodiment of the present invention, a combination scheme is deployed for each node in the multimedia end-edge cloud system so that each node can allocate efficient energy consumption requirements and lower response delay requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1 A flow chart of a swarm intelligent control method for a multimedia end-edge-cloud system provided in an embodiment of the present invention.

[0056] Figure 2 A structural diagram of the node transmission method of the end-edge cloud system provided in an embodiment of the present invention.

[0057] Figure 3 This is a diagram of the implementation framework of the swarm intelligent control method provided in an embodiment of the present invention.

[0058] Figure 4 A framework diagram of the edge node autonomous combination optimization deployment solution provided in an embodiment of the present invention.

[0059] Figure 5 A functional block diagram of a swarm intelligent control device for a multimedia end-edge-cloud system provided in an embodiment of the present invention.

[0060] Figure 6 This is a block diagram of the internal structure principle of the intelligent terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0061] The present invention discloses a swarm intelligent control method for a multimedia end-edge cloud system. In order to make the purpose, technical solution and effect of the present invention clearer and more specific, the present invention is further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described here are only used to explain the present invention and are not used to limit the present invention.

[0062] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0063] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless specifically defined as herein.

[0064] Due to the method in the prior art: an edge network control system including a cloud platform, a heterogeneous network, an edge control system, an edge control subsystem and an intelligent terminal, in heterogeneous data processing, supports multiple links, realizes multi-channel heterogeneous data transmission through autonomous switching of heterogeneous networks, and in task scheduling, sets a priority level for the communication mode and prioritizes the services according to the delay requirements, and proposes an improved static priority multi-task scheduling method, which gives the device's task unloading strategy and the end-edge cloud resource allocation strategy according to the task delay and energy consumption, so as to achieve the purpose of concurrent operation and delay of the service and reduce energy consumption. The defect of this method is that the method itself only targets heterogeneous, multi-channel terminal data, but ignores the limited resources that can be provided by the cloud server and the edge server, and the cooperation and competition relationship between the servers themselves. Especially in the current multimedia transmission, when wireless devices and computing resource allocation affect the data transmission rate, device energy consumption and task delay respectively, this method cannot give specific requirements for delay and energy consumption that are suitable for actual life.

[0065] In order to solve the problems of the prior art, this embodiment provides a group intelligent control method for a multimedia end-edge cloud system, which deploys a combination scheme for each node in the multimedia end-edge cloud system so that each node can allocate efficient energy consumption requirements and low response delay requirements. In specific implementation, first obtain the attribute information of the task request in the heterogeneous network; wherein the task request is used to characterize the request instruction of the terminal to send the task to the edge server; then, according to the attribute information of the task request, prioritize the tasks, obtain several priority levels, and calculate the total energy consumption value of the task and the total response time value of the task; wherein one priority level corresponds to multiple tasks; finally, according to the attribute information of the task request, the total energy consumption value and the total response time value, optimize the combination scheme of the terminal, edge server and cloud server in the multimedia end-edge cloud system and obtain the target deployment scheme of the multimedia end-edge cloud system, so that the total energy consumption value and the total response time value corresponding to the target deployment scheme are minimized.

[0066] Exemplary Methods

[0067] This embodiment provides a group intelligent control method for a multimedia end-edge-cloud system, which can be applied to intelligent terminals for network resource allocation. Figure 1 As shown, the method includes:

[0068] Step S100, obtaining attribute information of a task request in a heterogeneous network; wherein the task request is used to represent a request instruction for a terminal to send a task to an edge server;

[0069] Specifically, heterogeneous networks such as Figure 2 As shown, it is a network composed of terminals and servers produced by different manufacturers; the task request is used to represent the request instruction of the terminal to send a task to the edge server; the attribute information of the task request includes the delay requirement of the task transmission, the sequence number of the process sequence of the task and the resource requirement of the task, such as the arrival time of the task, the process sequence number of the task, and the tolerable bandwidth limit and computing power limit of the task.

[0070] After obtaining the attribute information of the task request in the heterogeneous network, we can execute Figure 1 The following steps are shown: S200, according to the attribute information of the task request, the tasks are prioritized to obtain several priority levels, and the total energy consumption value of the tasks and the total response time value of the tasks are calculated; wherein one priority level corresponds to multiple tasks;

[0071] Specifically, according to the urgency of the delay, the sequence number of the task in the process sequence, and the degree of bandwidth and resource demand, the balance weight is set to prioritize the tasks. The more urgent the delay demand, the smaller the flow number, and the higher the resource demand, the higher the priority. In this way, K division levels are given: ζ 1 ,ζ 2 ,...,ζ K ,A priority level contains multiple tasks, and there are divisible and indivisible tasks among the tasks. Moreover, multiple tasks in a priority level have parallel computing properties. In practice, multiple tasks in a priority level are input into the multimedia edge cloud system in the order of high priority, second highest priority, ... low priority.

[0072] The method involved in the present invention needs to be performed between multiple edge servers and multiple cloud servers, and there is a topological structure between servers, such as Figure 2 As shown in the figure, each task can be performed on the terminal, edge server and cloud server respectively. There is more than one cloud server, and there is a competition and cooperation relationship between the cloud servers. For devices that cannot be unloaded locally, such as cameras, they can only be unloaded on edge servers or cloud servers. Considering the distance between the terminal and the cloud server, the unloading method from the terminal directly to the cloud server is not considered here; Figure 3As shown, the deployment of dynamic management of edge nodes in a complex end-edge-cloud system needs to consider the energy consumption and response time of executing tasks, and the design of minimizing energy consumption and the shortest response time are the two major goals of the end-edge-cloud system evaluation model; the data volume, available computing power, consumed computing power and energy consumption of the terminal task, as well as the available bandwidth, transmission bandwidth, available computing power, consumed computing power and energy consumption parameters of the cloud server are transmitted to the group intelligent control of the complex multimedia end-edge-cloud system at the edge node, and the autonomous deployment of the edge node and the task issuance are implemented; this is a multi-objective problem based on energy consumption and response time. The present invention needs to give the task deployment strategy of the device and the resource allocation strategy in the end-edge-cloud node according to the delay and energy consumption of the task. The resource allocation strategy has two goals: Goal one: minimizing energy consumption. Minimizing energy consumption means that the sum of the energy consumed by the task execution processing on the terminal, edge server and cloud server, as well as the overall transmission process and server migration process is minimized; Goal two: minimizing response time. Minimizing response time means that the sum of the execution time of the task on the terminal, edge server and cloud server, as well as the overall transmission time, server migration time and task queuing waiting time is minimized. To this end, the total energy consumption value of the task and the total response time value of the task must be calculated first. In this embodiment, the number of tasks is N, indexed by i=1,2,...,N; the number of mobile terminals is U, indexed by u=1,2,...U, and each terminal may have multiple task requests at the same time. Tasks are only executed on local terminals, edges, and cloud nodes, and because the delay and energy consumption of task request returns are very small compared to the request, they are ignored here; while the task is executed on the server, there is a task migration process. Each task is only migrated once between edges and once between clouds, and cannot be returned. The number of edge servers is S, indexed by s=1,2,...S; the number of cloud servers is M, indexed by m=1,2,...M. Here, tasks are divided into divisible and indivisible. For tasks that can be divided, for each task i, there is a corresponding set of J matching subtasks j = {1, 2, ... J}. Assuming that the i-th task can be divided into 3 subtasks, then J = 3, and the subtask set is {(i, 1), (i, 2), (i, 3)}. The subtasks can be calculated in parallel. For tasks that cannot be divided, it means J = 1.

[0073] In step S200, the number of tasks is several, and each task includes several subtasks; the number of terminals is several; the calculation of the total energy consumption value of the task and the total response time value of the task includes the following steps: for each subtask in each terminal, calculate the first energy consumption value of each subtask in each terminal and the first response time value of each subtask; according to the first energy consumption value of each subtask in each terminal and the first response time value of each subtask, obtain the total energy consumption value of the task and the total response time value of the task.

[0074] In one implementation, for each subtask in each terminal, calculating the first energy consumption value of each subtask in each terminal and the first response time value of each subtask in each terminal includes the following steps: for each terminal, calculating the first initial energy consumption value and the first initial response time value when each terminal executes each subtask; calculating the second initial energy consumption value and the second initial response time value when each subtask is executed on the path from each terminal to the edge server; calculating the third initial energy consumption value and the third initial response time value when each subtask is executed on the path from each terminal via the edge server to the cloud server; adding the first initial energy consumption value, the second initial energy consumption value and the third initial energy consumption value to obtain the first energy consumption value of each subtask in each terminal; obtaining the maximum value of the first initial response time value, the second initial response time value and the third initial response time value to obtain the first response time value of each subtask in each terminal.

[0075] Specifically, for each terminal u, the first initial energy consumption value when each terminal executes each subtask is calculated by formula (1), and the first initial response time value when each terminal executes each subtask is calculated by formula (2):

[0076]

[0077]

[0078] in, represents the energy consumption of executing the jth subtask of the i-th task on terminal u; represents the response time of executing the jth subtask of the i-th task on terminal u; C(i,j) is the number of CPU resources required to execute the jth subtask of the i-th task; is the computing service capability of terminal u; is the computing power consumption per unit time of terminal u.

[0079] The second initial energy consumption value when executing each subtask on the path from each terminal to the edge server is calculated by formula (3); the second initial response time value when executing each subtask on the path from each terminal to the edge server is calculated by formula (4):

[0080]

[0081]

[0082]

[0083] Among them, the first part is the transmission time of the task from the terminal to the first edge server w, the second part is the transmission time of migrating from edge server w to the adjacent edge server s, and the third part is the time spent on the execution of the jth subtask of the i-th task on the edge server s; represents the sum of the energy consumption of executing the jth subtask of the i-th task from the terminal u to the edge server s; represents the overall response time from terminal u to edge server s for executing the jth subtask of the i-th task; is the computing energy consumption per unit time of the terminal u transmitting to the edge server s; is the computing energy consumption per unit time of terminal u transmitting to edge server w; is the energy consumption per unit time of migrating from edge server w to edge server s. If w = s, it means that it is directly executed on edge server s without migration. Fixed parameters are introduced at this time As shown in formula (5); is the transmission bandwidth from terminal u to edge server s; is the transmission bandwidth from terminal u to edge server w; is the transmission bandwidth from edge server w to edge server s. If w = s, is the uplink bandwidth transmitted from terminal u to edge server s; It is the energy consumption per unit time of edge server s to calculate and execute tasks; is the computing service capacity of the edge server s; D(i,j) represents the data size of the jth subtask of the i-th task transmitted by the terminal to be executed on the edge server; represents the waiting time of the jth subtask of the i-th task on the edge server s due to other processes.

[0084] The third initial energy consumption value when each subtask is executed on the path from each terminal via the edge server to the cloud server is calculated by formula (6); the third initial response time value when each subtask is executed on the path from each terminal via the edge server to the cloud server is calculated by formula (7):

[0085]

[0086]

[0087]

[0088] The first part is the transmission time of the task from terminal u to edge server s; the second part is the transmission time from edge server s to cloud server z; the third part is the transmission time from cloud server z to the adjacent cloud server m; the fourth part is the time required for the jth subtask of the i-th task to be executed on cloud server m; represents the sum of the energy consumption of executing the jth subtask of the i-th task from the terminal u through the edge server s to the cloud server m; represents the overall response time of executing the jth subtask of the i-th task from the terminal u through the edge server s to the cloud server m; The energy consumed per unit time by edge server s to cloud server z; is the energy consumption per unit time of migration from cloud server z to cloud server m. If z = m, Fixed parameters are introduced at this time As shown in formula (8); is the uplink bandwidth from edge server s to cloud server z; is the transmission bandwidth of cloud server z migrating to cloud server m. If z = m, is the transmission bandwidth from edge server s to cloud server m, is the energy consumption per unit time of cloud server m to calculate and execute tasks; is the computing service capability of cloud server m; represents the waiting time of the jth subtask of the ith task on cloud server m due to other processes.

[0089] Then, the first initial energy consumption value, the second initial energy consumption value and the third initial energy consumption value are added to obtain the first energy consumption value of each subtask in each terminal; the maximum value of the first initial response time value, the second initial response time value and the third initial response time value is obtained to obtain the first response time value of each subtask in each terminal; then the first energy consumption values ​​of all subtasks in each terminal are added to obtain the second energy consumption value of each task in each terminal; the maximum value of the first response time values ​​of all subtasks in each terminal is obtained to obtain the second response time value of each task in each terminal; the second energy consumption values ​​of all tasks in each terminal are added to obtain the third energy consumption value in each terminal; the third energy consumption values ​​of all terminals are added to obtain the total energy consumption value of the multimedia terminal-edge-cloud system. In order to minimize the total energy consumption value, as shown in formula (9), the maximum value of the second response time value of each task in all terminals is obtained, that is, the third response time value of each task in all terminals is obtained as follows:

[0090]

[0091] Where T(i) is the maximum response time of all tasks at each priority level. The third response time value of each task in all terminals is added together to obtain the total response time value of the multimedia end-edge-cloud system. In order to minimize the total response time value, as shown in formula (10):

[0092]

[0093]

[0094] Constraints:

[0095]

[0096]

[0097]

[0098] in, Indicates that the jth subtask of the i-th task is not executed locally. Indicates that the jth subtask of the i-th task is executed locally; there are multiple subtasks executed in parallel on the same terminal, and the total number of resources is less than the computing capacity of the terminal, that is, formula (12); Indicates whether the jth subtask of the i-th task is executed by edge server s; when the jth subtask of the i-th task is executed on edge server s Satisfied at this time otherwise There are multiple subtasks from different terminals that are executed in parallel on the same edge server, and the total number of resources is less than the computing power of the terminal, that is, formula (13); A parameter indicating whether the jth subtask of the i-th task is executed by cloud server m; when the jth subtask of the i-th task is executed on cloud server m Satisfied at this time otherwise There are multiple subtasks from different edge servers executed in parallel on the same cloud server, and the total number of resources is less than the computing power of the terminal, that is, formula (14).

[0099] After obtaining the total energy consumption value of the task and the total response time value of the task, the following can be performed: Figure 1 The following steps are shown: S300, according to the attribute information of the task request, the total energy consumption value and the total response time value, the combination of the terminal, edge server and cloud server in the multimedia end-edge-cloud system is optimized and deployed to obtain the target deployment scheme of the multimedia end-edge-cloud system, so that the total energy consumption value and the total response time value corresponding to the target deployment scheme are minimized.

[0100] Step S300 specifically includes the following steps:

[0101] S301, based on a preset first rule, and according to the attribute information of the task request, a node is allocated to each task to obtain a plurality of nodes corresponding to each task; wherein the node is used to represent a connection point of a terminal, an edge server or a cloud server;

[0102] S302: when the node is an edge server or a cloud server, obtaining the computing capacity of the node, wherein the computing capacity is used to characterize the computing processing capacity of the server; based on a preset second rule, grouping each node according to the computing capacity of the node to obtain two subset groups; updating a plurality of the nodes according to the two subset groups to obtain a plurality of updated nodes;

[0103] S303. According to a number of priority levels, a number of updated nodes, the total energy consumption value and the total response time value, the combination of terminals, edge servers and cloud servers in the multimedia end-edge-cloud system is optimized and deployed to obtain a target deployment plan for the multimedia end-edge-cloud system.

[0104] Before step S301, initialization is performed first, such as initializing the attribute information of the terminal, edge server, and cloud server, as well as the maximum number of individuals in the population and the maximum number of iterations. The maximum number of individuals in the population (such as 40) and the maximum number of iterations (such as 4000) need to be determined through parameter analysis later, and initialization only gives an estimated value based on past experience. The first preset rule is: eliminate edge servers and cloud servers that are far away from the terminal; eliminate edge servers and cloud servers with weak computing power, such as edge servers and cloud servers whose computing power is less than the preset computing power threshold. According to the priority level obtained based on the attributes of the task (the delay requirement of task transmission, the sequence number of the process sequence in which the task is located, and the resource requirement of the task), each task is assigned a terminal, edge, or cloud node in order of priority level from high to low to reduce the solution space.

[0105] In step S302, when the node is an edge server or a cloud server, the computing power of the node is obtained, and the preset second rule is that the nodes (edge ​​servers or cloud servers) in the subset have sufficient computing power to perform the computing task, such as the computing power of the edge server or cloud server is greater than the number of resources of the task. In this way, according to the preset second rule, each node is grouped to obtain two subset groups: the first subset group contains a set of servers with sufficient computing power to perform the computing task, and the second subset group refers to the remaining server combination that cannot meet all of its own services for the task, and thus needs to transfer their tasks to other edge servers. In this way, for the second subset group, the initial node needs to be updated to obtain several updated nodes.

[0106] Step S303 includes the following steps:

[0107] S3031, obtaining transmission modes of several updated nodes;

[0108] S3032, obtaining a total energy consumption value corresponding to the transmission mode and a total response time value corresponding to the transmission mode;

[0109] S3033. For each priority level, according to the node, the transmission mode corresponding to the node, the total energy consumption value corresponding to the transmission mode, and the total response time value corresponding to the transmission mode, the combination scheme of the terminal, edge server, and cloud server in the multimedia end-edge-cloud system is optimized and deployed to obtain the target deployment scheme of the multimedia end-edge-cloud system.

[0110] Specifically, for the second subset group, if the initial node is to be updated, the transmission mode of the corresponding node is also changed, so the transmission modes of several updated nodes are to be obtained; then the total energy consumption value corresponding to the transmission mode and the total response time value corresponding to the transmission mode are obtained. In one transmission mode, the task may be executed in the terminal, edge server or cloud server, according to the constraint condition of formula (11), therefore, one transmission mode corresponds to one total energy consumption value and one total response time value. In practice, the transmission process is to process all tasks of the same priority level in parallel to improve the efficiency of the network. The processing of tasks of different priority levels has been involved in the prior art and will not be repeated here. In this embodiment, the following steps are only performed for all tasks in the same priority level: the transmission mode with a total energy consumption value greater than a preset energy consumption threshold is classified into a disposable solution set, and the total energy consumption value in the disposable solution set is recorded; the transmission mode with a total energy consumption value less than or equal to the preset energy consumption threshold is classified into a non-disposable solution set; wherein the non-disposable solution set is an optimal solution set; the disposable solution set is a set other than the optimal solution set; the transmission mode with a total response time value less than a preset response time threshold is classified into a non-disposable solution set, and the total response time value in the non-disposable solution set is recorded; the transmission mode with a total response time value greater than or equal to the preset response time threshold is classified into a disposable solution set; The nodes and the transmission modes corresponding to the nodes in the set of controllable solutions are used as the first initial solution; the nodes and the transmission modes corresponding to the nodes in the set of non-controllable solutions are used as the second initial solution; the first initial solutions in the set of controllable solutions are sorted; based on the congestion strategy, the first initial solutions in the set of controllable solutions are weighted so that the difference between the weight of the first initial solution ranked first and the weight of the first initial solution at the end of the sorting is greater than a preset threshold; wherein the congestion strategy is the congestion distance, that is, the distance between the weight of the first initial solution ranked first and the weight of the first initial solution at the end of the sorting approaches infinity, which is impossible to be infinite in practice, and infinity can be set as a larger threshold (such as 40000). Based on the optimal learning strategy and the suboptimal learning strategy, the weighted first initial solution is updated and ranked by performance to obtain an updated first initial solution; wherein the optimal learning strategy and the suboptimal learning strategy are expressed by the following formula:

[0111]

[0112] Among them, Φ is the set of updated population strategies; X set is the population strategy set before updating; η is the update step size; and They are respectively towards the optimal strategy X optimal and the suboptimal strategy X suboptimalThe learning weight of . Based on the preset elimination strategy, the updated first initial plan is eliminated to obtain the intermediate deployment plan; wherein, the preset elimination strategy is to adopt the competition relationship between servers and set a random probability threshold. When the computing power of the server is less than the random probability threshold, the node corresponding to the server (edge ​​server or cloud server) is deleted, thereby ensuring the stability of the number of strategies. The intermediate deployment plan is copied to obtain the target deployment plan; the corresponding energy consumption value, response time value and optimal deployment plan are recorded; the second initial plan in the non-dominable solution set is sorted by performance, and the second initial plan ranked first is obtained. When the total energy consumption value in the second initial plan ranked first is smaller than the total energy consumption value in the target deployment plan and the total response time value in the second initial plan ranked first is smaller than the total response time value in the target deployment plan, the second initial plan ranked first is used as the target deployment plan; the first initial plan in the dominable solution set is weighted repeatedly based on the congestion strategy so that the difference between the weight of the first initial plan ranked first and the weight of the first initial plan at the end of the sorting is greater than a preset threshold, until the optimal target deployment plan is obtained, that is, the iteration converges, or, when the iteration exceeds the maximum number of iterations (such as 4000), the solution set makes the energy consumption value and the response time value unable to be better, then the result of the target deployment plan is output.

[0113] In one implementation, Figure 4 As shown, when the task volume has a random traffic demand or the server is under maintenance, the priority level of the attribute information of the task request changes, and it is necessary to dynamically manage it through the control module of the edge node. When the attribute information of the task request is updated, the priority sorting of the task is re-executed according to the updated attribute information of the task request to obtain several priority levels. Similarly, the task with a high priority level is still deployed first. Then the total energy consumption value of the task and the total response time value of the task are calculated; according to the attribute information of the task request, the total energy consumption value and the total response time value, the combination of the terminal, edge server and cloud server in the multimedia end-edge cloud system is optimized and deployed to obtain the target deployment scheme of the multimedia end-edge cloud system, so as to minimize the total energy consumption value and total response time value corresponding to the target deployment scheme, that is, to update the deployment in real time, re-give the edge node deployment scheme, obtain the optimal energy consumption value and response time value corresponding to the new node deployment scheme, record the optimal deployment result, and re-output the optimal deployment, that is, the target deployment scheme.

[0114] Highlights of the invention:

[0115] 1. Analysis, sorting and regional teaming of the resource volume and request time of multiple tasks provide clear data support for the subsequent calculation of response time;

[0116] 2. Considering the cooperation and competition between servers, an edge-cloud model of multimedia system based on energy consumption and response time is constructed, and the dynamic topology structure of nodes between edge servers and cloud servers is designed to implement the specific solution of edge node deployment;

[0117] 3. Dynamic management of edge nodes is realized. When the workload changes dramatically, server failures occur, or there are increases or decreases, the swarm intelligent control method of the present invention can provide corresponding deployment plans.

[0118] Exemplary Devices

[0119] like Figure 5 As shown in , an embodiment of the present invention provides a swarm intelligent control device for a multimedia end-edge-cloud system, the device comprising a task request attribute information acquisition module 401, a total energy consumption value and total response time value calculation module 402 and a scheme optimization deployment module 403: the task request attribute information acquisition module 401 is used to obtain attribute information of a task request in a heterogeneous network; wherein the task request is used to represent a request instruction for a terminal to send a task to an edge server;

[0120] The total energy consumption value and total response time value calculation module 402 is used to prioritize the tasks according to the attribute information of the task request, obtain a number of priority levels, and calculate the total energy consumption value and the total response time value of the tasks; wherein one priority level corresponds to a plurality of tasks;

[0121] The solution optimization deployment module 403 is used to optimize the deployment of the combination of terminals, edge servers and cloud servers in the multimedia end-edge-cloud system according to the attribute information of the task request, the total energy consumption value and the total response time value, and obtain the target deployment solution of the multimedia end-edge-cloud system so that the total energy consumption value and the total response time value corresponding to the target deployment solution are minimized.

[0122] Based on the above embodiments, the present invention further provides an intelligent terminal, whose principle block diagram can be shown as follows: Figure 6As shown. The smart terminal includes a processor, a memory, a network interface, a display screen, and a temperature sensor connected via a system bus. Among them, the processor of the smart terminal is used to provide computing and control capabilities. The memory of the smart terminal includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the smart terminal is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a group intelligent control method of a multimedia end-edge cloud system is implemented. The display screen of the smart terminal can be a liquid crystal display screen or an electronic ink display screen, and the temperature sensor of the smart terminal is pre-set inside the smart terminal to detect the operating temperature of the internal device.

[0123] Those skilled in the art will understand that Figure 6 The schematic diagram in the figure is only a block diagram of a part of the structure related to the scheme of the present invention, and does not constitute a limitation on the smart terminal to which the scheme of the present invention is applied. The specific smart terminal may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0124] In one embodiment, a smart terminal is provided, comprising a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors, and the one or more programs include instructions for performing the following operations:

[0125] Acquire attribute information of a task request in a heterogeneous network; wherein the task request is used to represent a request instruction for a terminal to send a task to an edge server;

[0126] According to the attribute information of the task request, the tasks are prioritized to obtain several priority levels, and the total energy consumption value of the tasks and the total response time value of the tasks are calculated; wherein one priority level corresponds to multiple tasks;

[0127] According to the attribute information of the task request, the total energy consumption value and the total response time value, the combination of the terminal, edge server and cloud server in the multimedia end-edge-cloud system is optimized and deployed to obtain the target deployment scheme of the multimedia end-edge-cloud system so that the total energy consumption value and the total response time value corresponding to the target deployment scheme are minimized.

[0128] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0129] In summary, the present invention discloses a swarm intelligent control method for a multimedia end-edge-cloud system, the method comprising: obtaining attribute information of task requests in a heterogeneous network; prioritizing tasks according to the attribute information of task requests to obtain several priority levels; calculating the total energy consumption value of tasks and the total response time value of tasks; wherein one priority level corresponds to multiple tasks; and optimizing the deployment of a combination of terminals, edge servers, and cloud servers in a multimedia end-edge-cloud system according to the attribute information, total energy consumption value, and total response time value of task requests to obtain a target deployment scheme of the multimedia end-edge-cloud system, wherein the total energy consumption value and total response time value of the target deployment scheme are the smallest among all deployment schemes. The embodiment of the present invention deploys a combination scheme of each node in a multimedia end-edge-cloud system so that each node can allocate efficient energy consumption requirements and lower response delay requirements.

[0130] Based on the above embodiments, the present invention discloses a swarm intelligent control method for a multimedia end-edge cloud system. It should be understood that the application of the present invention is not limited to the above examples. For ordinary technicians in this field, it can be improved or transformed according to the above description. All these improvements and transformations should fall within the scope of protection of the claims attached to the present invention.

Claims

1. A swarm intelligent control method for a multimedia edge-cloud system, characterized in that: The method comprises: Acquire attribute information of a task request in a heterogeneous network; wherein the task request is used to represent a request instruction for a terminal to send a task to an edge server; According to the attribute information of the task request, the tasks are prioritized to obtain a plurality of priority levels, and the total energy consumption value of the tasks and the total response time value of the tasks are calculated; wherein one priority level corresponds to a plurality of tasks; According to the attribute information of the task request, the total energy consumption value and the total response time value, the combination scheme of the terminal, the edge server and the cloud server in the multimedia end-edge-cloud system is optimized and deployed to obtain the target deployment scheme of the multimedia end-edge-cloud system, so that the total energy consumption value and the total response time value corresponding to the target deployment scheme are minimized; According to the attribute information of the task request, the total energy consumption value and the total response time value, the combination scheme of the terminal, the edge server and the cloud server in the multimedia end-edge-cloud system is optimized and deployed, and the target deployment scheme of the multimedia end-edge-cloud system is obtained, including: Based on a preset first rule, and according to the attribute information of the task request, a node is allocated to each task to obtain a plurality of nodes corresponding to each task; wherein the node is used to represent a connection point of a terminal, an edge server or a cloud server; When the node is an edge server or a cloud server, the computing capacity of the node is obtained, wherein the computing capacity is used to characterize the computing processing capacity of the server; based on a preset second rule, each node is grouped according to the computing capacity of the node to obtain two subset groups; according to the two subset groups, a number of the nodes are updated to obtain a number of updated nodes; According to a number of priority levels, a number of updated nodes, the total energy consumption value and the total response time value, a combination of terminals, edge servers and cloud servers in the multimedia end-edge-cloud system is optimally deployed to obtain a target deployment plan of the multimedia end-edge-cloud system; The combination scheme of the terminal, edge server and cloud server in the multimedia end-edge-cloud system is optimized and deployed according to the priority levels, the updated nodes, the total energy consumption value and the total response time value, and the target deployment scheme of the multimedia end-edge-cloud system is obtained, including: Get the transmission mode of several updated nodes; Obtaining a total energy consumption value corresponding to the transmission mode and a total response time value corresponding to the transmission mode; For each priority level, according to the node, the transmission mode corresponding to the node, the total energy consumption value corresponding to the transmission mode, and the total response time value corresponding to the transmission mode, the combination scheme of the terminal, the edge server, and the cloud server in the multimedia end-edge-cloud system is optimized and deployed to obtain the target deployment scheme of the multimedia end-edge-cloud system; According to the node, the transmission mode corresponding to the node, the total energy consumption value corresponding to the transmission mode, and the total response time value corresponding to the transmission mode, the combination scheme of the terminal, the edge server, and the cloud server in the multimedia end-edge-cloud system is optimized and deployed, and the target deployment scheme of the multimedia end-edge-cloud system is obtained, including: Classify the transmission mode whose total energy consumption value is greater than the preset energy consumption threshold into the disposable solution set, and record the total energy consumption value in the disposable solution set; classify the transmission mode whose total energy consumption value is less than or equal to the preset energy consumption threshold into the non-disposable solution set; wherein the non-disposable solution set is the optimal solution set; the disposable solution set is the set other than the optimal solution set; classify the transmission mode whose total response time value is less than the preset response time threshold into the non-disposable solution set, and record the total response time value in the non-disposable solution set; classify the transmission mode whose total response time value is greater than or equal to the preset response time threshold into the disposable solution set; take the nodes in the disposable solution set and the transmission mode corresponding to the nodes as the first initial solution; take the nodes in the non-disposable solution set and the transmission mode corresponding to the nodes as the second initial solution; sorting the first initial solutions in the set of dominable solutions; Based on the congestion strategy, the first initial solution in the set of disposable solutions is weighted so that the difference between the weight of the first initial solution ranked first and the weight of the first initial solution ranked last is greater than a preset threshold; Based on the optimal learning strategy and the suboptimal learning strategy, the weighted first initial solution is updated and ranked by performance to obtain an updated first initial solution; Based on a preset elimination strategy, the updated first initial plan is eliminated to obtain an intermediate deployment plan; Copy the intermediate deployment plan to obtain the target deployment plan; Rank the second initial solutions in the non-dominant solution set by performance, and obtain the second initial solution ranked first; When the total energy consumption value in the first-ranked second initial solution is smaller than the total energy consumption value in the target deployment solution and the total response time value in the first-ranked second initial solution is smaller than the total response time value in the target deployment solution, the first-ranked second initial solution is used as the target deployment solution; Repeat the steps of weighting the first initial solution in the set of available solutions based on the congestion strategy so that the difference between the weight of the first initial solution ranked first and the weight of the first initial solution ranked last is greater than a preset threshold until the optimal target deployment solution is obtained.

2. The swarm intelligent control method of the multimedia end-edge-cloud system according to claim 1 is characterized in that: The attribute information of the task request includes the delay requirement of task transmission, the sequence number of the process sequence in which the task is located, and the resource requirement of the task.

3. The swarm intelligent control method of the multimedia end-edge-cloud system according to claim 1 is characterized in that: The number of tasks is several, each task includes several subtasks; the number of terminals is several; the calculation of the total energy consumption value of the task and the total response time value of the task includes: For each subtask in each terminal, calculating a first energy consumption value of each subtask in each terminal and a first response time value of each subtask; According to the first energy consumption value of each subtask and the first response time value of each subtask in each terminal, the total energy consumption value of the task and the total response time value of the task are obtained.

4. The swarm intelligent control method of the multimedia end-edge-cloud system according to claim 3 is characterized in that: The calculating, for each subtask in each terminal, a first energy consumption value of each subtask in each terminal and a first response time value of each subtask comprises: For each terminal, calculating a first initial energy consumption value and a first initial response time value when each terminal executes each subtask; Calculating a second initial energy consumption value and a second initial response time value when executing each subtask on a path from each terminal to the edge server; Calculate a third initial energy consumption value and a third initial response time value when executing each subtask on a path from each terminal via the edge server to the cloud server; Adding the first initial energy consumption value, the second initial energy consumption value, and the third initial energy consumption value to obtain a first energy consumption value for each subtask in each terminal; The maximum value among the first initial response time value, the second initial response time value and the third initial response time value is obtained to obtain the first response time value of each subtask in each terminal.

5. The swarm intelligent control method of the multimedia end-edge-cloud system according to claim 3 is characterized in that: The obtaining, according to the first energy consumption value of each subtask in each terminal and the first response time value of each subtask, the total energy consumption value of the task and the total response time value of the task comprises: Adding the first energy consumption values ​​of all subtasks in each terminal to obtain the second energy consumption value of each task in each terminal; Obtain the maximum value of the first response time of all subtasks in each terminal, and obtain the second response time value of each task in each terminal; According to the second energy consumption value of each task in each terminal and the second response time value of each task in each terminal, the total energy consumption value of the task and the total response time value of the task are obtained.

6. The swarm intelligent control method of the multimedia end-edge-cloud system according to claim 5 is characterized in that: The obtaining, according to the second energy consumption value of each task in each terminal and the second response time value of each task in each terminal, the total energy consumption value of the task and the total response time value of the task comprises: Adding the second energy consumption values ​​of all tasks in each terminal to obtain a third energy consumption value in each terminal; Add the third energy consumption values ​​of all terminals to obtain the total energy consumption value of the multimedia end-edge-cloud system; Obtaining the maximum value of the second response time of each task in all terminals, and obtaining the third response time value of each task in all terminals; Add the third response time value of each task in all terminals to obtain the total response time value of the multimedia end-edge-cloud system.

7. The swarm intelligent control method of a multimedia end-edge-cloud system according to claim 1, characterized in that: The method further comprises: When the attribute information of the task request is updated, the priority sorting of the task is re-executed according to the updated attribute information of the task request to obtain several priority levels; the total energy consumption value of the task and the total response time value of the task are calculated; according to the attribute information of the task request, the total energy consumption value and the total response time value, the combination of the terminal, the edge server and the cloud server in the multimedia end-edge-cloud system is optimized and deployed to obtain the target deployment scheme of the multimedia end-edge-cloud system, so as to minimize the total energy consumption value and the total response time value corresponding to the target deployment scheme.

8. A swarm intelligent control device for a multimedia edge-cloud system, characterized in that: The device is applied to the swarm intelligent control method of the multimedia end-edge-cloud system according to any one of claims 1 to 7, and the device comprises: The task request attribute information acquisition module is used to obtain the attribute information of the task request in the heterogeneous network; wherein the task request is used to represent the request instruction for the terminal to send the task to the edge server; A total energy consumption value and total response time value calculation module, used to prioritize the tasks according to the attribute information of the task request, obtain a number of priority levels, and calculate the total energy consumption value of the tasks and the total response time value of the tasks; wherein one priority level corresponds to a plurality of tasks; The solution optimization deployment module is used to optimize the deployment of the combination of terminals, edge servers and cloud servers in the multimedia end-edge-cloud system according to the attribute information of the task request, the total energy consumption value and the total response time value, and obtain the target deployment solution of the multimedia end-edge-cloud system so that the total energy consumption value and the total response time value corresponding to the target deployment solution are minimized.

9. An intelligent terminal, characterized in that: The device comprises a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors, and the one or more programs include being used to execute the method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that: When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the method as described in any one of claims 1 to 7.

Images