Green edge computing task unloading and resource optimization method

Abstract

The invention discloses a green edge computing task unloading and resource optimization method, which comprises the following steps of: acquiring task types of an intelligent port in processing, and classifying the task types by considering a time delay demand and an energy demand; establishing a network optimization resource model, processing the classified task types, and taking the minimum energy consumption and time delay satisfaction degree as a task unloading target; performing global optimization on a task unloading target based on a grey wolf algorithm to obtain an optimal network resource allocation solution; the method can dynamically adjust task allocation, reduce system energy consumption, optimize response speed of delay sensitive tasks, improve resource utilization efficiency of the system, reduce network load and improve task processing stability, is especially suitable for scenes with high energy efficiency requirements such as smart ports, and has high practicability.

Description

Technical Field

[0001] The present invention relates to the technical field of energy management and resource optimization, and in particular to a green edge computing task offloading and resource optimization method. Background Art

[0002] With the rapid development of information technology, the operation of modern smart ports increasingly relies on the real-time processing and analysis of large-scale data. The gradual maturity of technologies such as the Internet of Things (IoT), edge computing, and cloud computing has provided strong support for the automated management and optimization of smart ports. In the smart port scenario, ship berthing, cargo loading and unloading, equipment maintenance, and other links all require the use of intelligent devices for real-time monitoring and scheduling. These tasks often have strict requirements on processing latency and energy consumption. Therefore, how to optimize resource allocation and reduce computing energy consumption under the premise of ensuring efficient completion of tasks has become a problem that needs to be solved urgently. Traditional computing tasks are usually processed centrally in the cloud. Although they have strong computing power, due to the high latency of data transmission, it is difficult to meet the low latency requirements of some tasks in the port. At the same time, edge computing provides a new idea for solving the latency problem. It allows some computing tasks to be offloaded to edge devices close to the data source for processing, thereby reducing network pressure. However, when facing different tasks, the existing edge computing solutions lack the classification and distinction of latency sensitivity and energy consumption, making it difficult to achieve the optimal task offloading strategy.

[0003] Although the current research on task offloading in smart ports can optimize the allocation of tasks to a certain extent, it generally has the following shortcomings: First, in terms of task classification and processing, the existing methods lack a detailed division of task types, especially fail to fully consider the differences in task latency requirements and energy consumption requirements; second, most of the existing task offloading strategies adopt static or pre-set rules, and fail to dynamically adjust the allocation and offloading methods of resources according to the actual operating status, resulting in difficulty in achieving the optimal system energy efficiency and response time; finally, in terms of resource optimization, the global optimization algorithm is not combined for dynamic task allocation, making the task offloading strategy between multiple nodes not intelligent and flexible enough. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the above existing problems, the present invention is proposed. Therefore, the present invention provides a green edge computing task offloading and resource optimization method to solve the problems raised in the background technology.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a green edge computing task offloading and resource optimization method, comprising:

[0007] Obtaining the task types being processed by the smart port, and classifying the task types considering latency requirements and energy requirements;

[0008] Establish a network optimization resource model to process the classified task types and take minimizing energy consumption and delay satisfaction as the task offloading goal;

[0009] The task offloading target is globally optimized based on the Grey Wolf Algorithm to obtain the optimal network resource allocation solution.

[0010] As a preferred solution of the green edge computing task offloading and resource optimization method of the present invention, the task types are classified considering the delay requirements and energy requirements, including:

[0011] Delay-sensitive tasks, energy-sensitive tasks and mixed tasks, and corresponding delay thresholds are set for the delay-sensitive tasks, energy-sensitive tasks and mixed tasks respectively.

[0012] As a preferred solution of the green edge computing task offloading and resource optimization method of the present invention, the classified task types are processed, including:

[0013] For each task, an offloading decision is made based on the task's latency requirement and the node's remaining energy.

[0014] As a preferred solution of the green edge computing task offloading and resource optimization method of the present invention, the offloading decision includes:

[0015] Set the latency and energy requirements of the current task;

[0016] If the delay requirement is less than or equal to the delay threshold of the delay-sensitive task and the remaining energy is greater than the energy threshold, processing is performed on the local device of the smart port;

[0017] If the delay requirement is greater than the delay threshold of the delay-sensitive task and less than or equal to the delay threshold of the hybrid task, the task is offloaded to the associated base station for processing;

[0018] If the delay requirement is greater than the delay threshold of the hybrid task and the actual load of the current base station is greater than the load threshold of the base station, the task is forwarded to a neighboring base station of the associated base station for processing;

[0019] If the delay requirement is greater than the delay threshold of the energy-sensitive task or the complexity of the current task exceeds the calculation range of the local device and the base station, the task is offloaded to the cloud for processing.

[0020] As a preferred solution of the green edge computing task offloading and resource optimization method described in the present invention, wherein: minimizing energy consumption and delay satisfaction is used as the task offloading goal, including:

[0021] The task offloading target formula is:

[0022]

[0023] Among them, w 1 and w 2 Represent the dynamic weights of energy consumption and delay satisfaction, Indicates the energy consumption of the current node, EC is the total energy consumption of the node required from the start to the completion of the task, and E max represents the maximum energy of the node; It is expressed as the delay satisfaction of the task, Latency, the actual processing delay of the current task, T required is the latency requirement of the current task.

[0024] As a preferred solution of the green edge computing task offloading and resource optimization method of the present invention, it also includes:

[0025]

[0026] Among them, E available is the residual energy of the node.

[0027] As a preferred solution of the green edge computing task offloading and resource optimization method described in the present invention, the task offloading target is globally optimized based on the gray wolf algorithm, including:

[0028] The size N of the gray wolf population and the maximum number of iterations M are set. The initial position of each gray wolf individual is a set of randomly generated task offloading schemes, indicating whether the task is processed locally, at the base station, or in the cloud.

[0029] Calculate the fitness value of each individual gray wolf;

[0030] In each iteration, the individuals in the population are sorted according to their fitness values, and the three best individuals are selected, marked as α, β, and δ, representing the best, second-best, and third-best task offloading solutions, respectively. The other gray wolf individuals are updated according to the positions of α, β, and δ;

[0031] The iteration is stopped when the number of iterations reaches the maximum number of iterations M, or when the change rate of the optimal individual fitness value in two consecutive iterations is less than the change tolerance of the fitness value.

[0032] As a preferred solution of the green edge computing task offloading and resource optimization method of the present invention, it also includes:

[0033] The next position of each individual gray wolf is updated using the following formula:

[0034]

[0035] in, It is represented as the current task offloading scheme of individuals α, β, and δ, represents the position of the current individual, that is, the current position of the current gray wolf individual in the tth iteration; Represents the position of the current leader, which can be any one of α, β, and δ; is the direction and distance that the gray wolf is currently moving. is the intensity of the gray wolf's proximity to the leader,

[0036] Compared with the prior art, the invention has the following beneficial effects:

[0037] 1. In the prior art, there is often a lack of refined task classification and dynamic decision-making mechanism for the processing of diversified tasks in smart ports. In particular, when facing delay-sensitive tasks and energy-sensitive tasks, it is impossible to flexibly adjust according to actual task requirements, resulting in low resource utilization efficiency or task delays. The present invention classifies tasks in a refined manner, divides tasks into delay-sensitive tasks, energy-sensitive tasks and mixed tasks, and sets corresponding delay thresholds and energy requirements respectively. Compared with the existing static task allocation scheme, it can dynamically classify and process tasks according to their different characteristics, thereby realizing refined management, which not only improves the system response speed, but also ensures the low-energy execution of energy-intensive tasks, effectively avoiding the waste of resources and the reduction of task timeliness caused by insufficient task characteristic processing in the prior art;

[0038] 2. In the prior art, task offloading mostly adopts preset rules or static strategies, lacking a mechanism for dynamic adjustment according to the actual environment, resulting in the system being unable to flexibly cope with the complex and changing task loads in the port. The present invention calculates the delay weight and energy consumption weight of each task, combined with the energy state of the current node and the network load, to achieve real-time task offloading decisions. The dynamic adjustment mechanism can effectively solve the problem of low system energy efficiency caused by fixed strategies in the prior art, ensuring that in a complex port environment, task offloading can be flexibly optimized and allocated between local processing, base station processing and cloud processing, thereby significantly reducing energy consumption and meeting delay requirements;

[0039] 3. The existing technology is insufficient in the global coordination of resource optimization and task offloading, and can usually only achieve local optimization, but cannot consider the resource utilization and task processing load of different nodes from the perspective of the whole system; the present invention introduces the gray wolf algorithm to perform global optimization in the task offloading process, and reasonably allocates system resources to different types of tasks, so as to balance the contradiction between latency and energy consumption; the present invention not only improves the accuracy of task offloading, but also can reduce resource waste on a global scale, and improve the processing efficiency and stability of the system; in addition, it also enables the system to adaptively adjust as the task load changes, ensuring the optimal resource utilization and task responsiveness in different scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0041] Figure 1 The present invention is an overall flow chart of a green edge computing task offloading and resource optimization method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0042] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0043] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0044] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0045] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0046] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0047] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0048] Example 1

[0049] Reference Figure 1 , which is the first embodiment of the present invention, and provides a green edge computing task offloading and resource optimization method, including:

[0050] S1. Obtain the task types being processed by the smart port and classify the task types considering the delay requirements and energy requirements;

[0051] Furthermore, delay-sensitive tasks, energy-sensitive tasks and mixed tasks, and corresponding delay thresholds are set for the delay-sensitive tasks, energy-sensitive tasks and mixed tasks respectively;

[0052] Specifically, the delay threshold of the delay-sensitive task is the low delay T low , such as real-time tracking of cargo in ports and monitoring of port equipment. Such tasks have high real-time requirements and must be completed within strict delay limits to ensure the efficiency and safety of port operations;

[0053] Specifically, the latency threshold of energy-sensitive tasks is high latency T high , such as environmental monitoring and long-term data analysis of ports. Such tasks allow longer delays, but have strict requirements on energy consumption. It is necessary to minimize the energy consumption of the system to ensure long-term stable operation of the equipment;

[0054] Specifically, the delay threshold of the mixed task is medium delay T medium , there are both latency requirements and energy consumption considerations, such as ship berthing management and port departure scheduling at ports. Such tasks must dynamically balance energy consumption and latency to ensure the overall efficiency and energy saving of port operations;

[0055] S2. Establish a network optimization resource model to process the classified task types and take minimizing energy consumption and delay satisfaction as the task offloading goal;

[0056] Furthermore, for each task, an offloading decision is made based on the task’s latency requirement and the node’s remaining energy;

[0057] Specifically, the offloading decisions are divided into: local processing, offloading to the associated base station for processing, forwarding to the adjacent base station of the associated base station through the associated base station for processing, and offloading to the cloud for processing;

[0058] Furthermore, the uninstall decision rule is as follows:

[0059] Set the latency requirement T for the current task required and energy requirement E required ;

[0060] Local processing: If the delay requirement is less than or equal to the delay threshold of the delay-sensitive task and the remaining energy E available Greater than the energy threshold E threshold , it is processed on the local equipment of the smart port;

[0061] Specifically, the conditions in local processing are expressed as follows:

[0062] T required ≤T low And E available >Ethreshold

[0063] It should be noted that when E available <E threshold When , it means that the node is energy-constrained and cannot continue to process more tasks;

[0064] Offload to the associated base station for processing: If the latency requirement is greater than the latency threshold of the latency-sensitive task and less than or equal to the latency threshold of the hybrid task, the task is offloaded to the associated base station for processing;

[0065] Specifically, the condition for offloading to the associated base station is expressed by the formula:

[0066] T low <T required ≤T medium

[0067] Forwarded by the associated base station to the adjacent base station for processing: If the delay requirement is greater than the delay threshold of the hybrid task and the actual load L of the current base station current Greater than the base station load threshold L threshold , then the task is forwarded to the neighboring base station of the associated base station for processing;

[0068] Specifically, the condition for forwarding to the neighboring base station of the associated base station for processing is expressed by the formula:

[0069] T reguired >T medium And L current >L threshold

[0070] Offloading to the cloud for processing: If the latency requirement is greater than the latency threshold of the energy-sensitive task or the complexity of the current task exceeds the calculation range of the local device and base station, the task will be offloaded to the cloud for processing;

[0071] Specifically, the conditions for offloading to the cloud for processing are expressed as follows:

[0072] T require d>T high

[0073] It should be explained that the current task complexity exceeds the computing range of local devices and base stations, which means that the computing resources, processing delay and multi-task concurrent processing capabilities required by the task exceed the computing capabilities of local devices and base stations; that is, when the computing requirements of the task exceed the available CPU cycles, memory or storage resources, or the processing cannot be completed within the specified delay threshold in the local device and base station, the task needs to be offloaded to the cloud for processing; in addition, when multiple tasks are running concurrently, if the resource competition between the local device and the base station causes the task to be unable to be effectively executed, the task also needs to be offloaded to the cloud;

[0074] Furthermore, the task offloading target formula is:

[0075]

[0076] Among them, w 1 and w 2 Represent the dynamic weights of energy consumption and delay satisfaction, Indicates the energy consumption of the current node, EC is the total energy consumption of the node required from the start to the completion of the task, and E max represents the maximum energy of the node; It is expressed as the delay satisfaction of the task, Latency, the actual processing delay of the current task, T required The latency requirement of the current task;

[0077] Furthermore,

[0078]

[0079] Among them, E available is the residual energy of the node;

[0080] It should be noted that w 1 and w 2 The value of will change dynamically according to the actual needs of the task, ensuring that latency and energy consumption can be balanced during the execution of the task; for example, when the latency requirement is urgent, w 1 The value of will increase, and when the remaining energy of the node is small, w 2 will increase, indicating the need to prioritize energy consumption;

[0081] S3, perform global optimization on the task offloading target based on the Grey Wolf Algorithm to obtain the optimal network resource allocation solution;

[0082] It needs to be explained that the Grey Wolf Optimizer (GWO) is a swarm intelligence algorithm that imitates the hunting behavior of grey wolf groups and approaches the optimal solution by continuously updating the positions of individual grey wolves.

[0083] Furthermore, the size N of the gray wolf population and the maximum number of iterations M are set, and the initial position of each gray wolf individual is a set of randomly generated task offloading schemes, indicating whether the task is processed locally, at the base station, or in the cloud;

[0084] Furthermore, the fitness value of each individual gray wolf is calculated;

[0085] Specifically, the fitness value is:

[0086]

[0087] Where x represents the specific offloading scheme of the task, EC(x) represents the EC corresponding to the offloading scheme x, and Latency(x) represents the Latency corresponding to the offloading scheme x;

[0088] Furthermore, in each iteration, the individuals in the population are sorted according to the fitness values, and the three best individuals are selected, which are marked as α, β, and δ, representing the best, second-best, and third-best task offloading solutions, respectively. The other gray wolf individuals are updated according to the positions of α, β, and δ;

[0089] Specifically, the next position of each individual gray wolf is updated using the following formula:

[0090]

[0091] in, It is represented as the current task offloading scheme of individuals α, β, and δ, represents the position of the current individual, that is, the current position of the current gray wolf individual in the tth iteration; Represents the position of the current leader, which can be any one of α, β, and δ; is the direction and distance that the gray wolf is currently moving. is the intensity of the gray wolf's proximity to the leader,

[0092] It should be noted that is a dynamic control parameter, which gradually decreases with the number of algorithm iterations; and is a random number vector ranging from [0,1]; Determines the degree to which the individual approaches or moves away from the prey (i.e., the optimal solution individuals α, β, δ); when When it is close to 0, the individual will move closer to the optimal solution; When it is close to 1, the individual will move away from the optimal solution, thereby expanding the exploration space; this allows the gray wolf individual to conduct global exploration in the early stage, and gradually converge to the local area for development in the later stage; It is used to determine the distance between the individual gray wolf and the prey (optimal solution). It can determine the direction in which the individual "aims" during the update process. It will affect the direction in which each gray wolf approaches or moves away from the prey. It enables individuals to continuously adjust their directions in space to find the optimal solution;

[0093] Furthermore, the iteration is stopped when the number of iterations reaches the maximum number of iterations M, or when the change rate of the best individual fitness value in two consecutive iterations is less than the change tolerance of the fitness value;

[0094] Specifically, when the change rate of the optimal individual fitness value of two consecutive iterations is less than the change tolerance of the fitness value, the iteration is stopped and expressed by the formula:

[0095]

[0096] Among them, f α (t) represents the fitness value of the best individual (i.e., individual α) in the tth iteration; f α (t+1) represents the fitness value of the best individual in the t+1th iteration, Δf α It represents the rate of change of the best individual fitness value between two consecutive iterations, ∈ represents the tolerance of the fitness value change, which is a very small positive number, usually 10 -6 or 10 -5 ;

[0097] It should be noted that a rate of change less than ∈ means that the optimal solution tends to be stable during the iteration, and the improvement of further iterations is very small, and the algorithm can be considered to have converged.

[0098] Example 2

[0099] Referring to Table 1, the second embodiment of the present invention provides a green edge computing task offloading and resource optimization method, including: verifying the overall performance of the scheme of the present invention by designing a simulation experiment; the simulation experiment scene is set in a smart port, and the task offloading includes local device processing, associated base station processing, neighboring base station forwarding and cloud processing; delay-sensitive tasks, energy-sensitive tasks and hybrid tasks are introduced in the simulation experiment; the purpose is to compare the task offloading optimization scheme of the present invention with the traditional task offloading scheme, focusing on the energy consumption, delay and system response performance during task execution;

[0100] First, define the delay thresholds of different tasks; the delay requirement of delay-sensitive tasks is set to T low =50ms, the delay requirement of energy-sensitive tasks is set to T high = 200ms, the delay requirement of the mixed task is T medium = 100ms; During the processing of each task, the system dynamically makes unloading decisions based on the task's latency requirements and the node's remaining energy, ensuring that the task minimizes energy consumption while meeting latency requirements. For the energy limits of local devices and associated base stations, the maximum energy of the node is set to E max =100J, the energy threshold is E threshold =10J;

[0101] Secondly, the defined tasks will be allocated to different processing nodes (local devices, base stations or cloud) according to the remaining energy and delay requirements; in the inventive method, the task offloading scheme is globally optimized based on the gray wolf algorithm, and the energy consumption and delay of the task are dynamically adjusted to optimize the offloading decision in real time; the gray wolf population size set in the simulation is 20, and the number of iterations is 100, and the energy consumption and delay of each task under different processing methods are tested; for comparative analysis, the traditional fixed task offloading scheme (based on preset rules) is compared with the dynamic optimization scheme invented in this article; in the traditional scheme, all tasks are fixedly processed locally or in the cloud according to the preset rules, and no dynamic adjustment is made according to energy and delay. The results are obtained by referring to Table 1;

[0102] Table 1

[0103]

[0104]

[0105] It can be seen from Table 1 that for delay-sensitive tasks, in the traditional solution, due to the lack of dynamic optimization, the control effect of energy consumption and delay is poor, especially when the local device is processing, the energy consumption is as high as 50J, and the delay is increased to 60ms; under the invention solution, the energy consumption is greatly reduced to 30J, the delay is also controlled within 45ms, and the task completion rate is increased to 95%; this improvement is mainly due to the fact that the invention gives priority to meeting the low delay requirements of delay-sensitive tasks and reasonably allocates computing resources through dynamic weight adjustment, and reduces unnecessary energy consumption through optimization algorithms; for energy-sensitive tasks, Table 1 shows that the energy consumption of the invention solution when processed by the local device is 20J, which saves more than 30% of energy consumption compared with the 30J of the traditional solution, and The latency and completion rate of the task are maintained at a high level (latency 150ms, completion rate 98%). This is because the gray wolf algorithm is introduced in the present invention for global optimization, which can monitor the energy consumption of each node in real time, and by adjusting the task allocation, give priority to nodes with lower energy consumption for task processing. For mixed tasks, the traditional solution fails to balance latency and energy consumption well during processing, resulting in high energy consumption during base station and cloud processing (45J and 50J, respectively), and latency exceeds expectations (110ms and 130ms, respectively). However, the solution of the present invention can flexibly balance latency and energy consumption, select appropriate nodes for processing through a dynamic optimization algorithm, and the energy consumption during base station processing is 35J, and the latency is only 90ms, which is significantly better than the traditional solution.

[0106] Through the comparative data in Table 1, it can be clearly seen that the present invention realizes the dual optimization of energy consumption and delay in the scenario of smart port, avoiding the low energy efficiency problem caused by unreasonable resource allocation in traditional methods; it illustrates that the dynamic task offloading and resource optimization method of the present invention can adapt to the energy state and task delay requirements of each node in the system, and introduce the gray wolf algorithm for global optimization to ensure the optimal resource allocation and task processing efficiency in a complex network environment; compared with the traditional fixed offloading scheme, the task processing effect is significantly improved, ensuring that while meeting the delay requirements, the energy consumption of the system is minimized, and it is practical.

[0107] Those skilled in the art will appreciate that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program codes. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0108] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0109] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0110] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0111] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0112] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A green edge computing task offloading and resource optimization method, characterized in that: include: Obtaining the task types being processed by the smart port, and classifying the task types considering latency requirements and energy requirements; Establish a network optimization resource model to process the classified task types and take minimizing energy consumption and delay satisfaction as the task offloading goal; The task offloading target is globally optimized based on the Grey Wolf Algorithm to obtain the optimal network resource allocation solution.

2. The green edge computing task offloading and resource optimization method according to claim 1, characterized in that: The task types are classified by considering the delay requirements and energy requirements, including: Delay-sensitive tasks, energy-sensitive tasks and mixed tasks, and corresponding delay thresholds are set for the delay-sensitive tasks, energy-sensitive tasks and mixed tasks respectively.

3. The green edge computing task offloading and resource optimization method according to claim 1, characterized in that: Process the classified task types, including: For each task, an offloading decision is made based on the task's latency requirement and the node's remaining energy.

4. The green edge computing task offloading and resource optimization method according to claim 2 or 3, characterized in that: The uninstall decision includes: Set the latency and energy requirements of the current task; If the delay requirement is less than or equal to the delay threshold of the delay-sensitive task and the remaining energy is greater than the energy threshold, processing is performed on the local device of the smart port; If the delay requirement is greater than the delay threshold of the delay-sensitive task and less than or equal to the delay threshold of the hybrid task, the task is offloaded to the associated base station for processing; If the delay requirement is greater than the delay threshold of the hybrid task and the actual load of the current base station is greater than the load threshold of the base station, the task is forwarded to a neighboring base station of the associated base station for processing; If the delay requirement is greater than the delay threshold of the energy-sensitive task or the complexity of the current task exceeds the calculation range of the local device and the base station, the task is offloaded to the cloud for processing.

5. The green edge computing task offloading and resource optimization method according to claim 4, characterized in that: Minimizing energy consumption and delay satisfaction is the goal of task offloading, including: The task offloading target formula is: Among them, w1 and w2 represent the dynamic weights of energy consumption and delay satisfaction, respectively. Indicates the energy consumption of the current node, EC is the total energy consumption of the node required from the start to the completion of the task, and E max represents the maximum energy of the node; It is expressed as the delay satisfaction of the task, Latency, the actual processing delay of the current task, T required is the latency requirement of the current task.

6. The green edge computing task offloading and resource optimization method according to claim 5, characterized in that: Also includes: Among them, E available is the residual energy of the node.

7. The green edge computing task offloading and resource optimization method according to claim 1 or 5, characterized in that: The task offloading target is globally optimized based on the Grey Wolf Algorithm, including: The size N of the gray wolf population and the maximum number of iterations M are set. The initial position of each gray wolf individual is a set of randomly generated task offloading schemes, indicating whether the task is processed locally, at the base station, or in the cloud. Calculate the fitness value of each individual gray wolf; In each iteration, the individuals in the population are sorted according to their fitness values, and the three best individuals are selected, marked as α, β, and δ, representing the best, second-best, and third-best task offloading solutions, respectively. The other gray wolf individuals are updated according to the positions of α, β, and δ; The iteration is stopped when the number of iterations reaches the maximum number of iterations M, or when the change rate of the optimal individual fitness value in two consecutive iterations is less than the change tolerance of the fitness value.

8. The green edge computing task offloading and resource optimization method according to claim 7, characterized in that: Also includes: The next position of each individual gray wolf is updated using the following formula: in, It is represented as the current task offloading scheme of individuals α, β, and δ, represents the position of the current individual, that is, the current position of the current gray wolf individual in the tth iteration; Represents the position of the current leader, which can be any one of α, β, and δ; The direction in which the gray wolf is currently moving and distance, is the intensity of the gray wolf's proximity to the leader,

Images