Material big data oriented computing, transmission resource and task joint allocation method

Abstract

The application provides a material big data oriented computing, transmission resource and task joint allocation method, and belongs to the technical field of material big data. The method comprises the following steps: modeling a material data collection and transmission process of a system structure composed of an internet of things layer, an edge layer and a cloud layer, to obtain a weighted cost model for material big data processing; introducing an adaptive dynamic inertia weight factor and a nonlinear convergence factor to improve a whale optimization algorithm, and solving the weighted cost model based on the improved whale optimization algorithm to obtain an optimal solution of a task allocation ratio and a computing resource and transmission resource allocation factor. The application can realize efficient and high-quality processing and transmission of material big data.

Description

Technical Field

[0001] This invention relates to the field of materials big data technology, and in particular to a method for the joint allocation of computing and transmission resources and tasks for materials big data. Background Technology

[0002] From personalized medicine to energy production and storage, many challenges of the 21st century share a common theme: materials are at the heart of the solutions. Big data technology has become a hot topic in materials science due to its ability to accelerate the discovery and design of new, advanced materials. In recent years, the Materials Genome Initiative (MGI) has been promoting the availability of big data in materials science, aiming to accelerate the discovery, development, manufacturing, and use of advanced materials, thereby fostering a resurgence in manufacturing.

[0003] By designing a materials big data sharing network based on caching and edge computing, and establishing a dedicated database platform for materials genome engineering, the research objective of materials big data aims to efficiently develop new materials and technologies, accelerate the industrial application of new materials, and fully utilize massive materials data resources. How to achieve efficient management and utilization of materials big data is one of the important issues in the materials genome engineering research process. Therefore, it is necessary to study an efficient collection and transmission technology for materials big data to accelerate the research and development cycle of materials and reduce the research and development costs of new materials. Summary of the Invention

[0004] This invention provides a method for jointly allocating computing and transmission resources and tasks for big data in materials, enabling high-efficiency and high-quality processing and transmission of such data. The technical solution is as follows:

[0005] This invention provides a method for jointly allocating computing and transmission resources and tasks for big data in materials, including:

[0006] A model is created for the material data collection and transmission process in a system structure consisting of an IoT layer, an edge layer, and a cloud layer, resulting in a weighted cost model for materials big data processing.

[0007] An adaptive dynamic inertia weighting factor and a nonlinear convergence factor are introduced to improve the whale optimization algorithm. Based on the improved whale optimization algorithm, the weighted cost model is solved to obtain the optimal solution of task allocation ratio and allocation factors of computing resources and transmission resources.

[0008] Furthermore, the weighted cost model is expressed as:

[0009]

[0010]

[0011]

[0012]

[0013]

[0014]

[0015]

[0016]

[0017]

[0018]

[0019]

[0020] Where Ψ represents the fitness value, α represents the task allocation ratio, and C and D represent the computing resource and transmission resource allocation factors, respectively; ψ r ψ ε These represent the control parameters for system delay and system energy consumption, respectively; ω is the system delay, and E is the system energy consumption. This represents the task allocation ratio of the terminal device n connected to the edge device m; This represents the task allocation ratio of edge device m to data from terminal device n; The rate at which the terminal device n, connected to the edge device m, acquires material data; r represents the data compression rate. These represent the computing capacity and transmission capacity of the terminal device n connected to the edge device m per unit time, respectively. This represents the transmission capacity of edge device m; This represents the rate at which data from terminal device n reaches edge device m; This represents the amount of processed data transmitted from terminal device n to edge device m; These represent the computing capacity and transmission capacity of the edge device m for data from the terminal device n, respectively. This represents the rate at which data from edge device m, connected to terminal device n, reaches the cloud center. C represents the computing capacity of the cloud center for data from edge device m; CC D represents the total computing capacity of the cloud center; CC M represents the total transmission capacity of the cloud center; N represents the number of edge devices; m and n represent the number of terminal devices; and C1 to C10 represent 10 constraints.

[0021] Furthermore, the system delay ω is expressed as:

[0022]

[0023] in, This represents the task allocation ratio of the cloud center for data from edge device m connected to terminal device n.

[0024] Furthermore, the system energy consumption E is expressed as:

[0025]

[0026] in, This represents the computational power of the terminal device n connected to the edge device m; This represents the transmission power of the terminal device n connected to the edge device m; This represents the computed power of edge device m; This represents the transmission power of the edge device m; This indicates the computing power of the cloud center; This represents the data processing time of terminal device n connected to edge device m; This represents the time it takes for data to be transmitted from terminal device n to edge device m; This represents the data processing time of edge device m; This indicates the time it takes for data to be transmitted from the edge device m to the cloud center; This indicates the time taken for data processing in the cloud center.

[0027] Furthermore, C1 is a constraint on the task allocation ratio between the IoT layer and the edge layer;

[0028] C2 is the constraint on the computing capacity of the Internet of Things (IoT) layer, and C3 and C4 are the constraints on the transmission capacity of the IoT layer.

[0029] C5 and C6 are constraints on the computing capacity of the edge layer, and C7 and C8 are constraints on the transmission capacity of the edge layer.

[0030] C9 and C10 are the computational capacity constraints for the cloud layer.

[0031] Furthermore, the whale optimization algorithm is improved by introducing an adaptive dynamic inertia weight factor and a nonlinear convergence factor. Based on the improved whale optimization algorithm, the weighted cost model is solved to obtain the optimal solution for task allocation ratios and the allocation factors of computing and transmission resources, including:

[0032] A1. Initialize the task allocation ratio and the computing and transmission resource allocation factors α, C, and D in the weighted cost model, and the maximum number of iterations k. max and the number of search agents;

[0033] A2, Calculate the initial fitness value Ψ based on the initialized parameters. 0 Humpback whales enter the stages of surrounding prey, using bubble nets to attack, and searching for prey. The task allocation ratio, computing resources, and transmission resource allocation factors in the weighted cost model are equivalent to the position vector of the humpback whale. Where k represents the current iteration number;

[0034] A3, Encircle the Prey Phase: During the algorithm iteration process, when Assuming the current optimal search agent is the target prey, the humpback whale continuously updates its position during algorithm iterations to search for the best search agent, thereby obtaining the next position vector. in, Represents the coefficient vector. Let [a] be a random vector between [0, 1]. Represents the nonlinear convergence factor.

[0035]

[0036] A4, Bubble Web Attack Phase: Establish a spiral equation between the prey's position and the humpback whale's position to simulate the humpback whale's spiral movement, thereby obtaining the next position vector.

[0037] A5, Prey Search Phase: During the algorithm iteration process, when The search agent's position during the exploration phase is updated based on the randomly selected search agent, thus obtaining the next position vector.

[0038] A6. Calculate the fitness value Ψ of each whale individual in the population, select the position of the individual with the smallest fitness value as the optimal position, and update the position vector.

[0039] A7. Let k = k + 1, and repeat steps A3 to A6 until the number of iterations reaches the maximum. Output the optimal solution for task allocation ratio and allocation factors for computing resources and transmission resources.

[0040] Furthermore, the behavior of step A3 is represented as follows:

[0041]

[0042]

[0043] in, This indicates the distance between the optimal search agent and the target prey. Represents the coefficient vector. This represents the optimal search proxy position, and k represents the current iteration number. Let w represent the position vector of the humpback whale, and w represent the adaptive dynamic inertia weighting factor.

[0044] Furthermore, the adaptive dynamic inertia weight factor w is expressed as:

[0045] w(k) = rand·w min ·sin(πk / 2k max )+w max ·(1-sin(πk / 2k max ))

[0046] Where rand represents a random number between [0,1], and w min and w max These represent the minimum and maximum weight factors, respectively.

[0047] Furthermore, the behavior of step A4 is represented as follows:

[0048]

[0049]

[0050] in, Indicates the current position Distance between and the best search agent location This represents the optimal search proxy position, and k represents the current iteration number. Let w represent the position vector of the humpback whale, w represent the adaptive dynamic weighting factor, ξ is a constant used to define the shape of the logarithmic spiral, and b is a random number between [-1, 1].

[0051] Furthermore, the behavior of step A5 is represented as follows:

[0052]

[0053]

[0054] in, This indicates the distance between the optimal search agent and the target prey. Represents the coefficient vector. Let k represent the random position vector of the reference whale, and k represent the current iteration number. Let w represent the position vector of the humpback whale, and w represent the adaptive dynamic weighting factor.

[0055] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0056] In this embodiment of the invention, the material data collection and transmission process of a system structure composed of an IoT layer, an edge layer, and a cloud layer is modeled to obtain a weighted cost model for material big data processing. An adaptive dynamic inertia weight factor and a nonlinear convergence factor are introduced to improve the whale optimization algorithm. Based on the improved whale optimization algorithm, the weighted cost model is solved to obtain the optimal solution of task allocation ratio and computing resource and transmission resource allocation factors. This reduces the system latency and energy consumption caused by long-distance material big data collection and transmission, and achieves high-efficiency and high-quality processing and transmission of material big data, so as to facilitate further analysis, modeling, and visualization research of material data. Attached Figure Description

[0057] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0058] Figure 1 This is a flowchart illustrating the method for jointly allocating computing and transmission resources and tasks for materials big data, as provided in an embodiment of the present invention.

[0059] Figure 2 This is a schematic diagram of the system structure provided in an embodiment of the present invention;

[0060] Figure 3 This is a schematic diagram of the process for finding the optimal solution of task allocation ratio and computational and transmission resource allocation factors provided in an embodiment of the present invention. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0062] like Figure 1 As shown, this embodiment of the invention provides a method for jointly allocating computing and transmission resources and tasks for big data in materials, including:

[0063] S101, Model the material data collection and transmission process of the system structure composed of the Internet of Things layer, edge layer and cloud layer, and obtain a weighted cost model for material big data processing;

[0064] like Figure 2 As shown, in this embodiment, the system structure includes an Internet of Things (IoT) layer, an edge layer, and a cloud layer. The edge layer can be expanded to multiple layers. This embodiment uses one layer as an example for explanation.

[0065] In this embodiment, a network comprising one cloud center, M edge devices, and N terminal devices is considered. Each node in the network has a certain computing capability. Raw material data is collected by the terminal devices, and the material data processing results from the edge devices are finally transmitted to the cloud center. Each terminal device can connect to at most one edge device, and each edge device can connect to at most one cloud center.

[0066] In this embodiment, to minimize system latency and device power consumption, the weighted cost model is expressed as follows:

[0067] P

[0068] C1

[0069] C2

[0070] C3

[0071] C4

[0072] C5

[0073] C6

[0074] C7

[0075] C8

[0076] C9

[0077] C10

[0078] Where Ψ represents the fitness value, α represents the task allocation ratio, and C and D represent the computing resource and transmission resource allocation factors, respectively; ψ r ψ ε These represent the control parameters for system delay and system energy consumption, respectively. The weighted cost model can be adjusted according to actual needs using these parameters; ω is the system delay, and E is the system energy consumption. This represents the task allocation ratio of the terminal device n connected to the edge device m; This represents the task allocation ratio of edge device m to data from terminal device n; The rate at which the terminal device n, connected to the edge device m, acquires material data; r represents the data compression rate. These represent the computing capacity and transmission capacity of the terminal device n connected to the edge device m per unit time, respectively. This represents the transmission capacity of edge device m; This represents the rate at which data from terminal device n reaches edge device m; This represents the amount of processed data transmitted from terminal device n to edge device m; These represent the computing capacity and transmission capacity of the edge device m for data from the terminal device n, respectively. This represents the rate at which data from edge device m, connected to terminal device n, reaches the cloud center. C represents the computing capacity of the cloud center for data from edge device m; CC D represents the total computing capacity of the cloud center; CC M represents the total transmission capacity of the cloud center; N represents the number of edge devices; m and n represent the number of terminal devices; and C1 to C10 represent 10 constraints.

[0079] In this embodiment, the system latency ω is the total latency of the data collected on the IoT layer per unit time, which can be expressed as:

[0080]

[0081] in, This represents the task allocation ratio of the cloud center for data from edge device m connected to terminal device n.

[0082] In this embodiment, the system energy consumption E is expressed as:

[0083]

[0084] in, This represents the computational power of the terminal device n connected to the edge device m; This represents the transmission power of the terminal device n connected to the edge device m; This represents the computed power of edge device m; This represents the transmission power of the edge device m; This indicates the computing power of the cloud center; This represents the data processing time of terminal device n connected to edge device m; This represents the time it takes for data to be transmitted from terminal device n to edge device m; This represents the data processing time of edge device m; This indicates the time it takes for data to be transmitted from the edge device m to the cloud center; This indicates the time taken for data processing in the cloud center.

[0085] In this embodiment, the optimization problem of task allocation ratio and computing and transmission resource allocation factors is considered under non-blocking conditions, with 10 constraints C1 to C10, among which...

[0086] C1 is a constraint on the task allocation ratio between the IoT layer and the edge layer;

[0087] C2 is the constraint on the computing capacity of the Internet of Things (IoT) layer, and C3 and C4 are the constraints on the transmission capacity of the IoT layer.

[0088] C5 and C6 are constraints on the computing capacity of the edge layer, and C7 and C8 are constraints on the transmission capacity of the edge layer.

[0089] C9 and C10 are the computational capacity constraints for the cloud layer.

[0090] S102, an adaptive dynamic inertia weight factor and a nonlinear convergence factor are introduced to improve the whale optimization algorithm. Based on the improved whale optimization algorithm, the weighted cost model is solved to obtain the optimal solution of task allocation ratio and allocation factors of computing resources and transmission resources.

[0091] In the context of collecting and transmitting big data on materials, this embodiment first proposes a new method for jointly allocating computing and transmission resources and tasks. Since the optimization problem is a non-convex optimization problem, the improved whale optimization algorithm is used to solve it, and finally an approximate optimal solution to the problem is obtained.

[0092] The whale optimization algorithm, based on the hunting behavior of whales, is a metaheuristic algorithm based on swarm intelligence. This algorithm searches for the optimal solution by simulating the hunting behavior of humpback whales and includes three search mechanisms:

[0093] I. The stage of surrounding the prey

[0094] In the whale optimization algorithm, assuming the current best search agent is the target prey, the humpback whale continuously updates its position during algorithm iterations to search for the best search agent. In this case, the task allocation ratio, computing resources, and transmission resource allocation factors in the weighted cost model can be represented as the humpback whale's position. This behavior can be mathematically expressed as:

[0095]

[0096]

[0097] in, It is a coefficient vector, where k is the current iteration number. It is the best search proxy location. This represents the position vector of the whale. The distance between the optimal search agent and the target prey is represented by w; w represents the adaptive dynamic weighting factor.

[0098] To avoid premature convergence of the algorithm, in this embodiment, when solving for the optimal task allocation ratio, computational resources, and transmission resource allocation factors for material data processing, an adaptive dynamic inertial weight factor w(k) is introduced into the position update formula based on the idea of ​​inertial weight guiding population optimization in particle swarm optimization, thereby improving the optimization accuracy of the algorithm. The formula is expressed as:

[0099] w(k) = rand·w min ·sin(πk / 2k max )+w max ·(1-sin(πk / 2k max ))

[0100] Where rand represents a random number between [0,1], and w min and w max These represent the minimum and maximum weight factors, respectively;

[0101] coefficient vector Represented as:

[0102]

[0103]

[0104] in, It is a random vector between [0,1]. is the convergence factor.

[0105] During the iteration process, the convergence factor of the traditional whale optimization algorithm The vector decreases linearly from 2 to 0, which makes the algorithm's convergence speed too slow and unsuitable for practical applications. To address this problem, this invention proposes a nonlinear convergence factor in the process of solving for the transmission resources, computing resources, and task allocation ratio for big data in materials science. The formula is expressed as:

[0106]

[0107] II. Bubble Web Attack Phase

[0108] In solving the optimal solution of the weighted cost model, the bubble web attack phase employs both a shrinking encirclement mechanism and a spiral position update to model the bubble web behavior of humpback whales; the shrinking encirclement mechanism is implemented by setting a coefficient vector. This will allow the new location to be situated between the current location and the location of the best search agent.

[0109] To simulate the spiral motion of a humpback whale, the spiral equation relating the prey's position and the whale's position is expressed as:

[0110]

[0111]

[0112] Where ξ is a constant used to define the shape of the logarithmic spiral; b is a random number between [-1, 1].

[0113] III. Hunting for Prey

[0114] The prey-hunting mechanism updates the search agent's position during the exploration phase based on a randomly selected search agent, rather than the best search agent so far. This behavior can be defined by the following formula:

[0115]

[0116]

[0117] in, This represents the random position vector of the whale used as a reference.

[0118] In this embodiment, the optimization variables are defined as the task allocation ratio during material data processing and the allocation factors α, C, and D for computing and transmission resources. It can be represented as:

[0119]

[0120] Based on the above, this embodiment proposes a new method for jointly allocating computing and transmission resources and tasks. After a certain number of iterations, the weighted cost of the system reaches near-optimal.

[0121] like Figure 3 As shown, the process of finding the optimal solution for task allocation ratio and resource allocation in material data processing using the improved whale algorithm in this embodiment includes:

[0122] A1. Initialize the task allocation ratio and the computing and transmission resource allocation factors α, C, and D in the weighted cost model, and the maximum number of iterations k. max And the number of search agents, etc.;

[0123] A2, Calculate the initial fitness value Ψ based on the initialized parameters. 0 Humpback whales enter the stages of surrounding prey, using bubble nets to attack, and searching for prey. The task allocation ratio, computing resources, and transmission resource allocation factors in the weighted cost model are equivalent to the position vector of the humpback whale. Where k represents the current iteration number;

[0124] A3, the prey-surrounding phase. Assuming the current best search agent is the target prey, the humpback whale continuously updates its position during algorithm iterations to search for the best search agent. That is, it calculates the next position vector according to the first strategy.

[0125] A4, the bubble-web attack phase. A spiral equation is established between the prey's position and the whale's position to simulate the humpback whale's spiral motion. That is, the next position vector is calculated according to the second strategy.

[0126] A5, the prey-hunting phase. The search agent's position during the exploration phase is updated based on the randomly selected search agent. That is, the next position vector is calculated according to the third strategy.

[0127] A6. Calculate the fitness value Ψ of each whale individual in the population, select the position of the individual with the smallest fitness value as the optimal position, and update the position vector.

[0128] A7. Let k = k + 1, and repeat steps A3 to A6 until the number of iterations reaches the maximum. Output the optimal solution for task allocation ratio and allocation factors for computing resources and transmission resources.

[0129] The first strategy involves: when solving for the minimum system delay and energy consumption during the material data collection and transmission process, the task allocation ratio, computing resources, and transmission resource allocation factor are equivalent to the position vector of a humpback whale; parameters are defined. According to the formula and Update parameters;

[0130] During the algorithm iteration process, when Assuming the current optimal search agent is the target prey, the humpback whale continuously updates its position during algorithm iterations to search for the best search agent, i.e., it searches for the optimal allocation factor for material data processing in the current iteration. This behavior is represented as:

[0131]

[0132]

[0133] The second strategy is to establish a spiral equation between the prey's position and the whale's position to simulate the spiral movement of a humpback whale; this behavior is represented as:

[0134]

[0135]

[0136] The third strategy is: during the algorithm iteration process, when Update the search agent's position during the exploration phase based on a randomly selected search agent. This behavior is represented as:

[0137]

[0138]

[0139] In this embodiment, edge computing technology is used to reduce system latency and energy consumption caused by long-distance data acquisition and transmission of materials. After establishing a weighted cost model based on the system structure, an adaptive dynamic inertia weight factor and a nonlinear convergence factor are introduced to improve the whale optimization algorithm, thereby improving the search accuracy in the later stages of iteration. Through continuous iteration, the optimal task allocation ratio and the allocation factors of computing and transmission resources are obtained, ultimately achieving efficient processing and transmission of materials big data, so as to facilitate further analysis, modeling, and visualization research of the materials data.

[0140] The invention's embodiment of the method for jointly allocating computing and transmission resources and tasks for materials big data models the materials data collection and transmission process of a system structure composed of an IoT layer, an edge layer, and a cloud layer, resulting in a weighted cost model for materials big data processing. An adaptive dynamic inertia weight factor and a nonlinear convergence factor are introduced to improve the whale optimization algorithm. Based on the improved whale optimization algorithm, the weighted cost model is solved to obtain the optimal solution for task allocation ratios and computing and transmission resource allocation factors. This reduces system latency and energy consumption caused by long-distance materials big data collection and transmission, achieving high-efficiency and high-quality processing and transmission of materials big data, facilitating further analysis, modeling, and visualization research of the materials data.

[0141] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for jointly allocating computing and transmission resources and tasks for big data in materials, characterized in that, include: A model is created for the material data collection and transmission process in a system structure consisting of an IoT layer, an edge layer, and a cloud layer, resulting in a weighted cost model for materials big data processing. An adaptive dynamic inertia weighting factor and a nonlinear convergence factor are introduced to improve the whale optimization algorithm. Based on the improved whale optimization algorithm, the weighted cost model is solved to obtain the optimal solution of task allocation ratio and allocation factors of computing resources and transmission resources. The weighted cost model is expressed as follows: Where Ψ represents the fitness value, α represents the task allocation ratio, and C and D represent the computing resource and transmission resource allocation factors, respectively; ψ r ψ ε These represent the control parameters for system delay and system energy consumption, respectively; ω is the system delay, and E is the system energy consumption. This represents the task allocation ratio of the terminal device n connected to the edge device m; This represents the task allocation ratio of edge device m to data from terminal device n; The rate at which the terminal device n, connected to the edge device m, acquires material data; r represents the data compression rate. These represent the computing capacity and transmission capacity of the terminal device n connected to the edge device m per unit time, respectively. This represents the transmission capacity of edge device m; This represents the rate at which data from terminal device n reaches edge device m; This represents the amount of processed data transmitted from terminal device n to edge device m; These represent the computing capacity and transmission capacity of the edge device m for data from the terminal device n, respectively. This represents the rate at which data from edge device m, connected to terminal device n, reaches the cloud center. C represents the computing capacity of the cloud center for data from edge device m; CC D represents the total computing capacity of the cloud center; CC M represents the total transmission capacity of the cloud center; N represents the number of edge devices; m and n represent the number of terminal devices; m and n are the m-th edge device in the edge layer and the n-th terminal device in the Internet of Things layer, respectively; C1 to C10 represent 10 constraints. Wherein, the system delay ω is expressed as: in, This represents the task allocation ratio of the cloud center for data from edge device m connected to terminal device n; Wherein, the system energy consumption E is expressed as: in, This represents the computational power of the terminal device n connected to the edge device m; This represents the transmission power of the terminal device n connected to the edge device m; This represents the computed power of edge device m; This represents the transmission power of the edge device m; This indicates the computing power of the cloud center; This represents the data processing time of terminal device n connected to edge device m; This represents the time it takes for data to be transmitted from terminal device n to edge device m; This represents the data processing time of edge device m; This indicates the time it takes for data to be transmitted from the edge device m to the cloud center; This indicates the time taken for data processing in the cloud center.

2. The method for joint allocation of computing and transmission resources and tasks for materials big data according to claim 1, characterized in that, C1 is a constraint on the task allocation ratio between the IoT layer and the edge layer; C2 is the constraint on the computing capacity of the Internet of Things (IoT) layer, and C3 and C4 are the constraints on the transmission capacity of the IoT layer. C5 and C6 are constraints on the computing capacity of the edge layer, and C7 and C8 are constraints on the transmission capacity of the edge layer. C9 and C10 are the computational capacity constraints for the cloud layer.

3. The method for joint allocation of computing and transmission resources and tasks for materials big data according to claim 1, characterized in that, The whale optimization algorithm is improved by introducing an adaptive dynamic inertia weight factor and a nonlinear convergence factor. Based on the improved whale optimization algorithm, the weighted cost model is solved to obtain the optimal solution for task allocation ratio and allocation factors of computing resources and transmission resources, including: A1, Initialize the task allocation ratio and computing resource and transmission resource allocation factors α, C, and D in the weighted cost model, and the maximum number of iterations k. max and the number of search agents; A2, Calculate the initial fitness value Ψ based on the initialized parameters. 0 Humpback whales enter the stages of surrounding prey, using bubble nets to attack, and searching for prey. The task allocation ratio, computing resources, and transmission resource allocation factors in the weighted cost model are equivalent to the position vector of the humpback whale. Where k represents the current iteration number; A3, Encircle the Prey Phase: During the algorithm iteration process, when Assuming the current optimal search agent is the target prey, the humpback whale continuously updates its position during algorithm iterations to search for the best search agent, thereby obtaining the next position vector. in, Represents the coefficient vector. Let [a] be a random vector between [0, 1]. Represents the nonlinear convergence factor. A4, Bubble Web Attack Phase: Establish a spiral equation between the prey's position and the humpback whale's position to simulate the humpback whale's spiral movement, thereby obtaining the next position vector. A5, Prey Search Phase: During the algorithm iteration process, when The search agent's position during the exploration phase is updated based on the randomly selected search agent, thus obtaining the next position vector. A6. Calculate the fitness value Ψ of each whale individual in the population, select the position of the individual with the smallest fitness value as the optimal position, and update the position vector. A7. Let k = k + 1, and repeat steps A3 to A6 until the number of iterations reaches the maximum. Output the optimal solution for task allocation ratio and allocation factors for computing resources and transmission resources.

4. The method for joint allocation of computing and transmission resources and tasks for materials big data according to claim 3, characterized in that, The behavior of step A3 is represented as follows: in, This indicates the distance between the optimal search agent and the target prey. Represents the coefficient vector. This represents the optimal search proxy position, and k represents the current iteration number. Let w represent the position vector of the humpback whale, and w represent the adaptive dynamic inertia weighting factor.

5. The method for joint allocation of computing and transmission resources and tasks for materials big data according to claim 4, characterized in that, The adaptive dynamic inertia weighting factor w is expressed as: w(k)=rand·w min ·sin(πk / 2k max )+in max (1-sin(πk / 2k max )) Where rand represents a random number between [0,1], and w min and w max These represent the minimum and maximum weight factors, respectively.

6. The method for joint allocation of computing and transmission resources and tasks for materials big data according to claim 3, characterized in that, The behavior of step A4 is represented as follows: in, Indicates the current position Distance between and the best search agent location This represents the optimal search proxy position, and k represents the current iteration number. Let w represent the position vector of the humpback whale, w represent the adaptive dynamic weighting factor, ξ is a constant used to define the shape of the logarithmic spiral, and b is a random number between [-1, 1].

7. The method for joint allocation of computing and transmission resources and tasks for materials big data according to claim 3, characterized in that, The behavior of step A5 is represented as follows: in, This indicates the distance between the optimal search agent and the target prey. Represents the coefficient vector. Let k represent the random position vector of the reference whale, and k represent the current iteration number. Let w represent the position vector of the humpback whale, and w represent the adaptive dynamic weighting factor.

Images