Cloud task scheduling method and device, storage medium and program product

Abstract

The invention provides a cloud task scheduling method and device, a storage medium and a program product, and relates to the field of financial science and tech, task scheduling equipment packages a service request of a network server into a cloud task, constructs a target optimization model with minimum completion time as a target, utilizes a randomly initialized dung beetle population model and iteratively optimizes, and finally optimizes the cloud task. And finally, an efficient cloud task scheduling scheme is generated, so that the problems of unbalanced and insufficient server resource utilization under the scene of task complexity and server performance difference are solved, and the resource utilization rate and the processing efficiency of a server cluster are effectively improved.

Description

Technical Field

[0001] The present application relates to the field of financial technology, and in particular to a cloud task scheduling method, device, storage medium and program product. Background Art

[0002] As banks' digital transformation deepens, the limitations of traditional centralized mainframe architectures are becoming increasingly apparent. Distributed microservices architectures, with their flexibility, scalability, and high availability, are becoming the preferred solution for upgrading core banking systems. In this process, cloud task scheduling, as a key component of distributed systems, plays a key role in orchestrating business processes, optimizing resources, and ensuring stability.

[0003] In the prior art, a hash algorithm is usually used to schedule cloud tasks. The cloud tasks are hashed through a hash function to obtain the corresponding hash value, so as to determine the server in the server cluster corresponding to the hash value, and then the cloud tasks are assigned to the server cluster.

[0004] However, since the completion time of a cloud task is closely related to the complexity of the cloud task and the performance of the server cluster, when the task complexity is different and the performance of the servers in the server cluster is uneven, the above-mentioned existing technologies lack awareness of the dynamic server load, which can easily lead to waste of computing resources and reduce the overall performance of the server cluster. Summary of the Invention

[0005] The present application provides a cloud task scheduling method, device, storage medium and program product, which solves the problem of unbalanced and insufficient server resource utilization in scenarios with different task complexity and server performance, and effectively improves the resource utilization and processing efficiency of the server cluster.

[0006] In a first aspect, an embodiment of the present application provides a cloud task scheduling method, which is applied to a task scheduling module, including:

[0007] Obtain multiple service requests sent by the network server and package the multiple service requests into multiple cloud tasks;

[0008] Based on the minimum completion time information of the server cluster, a target optimization model is constructed;

[0009] Based on the preset number of dung beetles and the total number of multiple cloud tasks, an initial dung beetle population model is constructed;

[0010] Based on the target optimization model, the initial dung beetle population model is iteratively optimized until the number of iterations is equal to the preset number of iterations to obtain the final dung beetle population model;

[0011] Based on the final dung beetle population model, determine the cloud task scheduling plan;

[0012] Based on the cloud task scheduling solution, multiple cloud tasks are assigned to the server cluster.

[0013] In one possible implementation, based on the target optimization model, the initial dung beetle population model is iteratively optimized until the number of iterations is equal to a preset number of iterations to obtain a final dung beetle population model, including:

[0014] The initial dung beetle population model is divided into four sub-dung beetle population models, and four dung beetle behaviors of ball rolling, foraging, egg laying, and stealing are simulated respectively. The four sub-dung beetle population models are updated to obtain four updated sub-dung beetle population models, and the four updated sub-dung beetle population models are merged to obtain a first dung beetle population model, wherein the dung beetle population model includes a preset number of dung beetle individuals.

[0015] Determining a second dung beetle population model based on the target optimization model, the initial dung beetle population model, and the first dung beetle population model, and determining the initial dung beetle population model for the next iterative optimization as the second dung beetle population model;

[0016] The next iterative optimization is performed until the number of iterations is equal to the preset number of iterations, and the final dung beetle population model is determined to be the second dung beetle population model corresponding to the last iterative optimization.

[0017] In one possible implementation, an initial dung beetle population model is divided into four sub-dung beetle population models, and four dung beetle behaviors of ball rolling, foraging, egg laying, and stealing are simulated respectively. The four sub-dung beetle population models are updated to obtain four updated sub-dung beetle population models, and the four updated sub-dung beetle population models are merged to obtain a first dung beetle population model, including:

[0018] A nonlinear boundary selection algorithm is used to determine the preset upper and lower boundary ranges, and boundary processing is performed on the dung beetle individuals that exceed the preset upper and lower boundary ranges in the four updated sub-dung beetle population models to obtain four sub-dung beetle population models after boundary processing, and the four sub-dung beetle population models after boundary processing are merged to obtain the first dung beetle population model.

[0019] In one possible implementation, determining the second dung beetle population model based on the target optimization model, the initial dung beetle population model, and the first dung beetle population model includes:

[0020] Based on the initial dung beetle population model and the first dung beetle population model, a third dung beetle population model is determined;

[0021] The third dung beetle population model is updated using a preset nonlinear Cauchy-Gaussian model to obtain the fourth dung beetle population model;

[0022] A preset greedy algorithm was used to screen the better dung beetle individuals in the third and fourth dung beetle population models to determine the second dung beetle population model.

[0023] In one possible implementation, an initial dung beetle population model is divided into four sub-dung beetle population models, and four dung beetle behaviors of ball rolling, foraging, egg laying, and stealing are simulated respectively. The four sub-dung beetle population models are updated to obtain four updated sub-dung beetle population models, and the four updated sub-dung beetle population models are merged to obtain a first dung beetle population model, including:

[0024] Based on the updated dung beetle population model corresponding to the rolling ball dung beetle behavior, a reverse learning dung beetle population model is generated through a reverse learning algorithm;

[0025] Using a preset elite selection model, the updated sub-dung beetle population model corresponding to the rolling ball dung beetle behavior and the reverse learning sub-dung beetle population model were screened for better dung beetle individuals to determine the better sub-dung beetle population model.

[0026] The better sub-dung beetle population model is merged with the remaining three updated sub-dung beetle population models to obtain the first dung beetle population model.

[0027] In one possible implementation, the initial dung beetle population model is divided into four sub-dung beetle population models, and four dung beetle behaviors, namely, ball rolling, foraging, egg laying, and stealing, are simulated respectively. The four sub-dung beetle population models are updated to obtain the updated four sub-dung beetle population models, including:

[0028] Based on the preset natural coefficient, the preset deflection coefficient and the preset light intensity, the behavior of the rolling ball dung beetle is simulated, and the dung beetle sub-population model is updated to obtain an updated dung beetle sub-population model corresponding to the rolling ball dung beetle behavior.

[0029] In one possible implementation, the initial dung beetle population model is divided into four sub-dung beetle population models, and four dung beetle behaviors, namely, ball rolling, foraging, egg laying, and stealing, are simulated respectively. The four sub-dung beetle population models are updated to obtain the updated four sub-dung beetle population models, including:

[0030] Based on the first random vector, the second random vector and the preset foraging boundary range, the foraging behavior of the dung beetle is simulated, and the dung beetle sub-population model is updated to obtain an updated dung beetle sub-population model corresponding to the foraging behavior of the dung beetle.

[0031] In one possible implementation, the initial dung beetle population model is divided into four sub-dung beetle population models, and four dung beetle behaviors, namely, ball rolling, foraging, egg laying, and stealing, are simulated respectively. The four sub-dung beetle population models are updated to obtain the updated four sub-dung beetle population models, including:

[0032] Based on the third random vector, the fourth random vector and the preset egg-laying boundary range, the egg-laying dung beetle behavior is simulated, and the daughter dung beetle population model is updated to obtain an updated daughter dung beetle population model corresponding to the egg-laying dung beetle behavior.

[0033] In one possible implementation, the initial dung beetle population model is divided into four sub-dung beetle population models, and four dung beetle behaviors, namely, ball rolling, foraging, egg laying, and stealing, are simulated respectively. The four sub-dung beetle population models are updated to obtain the updated four sub-dung beetle population models, including:

[0034] Based on the fifth random vector and the first preset value, the behavior of the stealing dung beetle is simulated, and the dung beetle sub-population model is updated to obtain an updated dung beetle sub-population model corresponding to the stealing dung beetle behavior.

[0035] In one possible implementation, a target optimization model is constructed based on the minimum completion time information of the server cluster, including:

[0036] A target optimization model is constructed based on the minimum completion time information of the server cluster, the mean square error of the minimum completion time information of the server cluster, the first weight value and the second weight value.

[0037] In a second aspect, an embodiment of the present application provides a cloud task scheduling device, comprising:

[0038] An acquisition module is used to acquire multiple service requests sent by the network server and package the multiple service requests into multiple cloud tasks;

[0039] A construction module is used to build a target optimization model based on the minimum completion time information of the server cluster;

[0040] The construction module is also used to construct an initial dung beetle population model based on a preset number of dung beetles and the total number of multiple cloud tasks;

[0041] a processing module, configured to iteratively optimize the initial dung beetle population model based on the target optimization model until the number of iterations is equal to a preset number of iterations, so as to obtain a final dung beetle population model;

[0042] The processing module is also used to determine the cloud task scheduling scheme based on the final dung beetle population model;

[0043] The scheduling module is used to allocate multiple cloud tasks to the server cluster based on the cloud task scheduling solution.

[0044] In a third aspect, an embodiment of the present application provides a task scheduling device, comprising: a memory, a processor;

[0045] Memory stores computer-executable instructions;

[0046] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementations of the first aspect.

[0047] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the first aspect above and / or various possible implementation methods of the first aspect.

[0048] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above first aspect and / or various possible implementation methods of the first aspect.

[0049] The cloud task scheduling method, device, storage medium and program product provided by the present application, the task scheduling device packages the service request of the network server into a cloud task, constructs a target optimization model with the minimum completion time as the goal, uses a randomly initialized dung beetle population model and iterative optimization, and finally generates an efficient cloud task scheduling solution, which solves the problem of unbalanced and insufficient server resource utilization in scenarios with different task complexity and server performance, and effectively improves the resource utilization and processing efficiency of the server cluster. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0051] Figure 1 A schematic diagram of a cloud task scheduling scenario provided in an embodiment of the present application;

[0052] Figure 2 Schematic diagram of the process of cloud task scheduling method provided in the embodiment of this application Figure 1 ;

[0053] Figure 3 Schematic diagram of the process of cloud task scheduling method provided in the embodiment of this application Figure 2 ;

[0054] Figure 4 A schematic diagram of the structure of a cloud task scheduling device provided in an embodiment of the present application;

[0055] Figure 5 A schematic diagram of the structure of the task scheduling device provided in an embodiment of the present application.

[0056] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0057] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0058] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data comply with the relevant laws, regulations and standards of the relevant regions, take necessary confidentiality measures, do not violate public order and good customs, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0059] In addition, this application involves big data analysis of user information (including but not limited to personal biometrics, identity data, consumption data, asset data, electronic terminal operation data, etc.), and the use of artificial intelligence technology for automated decision-making, and a technical solution for making decisions that have a significant impact on personal rights and interests based on the results of automated decision-making. The application provides users with corresponding operation entrances for users to choose to agree or reject the results of automated decision-making; if the user chooses to reject, the expert decision-making process will be entered.

[0060] It should be noted that the cloud task scheduling method, device, storage medium and program product provided in this application can be used in the field of financial technology, and can also be used in any field other than financial technology. The application field of the cloud task scheduling method, device, storage medium and program product in this application is not limited.

[0061] As banks' digital transformation deepens, the limitations of traditional centralized mainframe architectures are becoming increasingly apparent. Distributed microservices architectures, with their flexibility, scalability, and high availability, are becoming the preferred solution for upgrading core banking systems. In this process, cloud task scheduling, as a key component of distributed systems, plays a key role in orchestrating business processes, optimizing resources, and ensuring stability.

[0062] Distributed microservice architecture usually includes network servers, task scheduling devices and server clusters. Figure 1 A schematic diagram of a cloud task scheduling scenario provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the application scenario provided by the embodiment of the present application includes multiple client devices, network servers, task scheduling devices and server clusters.

[0063] The network server receives a large number of service requests sent by multiple client devices and forwards them to the task scheduling device. The task scheduling device packages the service requests into cloud tasks and distributes the large number of cloud tasks to multiple servers in the server cluster based on the preset cloud task scheduling algorithm.

[0064] In the prior art, a hash algorithm is usually used to schedule cloud tasks. The cloud tasks are hashed through a hash function to obtain the corresponding hash value, so as to determine the server in the server cluster corresponding to the hash value, and then the cloud tasks are assigned to the server cluster.

[0065] However, since the completion time of a cloud task is closely related to the complexity of the cloud task and the performance of the server cluster, when the task complexity is different and the performance of the servers in the server cluster is uneven, the above-mentioned existing technologies lack awareness of the dynamic server load, which can easily lead to waste of computing resources and reduce the overall performance of the server cluster.

[0066] In order to solve the above technical problems, the following technical concept is proposed: since the load balancing effect is poor when the cloud task complexity and server execution time are different in the existing technology, it is necessary to balance the server execution time and server load. The inventors thought of using the dung beetle optimization algorithm to dynamically iteratively optimize the server execution time and server load based on the dynamic behavior of dung beetles in finding the optimal path in the natural environment through rolling balls, foraging, laying eggs, stealing, etc., thereby improving the accuracy of cloud task allocation.

[0067] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0068] Figure 2 Schematic diagram of the process of cloud task scheduling method provided in the embodiment of this application Figure 1 ,like Figure 2 As shown, the method is applied to any task scheduling device, and the method includes:

[0069] S201. Acquire multiple service requests sent by a network server, and package the multiple service requests into multiple cloud tasks.

[0070] Specifically, the network server, as the entry point for client device requests, receives various service requests from different client devices. These service requests may involve different business functions such as account query, transaction processing, and data statistics. The network server needs to parse the received service requests and identify which requests require microservices (such as those involving complex business logic or requiring access to specific resources). These service requests requiring microservice support are assigned to the server cluster by the task scheduling device for further processing, while other simple service requests may be responded to directly by the network server.

[0071] For different service requests, the task scheduling module generates multiple task parameters to package the service requests into cloud tasks. The task parameters include one or more of the task type, task identifier, input data, priority, and dependency.

[0072] S202: Build a target optimization model based on the minimum completion time information of the server cluster.

[0073] If possible, based on the minimum completion time information of the server cluster, a meta-heuristic algorithm is used to build a target optimization model.

[0074] S203: constructing an initial dung beetle population model based on a preset number of dung beetles and the total number of multiple cloud tasks.

[0075] Specifically, the initial dung beetle population model is shown in the following formula (1):

[0076]

[0077] Where N is the preset number of dung beetles; d is the total number of multiple cloud tasks; and X is the initial dung beetle population model.

[0078] Specifically, each dung beetle individual in the initial dung beetle population model corresponds to an initial cloud task allocation plan.

[0079] S204 , based on the target optimization model, iteratively optimize the initial dung beetle population model until the number of iterations is equal to the preset number of iterations, so as to obtain the final dung beetle population model.

[0080] Possibly, based on the target optimization model, at least one of a gradient descent algorithm, a genetic algorithm, or a particle swarm optimization algorithm is used to iteratively optimize the initial dung beetle population model until the number of iterations is equal to a preset number of iterations to obtain a final dung beetle population model.

[0081] S205. Determine a cloud task scheduling solution based on the final dung beetle population model.

[0082] Specifically, in the final dung beetle population model, the target optimization model is used to calculate the fitness values ​​of all dung beetle individuals, and the cloud task scheduling plan is determined based on the dung beetle individual with the smallest fitness value and the mapping relationship between cloud tasks and server clusters.

[0083] S206: Allocate multiple cloud tasks to the server cluster based on the cloud task scheduling solution.

[0084] Specifically, the cloud task scheduling plan is parsed to generate a cloud task allocation list, and multiple cloud tasks are allocated to corresponding servers in the server cluster according to the cloud task allocation list.

[0085] The cloud task scheduling method provided in the embodiment of the present application is a task scheduling device that packages the service requests of the network server into cloud tasks, constructs a target optimization model with the minimum completion time as the goal, uses a randomly initialized dung beetle population model and iterative optimization, and finally generates an efficient cloud task scheduling solution, which solves the problem of unbalanced and insufficient server resource utilization in scenarios with different task complexity and server performance, and effectively improves the resource utilization and processing efficiency of the server cluster.

[0086] Figure 3 Schematic diagram of the process of cloud task scheduling method provided in the embodiment of this application Figure 2 ,like Figure 3 As shown, the method includes:

[0087] S301: Acquire multiple service requests sent by a network server, and package the multiple service requests into multiple cloud tasks.

[0088] It should be noted that the specific implementation of S301 can refer to the specific implementation of S201, and will not be described in detail here.

[0089] S302: Build a target optimization model based on the minimum completion time information of the server cluster.

[0090] Specifically, the minimum completion time information of the server cluster is the longest time for all servers in the server cluster to complete processing the cloud task, as shown in the following formula (2), including:

[0091] makespan=Max(Ti)(2)

[0092] Among them, makespan is the minimum completion time information of the server cluster; Ti is the time it takes for the i-th server to complete the cloud task.

[0093] The mean square error of the minimum completion time information of the server cluster is calculated, and the mean square error of the minimum completion time information of the server cluster is used to indicate the balance of the server load.

[0094] The linear weighted method is used to integrate the minimum completion time information of the server cluster and the mean square error of the minimum completion time information of the server cluster to construct the target optimization model, as shown in the following formula (3):

[0095]

[0096] Among them, fitness is the target optimization model; makespan is the minimum completion time information of the server cluster; σ is the mean square error of the minimum completion time information of the server cluster; w1 is the first weight value; w2 is the second weight value.

[0097] It is possible to balance the minimum completion time and load balancing of the server cluster by adjusting the first weight value and the second weight value. When the first weight value is greater than the second weight value, the minimum completion time of the server cluster can be accelerated. When the second weight value is greater than the first weight value, the load of the server cluster can be balanced and the performance utilization of the server cluster can be improved.

[0098] It is possible to construct constraints for cloud task allocation, where each cloud task can only be assigned to one server in the server cluster.

[0099] S303: constructing an initial dung beetle population model based on a preset number of dung beetles and the total number of multiple cloud tasks.

[0100] S304: Based on the target optimization model, the initial dung beetle population model is iteratively optimized until the number of iterations is equal to the preset number of iterations, so as to obtain the final dung beetle population model.

[0101] Specifically, S304 includes S3041, S3042 and S3043.

[0102] S3041. Divide the initial dung beetle population model into four sub-dung beetle population models, and simulate four dung beetle behaviors of rolling a ball, foraging, laying eggs, and stealing respectively. Update the four sub-dung beetle population models to obtain four updated sub-dung beetle population models, and merge the four updated sub-dung beetle population models to obtain a first dung beetle population model. The dung beetle population model includes a preset number of dung beetle individuals.

[0103] Specifically, the initial dung beetle population model is divided into four sub-dung beetle population models, and based on the feces source density, competition intensity and temperature difference data, the four dung beetle behaviors of rolling, foraging, laying eggs and stealing are simulated respectively, and the four sub-dung beetle population models are updated to obtain four updated sub-dung beetle population models. The four updated sub-dung beetle population models are merged, and the surviving dung beetle individuals in the above four updated sub-dung beetle population models are screened to obtain the first dung beetle population model.

[0104] In a possible implementation, based on a preset natural coefficient, a preset deflection coefficient, and a preset light intensity, the behavior of the ball-rolling dung beetle is simulated, and the sub-dung beetle population model is updated to obtain an updated sub-dung beetle population model corresponding to the ball-rolling dung beetle behavior.

[0105] Specifically, it is shown in the following formula (4):

[0106] x i (t+1)=x i (t)+α×k×x i (t-1)+b×Δx(4)

[0107] Δx=|x i (t)-X w |

[0108] Among them, x i (t) represents the position information of the i-th dung beetle individual in the t-th iteration of the dung beetle population model; x i (t+1) represents the position information of the i-th dung beetle in the updated sub-dung beetle population model; k∈(0,0.2] represents the deflection coefficient; b∈(0,1) is a constant; α is a natural coefficient, which simulates the deviation of the movement direction of the rolling dung beetle due to natural factors such as wind and uneven ground. When α=1, there is no deviation, and when α=-1, it deviates from the original direction; X w Indicates the current global worst position; Δx simulates the change in light intensity. The larger it is, the weaker the light is. Light intensity will affect the movement path of the dung beetle.

[0109] In one possible implementation, based on the updated sub-dung beetle population model corresponding to the rolling ball dung beetle behavior, a reverse-learned sub-dung beetle population model is generated through a reverse learning algorithm; a preset elite selection model is used to screen the better dung beetle individuals in the updated sub-dung beetle population model corresponding to the rolling ball dung beetle behavior and the reverse-learned sub-dung beetle population model to determine a better sub-dung beetle population model; the better sub-dung beetle population model is merged with the remaining three updated sub-dung beetle population models to obtain a first dung beetle population model.

[0110] Specifically, based on the updated sub-dung beetle population model corresponding to the rolling ball dung beetle behavior, a reverse learning sub-dung beetle population model is generated through a reverse learning algorithm as shown in the following formula (5). A preset elite selection model is used to screen the better dung beetle individuals in the updated sub-dung beetle population model corresponding to the rolling ball dung beetle behavior and the reverse learning sub-dung beetle population model to determine a better sub-dung beetle population model as shown in the following formula (6), including:

[0111] x OBL =k×(Ub+Lb)-x i (t+1)(5)

[0112]

[0113] Among them, x OBL is the i-th dung beetle individual in the reverse learning dung beetle population model; x i (t+1) represents the position information of the i-th dung beetle in the updated dung beetle population model; k is a random vector of size 1×d that obeys the normal distribution; Lb and Ub represent the lower and upper bounds of the boundary space respectively; f(x i (t+1)) is the fitness value of the i-th dung beetle individual in the updated dung beetle population model; f(x OBL ) is the fitness value of the i-th dung beetle individual in the reverse learning dung beetle population model.

[0114] It is possible that during the process of simulating the behavior of the ball-rolling dung beetle, there are obstacles, and the simulated ball-rolling dung beetle cannot distinguish the direction of movement. The direction of movement can be determined by simulating the behavior of the dancing dung beetle, as shown in the following formula (7):

[0115] x i (t+1)=x i (t)+tan(θ)|x i (t)-x i (t-1)|(7)

[0116] Among them, x i (t) is the i-th dung beetle individual in the t-th iteration of the dung beetle population model; x i (t-1) is the i-th dung beetle individual in the t-1-th iteration of the dung beetle population model; x i (t+1) is the i-th dung beetle individual in the t+1-th iteration of the dung beetle population model; θ∈(0,π).

[0117] In a possible implementation, based on the first random vector, the second random vector and a preset foraging boundary range, the foraging behavior of dung beetles is simulated, and the sub-dung beetle population model is updated to obtain an updated sub-dung beetle population model corresponding to the foraging behavior of dung beetles.

[0118] Specifically, it is shown in the following formula (8):

[0119] x i (t+1)=x i (t)+C1×(x i (t)-Lb b )+C2×(x i (t)-Ub b )(8)

[0120] Among them, x i(t) is the i-th dung beetle individual in the dung beetle population model; x i (t+1) is the i-th dung beetle individual in the updated dung beetle population model; C1 is the first random vector; C2 is the second random vector; Lb b and Ub b The upper and lower bounds of the preset foraging boundary range.

[0121] In one possible implementation, based on the third random vector, the fourth random vector and the preset egg-laying boundary range, the egg-laying dung beetle behavior is simulated, and the dung beetle population model is updated to obtain an updated dung beetle population model corresponding to the egg-laying dung beetle behavior.

[0122] Specifically, it is shown in the following formula (9):

[0123] B i (t+1)=X * +b1×(B i (t)-Lb * )+b2×(B i (t)-Ub * )(9)

[0124] Among them, X * is the local optimal dung beetle individual of the current iteration number; B i (t) is the position information of the i-th egg ball of the daughter dung beetle model; B i (t+1) is the position information of the i-th egg ball in the updated dung beetle population model; b1 is the third random vector of size 1×d; b2 is the fourth random vector of size 1×d; Lb * and Ub * The upper and lower bounds of the preset spawning boundary range.

[0125] In one possible implementation, a nonlinear boundary selection algorithm is used to determine preset upper and lower boundary ranges, and boundary processing is performed on the dung beetle individuals in the four updated sub-dung beetle population models that exceed the preset upper and lower boundary ranges to obtain four sub-dung beetle population models after boundary processing, and the four sub-dung beetle population models after boundary processing are merged to obtain a first dung beetle population model.

[0126] Specifically, a nonlinear boundary selection algorithm is used to determine the preset upper and lower boundary ranges. Taking the behavior of simulating egg-laying dung beetles as an example, the following formulas (10), (11), and (12) are shown:

[0127] Lb * =max(X * ×(1-R),Lb)(10)

[0128] Ub * =min(X* ×(1+R),Ub)(11)

[0129]

[0130] Among them, Lb * and Ub * The upper and lower bounds of the preset spawning range; X * is the local optimal dung beetle individual for the current number of iterations; Tmax represents the preset number of iterations; and R represents the preset egg-laying boundary range.

[0131] Furthermore, boundary processing is performed on the dung beetle individuals in the four updated sub-dung beetle population models that exceed the preset upper and lower boundary ranges to ensure that all dung beetle individuals are within the preset upper and lower boundary ranges, so as to obtain four sub-dung beetle population models after boundary processing, and the four sub-dung beetle population models after boundary processing are merged to obtain the first dung beetle population model.

[0132] Possibly, the boundary processing is a Gaussian distribution perturbation processing, including: applying Gaussian noise to dung beetle individuals that exceed the preset upper and lower boundary ranges, so that the dung beetle individuals move within the preset upper and lower boundary ranges.

[0133] In a possible implementation, based on the fifth random vector and the first preset value, the behavior of the stealing dung beetle is simulated, and the dung beetle sub-population model is updated to obtain an updated dung beetle sub-population model corresponding to the stealing dung beetle behavior.

[0134] Specifically, it is shown in the following formula (13):

[0135] x i (t+1)=X b +S×g×{|x i (t)-X * |+|x i (t)-X b |}(13)

[0136] Where g is the fifth random vector of size 1×d that obeys the normal distribution; S is a constant; X b is the global optimal dung beetle individual of the current iteration number; x i (t) is the i-th dung beetle individual in the dung beetle population model; x i (t+1) is the i-th dung beetle individual in the updated dung beetle population model; X * is the local optimal dung beetle individual for the current iteration number.

[0137] S3042: Determine a second dung beetle population model based on the target optimization model, the initial dung beetle population model, and the first dung beetle population model, and determine the initial dung beetle population model for the next iterative optimization as the second dung beetle population model.

[0138] Specifically, the fitness values ​​of the dung beetle individuals in the initial dung beetle population model and the fitness values ​​of the dung beetle individuals corresponding to the first dung beetle population model are calculated through the target optimization model, and the dung beetle individuals with lower fitness values ​​are determined as the corresponding dung beetle individuals in the second dung beetle population model, and the dung beetle individual with the lowest fitness among all dung beetle individuals is determined as the global optimal dung beetle individual.

[0139] In one possible implementation, a third dung beetle population model is determined based on the initial dung beetle population model and the first dung beetle population model; the third dung beetle population model is updated using a preset nonlinear Cauchy-Gaussian model to obtain a fourth dung beetle population model; and a preset greedy algorithm is used to screen better dung beetle individuals in the third dung beetle population model and the fourth dung beetle population model to determine a second dung beetle population model.

[0140] Specifically, the third dung beetle population model is updated using a preset nonlinear Cauchy-Gaussian model to obtain a fourth dung beetle population model as shown in formula (14). A preset greedy algorithm is used to screen the better dung beetle individuals in the third and fourth dung beetle population models to determine the second dung beetle population model as shown in formula (15), including:

[0141] X new =X b ×[1+w1×Cauchy+(1-w1)×Gaussian](14)

[0142] w1=1-t / T max

[0143]

[0144] Among them, X new is the updated dung beetle individual in the fourth dung beetle population model; X b is the global optimal dung beetle individual; Cauchy is a random variable that satisfies the Cauchy distribution; Gaussian is a random variable that satisfies the Gaussian distribution; w1 is the inertia weight; T max is the preset number of iterations; f(X new ) is the fitness value corresponding to the updated dung beetle individual in the fourth dung beetle population model; f(X b ) is the fitness value of the global optimal dung beetle individual.

[0145] S3043. Perform the next iterative optimization until the number of iterations is equal to the preset number of iterations, and determine that the final dung beetle population model is the second dung beetle population model corresponding to the last iterative optimization.

[0146] S305: Determine a cloud task scheduling solution based on the final dung beetle population model.

[0147] S306: Allocate multiple cloud tasks to the server cluster based on the cloud task scheduling solution.

[0148] It is possible. Assuming that there are n cloud tasks with different complexities, which need to be assigned to m servers with different hardware resources and processing performance, the mathematical model of cloud task scheduling is shown in the following formula (16):

[0149]

[0150]

[0151] Among them, p ij Represents the allocation relationship between cloud task i and server j.

[0152] Assuming that each cloud task can only be completed by one server, the task scheduling scheme of the above formula n*m can be simplified to a one-dimensional matrix n*1, where the dimension is equal to the number of tasks n, and the value of the element is rounded down to the virtual machine number in the range of (1, m+1). There are m feasible solutions for n tasks and m virtual machines. n kind.

[0153] Taking n = 6, m = 4 as an example, assuming that the individual dung beetles are [4.2, 1.4, 2.5, 3.7, 2.5, 3.6], the mapping relationship between cloud tasks and servers is shown in Table 1:

[0154] Table 1

[0155]

[0156] The cloud task scheduling method provided in the embodiment of the present application is a task scheduling module device that packages the service requests of the network server into cloud tasks, constructs a target optimization model with the minimum completion time as the goal, uses a randomly initialized dung beetle population model and iterative optimization, and finally generates an efficient cloud task scheduling solution, which solves the problem of unbalanced and insufficient server resource utilization in scenarios with different task complexity and server performance, and effectively improves the resource utilization and processing efficiency of the server cluster.

[0157] Furthermore, the initial dung beetle population model was divided into four sub-populations and updated by simulating four behaviors: ball rolling, foraging, egg laying, and stealing. Combined with the target optimization model, a new generation of the second dung beetle population model was determined. These four behaviors simulated different optimization strategies, covering global exploration and local development, preventing a single strategy from falling into a local optimum and improving the comprehensiveness and balance of the solution space search. The sub-population update results were screened based on the target optimization model to ensure that the population evolution direction always aligns with business needs, improving the targeted and effective optimization.

[0158] Through the nonlinear boundary selection algorithm, the preset upper and lower boundary ranges are determined, and the boundary reduction rate is slow in the early stage of iteration, which is convenient for full exploration of the global situation in the early stage. The boundary reduction rate is accelerated in the later stage of iteration, which is convenient for focusing on the vicinity of the current optimal solution in the later stage, performing local opening, and finding a better cloud task scheduling solution. For the dung beetle individuals that exceed the preset upper and lower boundary ranges in the four updated sub-dung beetle population models, boundary processing is performed to ensure that all dung beetle individuals are within the preset upper and lower boundary ranges, avoiding invalid solutions from participating in subsequent optimization, and improving the overall feasibility of the population.

[0159] The third dung beetle population model was updated by pre-setting a nonlinear Cauchy-Gaussian model. The Cauchy distribution, with its heavy-tailed nature, dominates variation in the early stages of the iteration, causing significant perturbations in individual positions, effectively escaping the local optimal solution and expanding the search range. The Gaussian distribution, with its centralization, dominates variation in the later stages of the iteration, fine-tuning the solution near the optimal solution with smaller steps, improving solution accuracy. Combined with a pre-set greedy algorithm, this ensures the dung beetle population model evolves monotonically toward the optimal solution, preventing the loss of high-quality solutions due to random mutations.

[0160] Through the reverse learning algorithm and the preset elite selection model, a better daughter dung beetle population model is determined. The daughter dung beetle population model generated by reverse learning is complementary to the atomic dung beetle population model. Especially in the early stage of iteration, it can significantly increase the solution space coverage density of the population, providing richer candidate solutions for subsequent optimization. The elite selection uses the objective function as the only screening criterion, so that the evolutionary direction of the population always fits the optimization goal, avoiding evolutionary deviation caused by random mutation.

[0161] Based on the preset natural coefficient, preset deflection coefficient and preset light intensity, the behavior of the rolling dung beetle is simulated, and the sub-dung beetle population model is updated to obtain the updated sub-dung beetle population model corresponding to the rolling dung beetle behavior, thereby improving the algorithm robustness and optimization efficiency.

[0162] Based on the first random vector, the second random vector and the preset foraging boundary range, the foraging behavior of dung beetles is simulated, and the sub-dung beetle population model is updated to obtain an updated sub-dung beetle population model corresponding to the foraging behavior of dung beetles, thereby enhancing the local search accuracy and diversity of the population.

[0163] Based on the third random vector, the fourth random vector and the preset egg-laying boundary range, the egg-laying dung beetle behavior is simulated, and the daughter dung beetle population model is updated to obtain the updated daughter dung beetle population model corresponding to the egg-laying dung beetle behavior. This model is used in conjunction with other dung beetle behaviors to improve the global optimization efficiency of the algorithm, and ultimately optimize the load balance and completion time of the task scheduling solution.

[0164] Based on the fifth random vector and the first preset value, the behavior of the stealing dung beetle is simulated, and the sub-dung beetle population model is updated to obtain an updated sub-dung beetle population model corresponding to the stealing dung beetle behavior, thereby improving the algorithm's global optimization ability and solution quality in task scheduling, and ultimately optimizing the load balancing and resource utilization of the service cluster.

[0165] A target optimization model is constructed based on the minimum completion time of the server cluster, its mean square error, the first weight value and the second weight value, which can flexibly balance the cluster efficiency and load balancing through linear weighting.

[0166] Figure 4 A schematic diagram of the structure of the cloud task scheduling device provided in the embodiment of the present application is shown as follows: Figure 4 As shown, the cloud task scheduling device 40 includes: an acquisition module 401, a construction module 402, a processing module 403 and a scheduling module 404.

[0167] The acquisition module 401 is used to acquire multiple service requests sent by the network server and package the multiple service requests into multiple cloud tasks.

[0168] The construction module 402 is used to construct a target optimization model based on the minimum completion time information of the server cluster.

[0169] The construction module 402 is further configured to construct an initial dung beetle population model based on a preset number of dung beetles and a total number of multiple cloud tasks.

[0170] The processing module 403 is used to iteratively optimize the initial dung beetle population model based on the target optimization model until the number of iterations is equal to the preset number of iterations, so as to obtain the final dung beetle population model.

[0171] The processing module 403 is further configured to determine a cloud task scheduling solution based on the final dung beetle population model.

[0172] The scheduling module 404 is used to allocate multiple cloud tasks to the server cluster based on the cloud task scheduling solution.

[0173] In a possible implementation, the processing module 403 is specifically configured to:

[0174] The initial dung beetle population model is divided into four sub-dung beetle population models, and four dung beetle behaviors of ball rolling, foraging, egg laying, and stealing are simulated respectively. The four sub-dung beetle population models are updated to obtain four updated sub-dung beetle population models, and the four updated sub-dung beetle population models are merged to obtain a first dung beetle population model, wherein the dung beetle population model includes a preset number of dung beetle individuals.

[0175] Determining a second dung beetle population model based on the target optimization model, the initial dung beetle population model, and the first dung beetle population model, and determining the initial dung beetle population model for the next iterative optimization as the second dung beetle population model;

[0176] The next iterative optimization is performed until the number of iterations is equal to the preset number of iterations, and the final dung beetle population model is determined to be the second dung beetle population model corresponding to the last iterative optimization.

[0177] In a possible implementation, the processing module 403 is specifically configured to:

[0178] A nonlinear boundary selection algorithm is used to determine the preset upper and lower boundary ranges, and boundary processing is performed on the dung beetle individuals that exceed the preset upper and lower boundary ranges in the four updated sub-dung beetle population models to obtain four sub-dung beetle population models after boundary processing, and the four sub-dung beetle population models after boundary processing are merged to obtain the first dung beetle population model.

[0179] In a possible implementation, the processing module 403 is specifically configured to:

[0180] Based on the initial dung beetle population model and the first dung beetle population model, a third dung beetle population model is determined;

[0181] The third dung beetle population model is updated using a preset nonlinear Cauchy-Gaussian model to obtain the fourth dung beetle population model;

[0182] A preset greedy algorithm was used to screen the better dung beetle individuals in the third and fourth dung beetle population models to determine the second dung beetle population model.

[0183] In a possible implementation, the processing module 403 is specifically configured to: generate a reverse-learned dung beetle population model based on the updated dung beetle population model corresponding to the ball-rolling dung beetle behavior through a reverse learning algorithm;

[0184] Using a preset elite selection model, the updated sub-dung beetle population model corresponding to the rolling ball dung beetle behavior and the reverse learning sub-dung beetle population model were screened for better dung beetle individuals to determine the better sub-dung beetle population model.

[0185] The better sub-dung beetle population model is merged with the remaining three updated sub-dung beetle population models to obtain the first dung beetle population model.

[0186] In a possible implementation, the processing module 403 is specifically configured to:

[0187] Based on the preset natural coefficient, the preset deflection coefficient and the preset light intensity, the behavior of the rolling ball dung beetle is simulated, and the dung beetle sub-population model is updated to obtain an updated dung beetle sub-population model corresponding to the rolling ball dung beetle behavior.

[0188] In a possible implementation, the processing module 403 is specifically configured to:

[0189] Based on the first random vector, the second random vector and the preset foraging boundary range, the foraging behavior of the dung beetle is simulated, and the dung beetle sub-population model is updated to obtain an updated dung beetle sub-population model corresponding to the foraging behavior of the dung beetle.

[0190] In a possible implementation, the processing module 403 is specifically configured to:

[0191] Based on the third random vector, the fourth random vector and the preset egg-laying boundary range, the egg-laying dung beetle behavior is simulated, and the daughter dung beetle population model is updated to obtain an updated daughter dung beetle population model corresponding to the egg-laying dung beetle behavior.

[0192] In a possible implementation, the processing module 403 is specifically configured to:

[0193] Based on the fifth random vector and the first preset value, the behavior of the stealing dung beetle is simulated, and the dung beetle sub-population model is updated to obtain an updated dung beetle sub-population model corresponding to the stealing dung beetle behavior.

[0194] In a possible implementation, the construction module 402 is specifically configured to:

[0195] A target optimization model is constructed based on the minimum completion time information of the server cluster, the mean square error of the minimum completion time information of the server cluster, the first weight value and the second weight value.

[0196] The cloud task scheduling device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effects are similar, and will not be described in detail in this embodiment.

[0197] Figure 5 A structural diagram of a task scheduling device provided in an embodiment of the present application is shown as follows: Figure 5 As shown, the task scheduling device 50 provided in this embodiment includes: at least one processor 501 and a memory 502. Optionally, the device 50 further includes a communication component 503. The processor 501, the memory 502 and the communication component 503 are connected via a bus.

[0198] In a specific implementation process, at least one processor 501 executes the computer-executable instructions stored in the memory 502, so that the at least one processor 501 performs the above method.

[0199] The specific implementation process of the processor 501 can be found in the above method embodiment. Its implementation principle and technical effects are similar and will not be repeated here in this embodiment.

[0200] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASICs), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly executed by a hardware processor or by a combination of hardware and software modules in the processor.

[0201] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (NVM), such as at least one disk memory.

[0202] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be classified into address buses, data buses, and control buses. For ease of illustration, the buses in the drawings of this application are not limited to just one bus or just one type of bus.

[0203] The present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0204] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.

[0205] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile memory device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0206] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0207] It should be noted that for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all optional embodiments, and the actions and modules involved are not necessarily required by this application.

[0208] It should be further noted that, although the various steps in the flowchart are shown in sequence as indicated by the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps may be performed in other orders. Moreover, at least a portion of the steps in the flowchart may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily performed at the same time, but may be performed at different times. The execution order of these sub-steps or stages is not necessarily to be performed in sequence, but may be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0209] It should be understood that the above-described device embodiments are merely illustrative, and the device of the present application may also be implemented in other ways. For example, the division of units / modules in the above-described embodiments is merely a logical functional division, and actual implementations may employ other division methods. For example, multiple units, modules, or components may be combined or integrated into another system, or some features may be omitted or not implemented.

[0210] In addition, unless otherwise specified, the functional units / modules in the various embodiments of the present application may be integrated into a single unit / module, each unit / module may exist physically separately, or two or more units / modules may be integrated together. The aforementioned integrated units / modules may be implemented in the form of hardware or software program modules.

[0211] If the integrated unit / module is implemented in hardware, the hardware may be digital circuits, analog circuits, etc. The physical implementation of the hardware structure includes, but is not limited to, transistors, memristors, etc. Unless otherwise specified, the processor may be any appropriate hardware processor, such as a CPU, GPU, FPGA, DSP, and ASIC. Unless otherwise specified, the storage unit may be any appropriate magnetic storage medium or magneto-optical storage medium, such as resistive random access memory (RRAM), dynamic random access memory (DRAM), static random access memory (SRAM), enhanced dynamic random access memory (EDRAM), high-bandwidth memory (HBM), hybrid memory cube (HMC), etc.

[0212] If the integrated unit / module is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes a number of instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes various media that can store program codes, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0213] In the above embodiments, the description of each embodiment has its own emphasis. For parts not described in detail in a particular embodiment, please refer to the relevant description of other embodiments. The technical features of the above embodiments can be combined in any way. To keep the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0214] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0215] It should be understood that the present application is not limited to the exact structure described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A cloud task scheduling method, characterized in that: Applied to the task scheduling module, including: Acquire multiple service requests sent by the network server, and package the multiple service requests into multiple cloud tasks; Based on the minimum completion time information of the server cluster, a target optimization model is constructed; constructing an initial dung beetle population model based on a preset number of dung beetles and a total number of the plurality of cloud tasks; Based on the target optimization model, the initial dung beetle population model is iteratively optimized until the number of iterations is equal to a preset number of iterations to obtain a final dung beetle population model; Determining a cloud task scheduling solution based on the final dung beetle population model; Based on the cloud task scheduling scheme, the multiple cloud tasks are allocated to the server cluster.

2. The method according to claim 1, characterized in that The iterative optimization of the initial dung beetle population model based on the target optimization model until the number of iterations is equal to a preset number of iterations to obtain a final dung beetle population model includes: The initial dung beetle population model is divided into four sub-dung beetle population models, and four dung beetle behaviors of ball rolling, foraging, egg laying, and stealing are simulated respectively. The four sub-dung beetle population models are updated to obtain four updated sub-dung beetle population models, and the four updated sub-dung beetle population models are merged to obtain a first dung beetle population model, wherein the dung beetle population model includes a preset number of dung beetle individuals. Determining a second dung beetle population model based on the target optimization model, the initial dung beetle population model, and the first dung beetle population model, and determining the initial dung beetle population model for the next iterative optimization to be the second dung beetle population model; The next iterative optimization is performed until the number of iterations is equal to the preset number of iterations, and the final dung beetle population model is determined to be the second dung beetle population model corresponding to the last iterative optimization.

3. The method according to claim 2, characterized in that The initial dung beetle population model is divided into four sub-dung beetle population models, and four dung beetle behaviors of rolling a ball, foraging, laying eggs, and stealing are simulated respectively. The four sub-dung beetle population models are updated to obtain four updated sub-dung beetle population models, and the four updated sub-dung beetle population models are merged to obtain a first dung beetle population model, including: A nonlinear boundary selection algorithm is used to determine the preset upper and lower boundary ranges, and boundary processing is performed on the dung beetle individuals in the four updated dung beetle sub-population models that exceed the preset upper and lower boundary ranges to obtain four dung beetle sub-population models after boundary processing, and the four dung beetle sub-population models after boundary processing are merged to obtain a first dung beetle population model.

4. The method according to claim 2, characterized in that Determining a second dung beetle population model based on the target optimization model, the initial dung beetle population model, and the first dung beetle population model includes: determining a third dung beetle population model based on the initial dung beetle population model and the first dung beetle population model; The third dung beetle population model is updated using a preset nonlinear Cauchy-Gaussian model to obtain the fourth dung beetle population model; A preset greedy algorithm was used to screen the better dung beetle individuals in the third and fourth dung beetle population models to determine the second dung beetle population model.

5. The method according to claim 2, characterized in that The initial dung beetle population model is divided into four sub-dung beetle population models, and four dung beetle behaviors of rolling a ball, foraging, laying eggs, and stealing are simulated respectively. The four sub-dung beetle population models are updated to obtain four updated sub-dung beetle population models, and the four updated sub-dung beetle population models are merged to obtain a first dung beetle population model, including: Based on the updated dung beetle population model corresponding to the rolling ball dung beetle behavior, a reverse learning dung beetle population model is generated through a reverse learning algorithm; Using a preset elite selection model, screening the updated dung beetle population model corresponding to the rolling ball dung beetle behavior and the reversely learned dung beetle population model for better dung beetle individuals to determine a better dung beetle population model; The better sub-dung beetle population model is combined with the remaining three updated sub-dung beetle population models to obtain a first dung beetle population model.

6. The method according to claim 2, characterized in that The initial dung beetle population model is divided into four sub-dung beetle population models, and four dung beetle behaviors of rolling, foraging, laying eggs, and stealing are simulated respectively, and the four sub-dung beetle population models are updated to obtain the updated four sub-dung beetle population models, including: Based on the preset natural coefficient, the preset deflection coefficient and the preset light intensity, the behavior of the rolling ball dung beetle is simulated, and the dung beetle sub-population model is updated to obtain an updated dung beetle sub-population model corresponding to the rolling ball dung beetle behavior.

7. The method according to claim 2, characterized in that The initial dung beetle population model is divided into four sub-dung beetle population models, and four dung beetle behaviors of rolling, foraging, laying eggs, and stealing are simulated respectively, and the four sub-dung beetle population models are updated to obtain the updated four sub-dung beetle population models, including: Based on the first random vector, the second random vector and the preset foraging boundary range, the foraging behavior of the dung beetle is simulated, and the dung beetle sub-population model is updated to obtain an updated dung beetle sub-population model corresponding to the foraging behavior of the dung beetle.

8. The method according to claim 2, characterized in that The initial dung beetle population model is divided into four sub-dung beetle population models, and four dung beetle behaviors of rolling, foraging, laying eggs, and stealing are simulated respectively, and the four sub-dung beetle population models are updated to obtain the updated four sub-dung beetle population models, including: Based on the third random vector, the fourth random vector and the preset egg-laying boundary range, the egg-laying dung beetle behavior is simulated, and the daughter dung beetle population model is updated to obtain an updated daughter dung beetle population model corresponding to the egg-laying dung beetle behavior.

9. The method according to claim 2, characterized in that The initial dung beetle population model is divided into four sub-dung beetle population models, and four dung beetle behaviors of rolling, foraging, laying eggs, and stealing are simulated respectively, and the four sub-dung beetle population models are updated to obtain the updated four sub-dung beetle population models, including: Based on the fifth random vector and the first preset value, the behavior of the stealing dung beetle is simulated, and the dung beetle sub-population model is updated to obtain an updated dung beetle sub-population model corresponding to the stealing dung beetle behavior.

10. The method according to claim 1, characterized in that The target optimization model is constructed based on the minimum completion time information of the server cluster, including: A target optimization model is constructed based on the minimum completion time information of the server cluster, the mean square error of the minimum completion time information of the server cluster, the first weight value and the second weight value.

11. A cloud task scheduling device, characterized in that: include: An acquisition module, configured to acquire multiple service requests sent by a network server and package the multiple service requests into multiple cloud tasks; A construction module is used to build a target optimization model based on the minimum completion time information of the server cluster; The construction module is further configured to construct an initial dung beetle population model based on a preset number of dung beetles and a total number of the plurality of cloud tasks; a processing module, configured to iteratively optimize the initial dung beetle population model based on the target optimization model until the number of iterations is equal to a preset number of iterations, so as to obtain a final dung beetle population model; The processing module is further configured to determine a cloud task scheduling solution based on the final dung beetle population model; A scheduling module is used to allocate the multiple cloud tasks to the server cluster based on the cloud task scheduling scheme.

12. A task scheduling device, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 10 when executed by a processor.

14. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 10 when executed by a processor.

Images