Virtual machine resource scheduling method and device based on improved sunflower algorithm

Abstract

The present invention discloses a virtual machine resource scheduling method based on an improved sunflower algorithm, including: initializing the resource scheduling state of each sunflower, and, based on the total energy obtained by calculation, determining the sunflower where the sun is located in the initial state; determining the energy absorbed by the sunflower; determining the maximum completion time of the total task and the utilization rate of the CPU; with the goal of minimizing the maximum completion time of the total task and maximizing the CPU utilization rate of the total resource consumption, pollinating to generate new sunflowers based on the global state update and local state update of the sunflower, and updating the position of the sunflower; based on the updated sunflowers, updating the sunflower where the sun is located. The method of the present application realizes more global searches at the beginning of the algorithm iteration and more local searches near the end through an improved pollination strategy, which can prevent entering the local optimal solution to a certain extent and also accelerate the convergence speed near the end.

Description

Technical Field

[0001] The present invention relates to the field of cloud computing technology, and in particular to a virtual machine resource scheduling method and device based on an improved sunflower algorithm. Background Art

[0002] As an important branch of distributed computing, cloud computing allows users to access on-demand computing services such as storage, servers, databases, software and other resources through the Internet. The goal of cloud task scheduling is to divide tasks on shared resources to reduce resource consumption and maximize server utilization. Task scheduling is an NP-hard problem. Three types of methods are usually used to solve this type of problem: game theory-based methods, dynamics-based methods, and heuristic algorithm-based methods. Heuristic algorithms usually have better results in solving specific problems. In order to achieve the above requirements and achieve more ideal allocation results, it is necessary to apply heuristic resource scheduling algorithms.

[0003] Heuristic algorithms usually formulate the problem as an optimization problem, solving it by finding the optimal solution. When users apply to use these resources, they need to submit their requests to the task manager module, after which the scheduler obtains the request from the task manager and analyzes all resource requests with the help of the resource information server. However, the current heuristic algorithms are inefficient and their accuracy cannot meet the requirements. Summary of the invention

[0004] The embodiment of the present invention provides a virtual machine resource scheduling method and device based on an improved sunflower algorithm, which can solve the problem of cloud resource scheduling with faster efficiency and higher accuracy.

[0005] The embodiment of the present invention provides a virtual machine resource scheduling method based on an improved sunflower algorithm, which is applied to a resource scheduling environment including M tasks, N virtual machines and Z sunflowers, wherein the sunflower includes M sub-items, and the sub-items are used to define the resource scheduling state of the task executed on the corresponding virtual machine. The virtual machine resource scheduling method includes:

[0006] Initialize the resource scheduling state of each sunflower, and determine the sunflower where the sun is located in the initial state based on the calculated total energy;

[0007] Determining the energy absorbed by the sunflower, wherein the energy absorbed by the sunflower is defined as being determined by minimizing the maximum completion time and maximizing the CPU utilization;

[0008] Determine the maximum total task completion time and CPU utilization;

[0009] With the goal of minimizing the maximum completion time of the total task and maximizing the CPU utilization of the total resource consumption, pollination is performed based on the global state update and local state update of the sunflower to produce new sunflowers, and the position of the sunflower is updated;

[0010] Based on the updated sunflowers, update the sunflowers where the sun is located;

[0011] Determine whether the iteration meets the conditions. If the conditions are met, perform resource scheduling based on the sunflower where the output sun's position is located.

[0012] In some embodiments, the sunflower includes M sub-items, and each sub-item is an integer. Initializing the resource scheduling state of each sunflower includes:

[0013] Within the upper and lower bounds of the numbers that the machine can obtain, the initial state of each sunflower is generated using a formula containing random parameters.

[0014] In some embodiments, the total task maximum completion time is determined as follows:

[0015] Calculate the processing time of the jth task on the i-th virtual machine and multiply it by the Boolean value that determines whether task j can be executed on virtual machine i;

[0016] Sum the processing results of M tasks on a virtual machine;

[0017] The maximum value of the time obtained in N virtual machines is taken as the maximum completion time of the total task.

[0018] In some embodiments, total resource consumption versus CPU utilization is determined as follows:

[0019] Sum the energy consumption of M tasks executed one by one on N virtual machines;

[0020] The average of the summed energy consumption is taken as the total CPU energy consumption.

[0021] In some embodiments, pollinating to generate a new sunflower based on the global state update and the local state update of the sunflower, and updating the position of the sunflower includes:

[0022] Perform pollination operation with probability p and update the position of sunflower;

[0023] Pollination operation meets:

[0024]

[0025] in, For the new sunflower produced after pollination, α tTo control the balance factor between global pollination and local pollination, A is the pollination strategy under local conditions. A is used to define the pollination operation between the current sunflower and the sunflower closest to it to generate new position information to meet B is the pollination strategy under global conditions. B is used to define the pollination operation between the current sunflower and the current sun position, calculate and generate new position information, and satisfy S * Indicates the current position of the sun, and β is the parameter that controls the amplitude of the search direction.

[0026] In some embodiments, pollinating to generate a new sunflower based on the global state update and the local state update of the sunflower, and updating the position of the sunflower further includes:

[0027] The movement operation is performed with a probability of 1-p, and the sunflower moves in the direction of the vector indicated by the line connecting the sun's position with a random step size within a preset range to update the position of the sunflower.

[0028] In some embodiments, based on the updated sunflowers, updating the sunflowers where the sun's position is located includes:

[0029] Calculate the energy absorbed by all sunflowers after the update;

[0030] The sunflower with the best energy absorption is used as the sun position.

[0031] The embodiment of the present invention further proposes a virtual machine resource scheduling device based on an improved sunflower algorithm, which is applied to a resource scheduling environment including M tasks, N virtual machines and Z sunflowers, wherein the sunflower includes M sub-items, and the sub-items are used to define the resource scheduling state of the task executed on the corresponding virtual machine. The virtual machine resource scheduling device includes a processor, which is configured to:

[0032] Initialize the resource scheduling state of each sunflower, and determine the sunflower where the sun is located in the initial state based on the calculated total energy;

[0033] Determining the energy absorbed by the sunflower, wherein the energy absorbed by the sunflower is defined as being determined by minimizing the maximum completion time and maximizing the CPU utilization;

[0034] Determine the maximum total task completion time and CPU utilization;

[0035] With the goal of minimizing the maximum completion time of the total task and maximizing the CPU utilization of the total resource consumption, pollination is performed based on the global state update and local state update of the sunflower to produce new sunflowers, and the position of the sunflower is updated;

[0036] Based on the updated sunflowers, update the sunflowers where the sun is located;

[0037] Determine whether the iteration meets the conditions. If the conditions are met, perform resource scheduling based on the sunflower where the output sun's position is located.

[0038] An embodiment of the present invention further proposes a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the virtual machine resource scheduling method based on the improved Sunflower algorithm described in each embodiment of the present disclosure are implemented.

[0039] The embodiment of the present invention regards various task allocation strategies as a sunflower group through the sunflower algorithm, and regards the current optimal allocation strategy as the sun. Through the improved pollination strategy, more global searches are performed at the beginning of the algorithm iteration, and more local searches are performed near the end, which can prevent entering the local optimal solution to a certain extent and also accelerate the convergence speed near the end.

[0040] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0042] Figure 1 A schematic diagram of the application framework of the virtual machine resource scheduling method of the present application;

[0043] Figure 2 This is a basic flow chart of the virtual machine resource scheduling method of this application;

[0044] Figure 3 This is a general flow chart of the virtual machine resource scheduling method of this application. DETAILED DESCRIPTION

[0045] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0046] The present invention provides a virtual machine resource scheduling method based on an improved sunflower algorithm. Figure 1 As shown, it is applied to a resource scheduling environment including M tasks, N virtual machines and Z sunflowers. The sunflower includes M sub-items, which are used to define the resource scheduling state of the task executed on the corresponding virtual machine. The specific sunflower group P is represented by P = {S1, S 2, S3……S z}, where each S represents a sunflower, initially consisting of Z sunflowers. Sunflower S is represented by S = {s i,1 ,s i,2 ,s i,3 ……s i,M}, where S i,j Represents the jth task in the strategy of this group in s i,j This virtual machine is executed, represented by a positive integer, and each S contains a total of M positive integers. Figure 2 As shown, the virtual machine resource scheduling method includes:

[0047] In step S101, the resource scheduling state of each sunflower is initialized, and the sunflower at the sun position in the initial state is determined based on the calculated total energy. The initial state P is a sunflower group consisting of Z sunflowers, Po p ={S1,S2,……,S z}, for the initial state Po p Determine an optimal solution state S*, which is called the position of the sun in the sunflower algorithm. The two indicators for evaluating the current sunflower state are mainly the minimum maximum completion time and the highest resource consumption to CPU utilization.

[0048] In step S102, the energy absorbed by the sunflower is determined, wherein the energy absorbed by the sunflower is defined as being determined according to the minimum maximum completion time and the maximum CPU utilization.

[0049] In step S103, the maximum total task completion time and CPU utilization are determined.

[0050] In step S104, with the goal of minimizing the maximum total task completion time and maximizing the total resource consumption to CPU utilization, pollination is performed based on the global state update and local state update of the sunflower to generate a new sunflower, and the position of the sunflower is updated.

[0051] In step S105, based on the updated sunflowers, the sunflower where the sun is located is updated.

[0052] In this example, the sunflower algorithm regards various task allocation strategies as a sunflower population, and the current optimal allocation strategy as the position of the sun. The sunflower population will move a certain distance towards the sun in each iteration. At the same time, each sunflower will release pollen, and have a certain probability of producing gametes and generating new individual sunflowers. In this example, it is assumed that each sunflower only produces one new pollen gamete, thereby bringing certain disturbances to the system and preventing it from reaching the local optimal solution.

[0053] In step S106, it is determined whether the iteration satisfies a condition. If the condition is satisfied, resource scheduling is performed based on the sunflower where the outputted sun's position is located.

[0054] The embodiment of the present invention regards various task allocation strategies as a sunflower group through the sunflower algorithm, and regards the current optimal allocation strategy as the sun. Through the improved pollination strategy, more global searches are performed at the beginning of the algorithm iteration, and more local searches are performed near the end, which can prevent entering the local optimal solution to a certain extent and also accelerate the convergence speed near the end.

[0055] In some embodiments, the sunflower includes M sub-items, and each sub-item is an integer. Initializing the resource scheduling state of each sunflower includes:

[0056] Within the upper and lower bounds of the numbers that the virtual machine can obtain, the initial state of each sunflower is generated using a formula containing random parameters. i,1 ,s i,2 ,s i,3 ……s i,M}, the random initialization of S should be the minimum value of s plus a random number between 0 and 1 multiplied by the difference between the maximum value of s and the minimum value of s, ensuring that the obtained s is within the value range and satisfies:

[0057]

[0058] The above min and max The upper and lower bounds of the numbers that the machine can obtain, φ is a random number in the closed interval from 0 to 1.

[0059] In some embodiments, the total task maximum completion time is determined as follows:

[0060] Calculate the processing time of the jth task on the i-th virtual machine and multiply it by the Boolean value that determines whether task j can be executed on virtual machine i; sum the processing results of M tasks on one virtual machine to satisfy:

[0061]

[0062] Where x represents a Boolean value indicating whether task i can be processed on virtual machine j, and P is the processing time of task j on machine i. Since virtual machines can work in parallel, the duration of the task with the longest execution time is considered in this example.

[0063] The maximum value of the time obtained in N virtual machines is taken as the maximum completion time of the total task.

[0064] In some embodiments, total resource consumption versus CPU utilization is determined as follows:

[0065] The energy consumption of M tasks executed one by one on N virtual machines is summed up, and the average value of the summed energy consumption is taken as the total CPU energy consumption, satisfying:

[0066]

[0067] Where E i,j is the resource consumed by task j on virtual machine i. After determining the number of tasks and the status of virtual machines, all E values ​​can be obtained. The average CPU utilization of this batch of tasks can be obtained by calculating and averaging.

[0068] In some embodiments, Figure 3 As shown, pollination is performed based on the global state update and local state update of the sunflower to produce a new sunflower, and the position of the sunflower is updated, including:

[0069] For each sunflower, pollination operation is performed with probability p, and the position of the sunflower is updated;

[0070] Pollination operation meets:

[0071]

[0072] in, is the new sunflower produced after pollination, A is the pollination strategy under local conditions, and A is used to define the pollination operation between the current sunflower and the sunflower closest to it to generate new position information, satisfying B is the pollination strategy under global conditions. B is used to define the pollination operation between the current sunflower and the current sun position, calculate and generate new position information, and satisfy S * Indicates the current position of the sun, β is a parameter that controls the amplitude of the search direction. Specifically, it can be a uniform random number between 3 and 4 to control the situation where the sun position and the sunflower position are too close to each other and cannot converge. t To control the balance factor between global pollination and local pollination, the specific calculation method is:

[0073]

[0074] The above α max With α min Get 0.6 and 0.4, I is the total number of iterations. This formula can be used to control α during the iteration process. t It decreases smoothly with the increase of iteration number. The purpose of shifting the attention mechanism of pollination strategy in the iteration process is achieved. Through such setting, the pollination strategy pays more attention to the operation in the global state at the beginning to prevent premature convergence into the local optimal solution, and pays more attention to the operation in the local state in the later stage, which speeds up the convergence of the final result and improves the efficiency of the scheduling algorithm.

[0075] In some embodiments, pollinating to generate a new sunflower based on the global state update and the local state update of the sunflower, and updating the position of the sunflower further includes:

[0076] For each sunflower, a motion operation is performed with a probability of 1-p, and the sunflower moves in the direction of the vector indicated by the line connecting the sun's position with a preset random step size, and the position of the sunflower is updated. That is, for the remaining sunflowers that do not perform the pollination operation, the motion operation is performed in this example, and the sunflowers that perform the motion operation move directly toward the sun's position for a certain distance. The calculation formula for the movement is as follows:

[0077]

[0078] The above r represents a random number in the open interval from 0 to 1, which is used to add disturbance to the movement strategy of the sunflower to ensure that it can converge in the end while preventing premature convergence. The sum of the two parts represents the movement change of the sunflower towards the sun's position.

[0079] In some embodiments, based on the updated sunflowers, updating the sunflower where the sun's position is located includes: calculating the absorbed energy of all sunflowers after the update, and taking the sunflower with the best absorbed energy as the sun's position. t and F w To evaluate which sunflower should be chosen as the location of the next sun.

[0080] Determine whether the iteration has reached convergence and whether the maximum number of iterations has been reached. If the conditions are not met, proceed to the next round of iteration.

[0081] The specific method of updating the sun's position is to obtain the energy absorption status of all sunflowers and obtain the best one as the new sun's position S*.

[0082] The update iteration round generally sets an upper limit value I to prevent the failure of fast convergence due to incorrect parameter settings. The output result is a set of M numbers in a sunflower S, which means that task j should be run on machine i to achieve the optimal scheduling method.

[0083] Finally, the sunflower information of the sun's position is output. The output information indicates that each task can achieve the highest resource utilization and the shortest running time when running on this virtual machine. For example, the output S*={2,3,2,1} means that task 1 runs on virtual machine 2, task 2 runs on virtual machine 3, task 3 runs on virtual machine 2, and task 4 runs on virtual machine 1 to achieve the best effect.

[0084] The method of the embodiment of the present invention, in the resource scheduling problem under the cloud platform, by simulating the movement of sunflowers absorbing solar energy, divides the entire resource scheduling process into two stages of pollination and movement, iterates and converges step by step, and approaches the optimal solution. In order to obtain the global optimal solution in the correct solution space, the present application introduces an operator that changes the proportion of pollination operations, thereby improving the convergence speed of the algorithm and the efficiency of finding the optimal solution in the scheduling process. The present invention solves the problem of unreasonable resource allocation that is prone to occur in the scheduling process by introducing a heuristic sunflower algorithm into the cloud scheduling task, and increases the resource utilization rate under the cloud platform.

[0085] The embodiment of the present invention further proposes a virtual machine resource scheduling device based on an improved sunflower algorithm, which is applied to a resource scheduling environment including M tasks, N virtual machines and Z sunflowers, wherein the sunflower includes M sub-items, and the sub-items are used to define the resource scheduling state of the task executed on the corresponding virtual machine. The virtual machine resource scheduling device includes a processor, which is configured to:

[0086] Initialize the resource scheduling state of each sunflower, and determine the sunflower where the sun is located in the initial state based on the calculated total energy;

[0087] Determining the energy absorbed by the sunflower, wherein the energy absorbed by the sunflower is defined as being determined by minimizing the maximum completion time and maximizing the CPU utilization;

[0088] Determine the maximum total task completion time and CPU utilization;

[0089] With the goal of minimizing the maximum completion time of the total task and maximizing the CPU utilization of the total resource consumption, pollination is performed based on the global state update and local state update of the sunflower to produce new sunflowers, and the position of the sunflower is updated;

[0090] Based on the updated sunflowers, update the sunflowers where the sun is located;

[0091] Determine whether the iteration meets the conditions. If the conditions are met, perform resource scheduling based on the sunflower where the output sun's position is located.

[0092] An embodiment of the present invention further proposes a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the virtual machine resource scheduling method based on the improved Sunflower algorithm described in each embodiment of the present disclosure are implemented.

[0093] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0094] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0095] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), including a number of instructions for a terminal (which can be a mobile phone, computer, server or network equipment, etc.) to execute the methods described in each embodiment of the present invention.

[0096] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. A virtual machine resource scheduling method based on an improved sunflower algorithm, characterized in that: Applied to a resource scheduling environment including M tasks, N virtual machines and Z sunflowers, the sunflower includes M sub-items, and the sub-items are used to define the resource scheduling state of the task executed on the corresponding virtual machine. The virtual machine resource scheduling method includes: Initialize the resource scheduling state of each sunflower, and determine the sunflower where the sun is located in the initial state based on the calculated total energy; Determining the energy absorbed by the sunflower, wherein the energy absorbed by the sunflower is defined as being determined by minimizing the maximum completion time and maximizing the CPU utilization; Determine the maximum total task completion time and CPU utilization; With the goal of minimizing the maximum completion time of the total task and maximizing the CPU utilization of the total resource consumption, pollination is performed based on the global state update and local state update of the sunflower to produce new sunflowers, and the position of the sunflower is updated; Based on the updated sunflowers, update the sunflowers where the sun is located; Determine whether the iteration meets the conditions. If the conditions are met, perform resource scheduling based on the sunflower where the output sun's position is located; Pollination is performed based on the global state update and local state update of the sunflower to produce a new sunflower, and the position of the sunflower is updated, including: Perform pollination operation with probability p and update the position of sunflower; Pollination operation meets: in, For the new sunflower produced after pollination, α t To control the balance factor between global pollination and local pollination, A is the pollination strategy under local conditions. A is used to define the pollination operation between the current sunflower and the sunflower closest to it to generate new position information to meet B is the pollination strategy under global conditions. B is used to define the pollination operation between the current sunflower and the current sun position, calculate and generate new position information, and satisfy S * represents the current position of the sun, and β is the parameter that controls the amplitude of the search direction; Pollination based on the global state update and local state update of the sunflower produces a new sunflower, and the position update of the sunflower also includes: Perform movement operation with a probability of 1-p, move in the direction of the vector indicated by the line connecting the sun's position with a preset range of random steps, and update the position of the sunflower; Based on the updated sunflowers, the sunflowers where the sun's position is updated include: Calculate the energy absorbed by all sunflowers after the update; The sunflower with the best energy absorption is used as the sun position.

2. The virtual machine resource scheduling method based on the improved sunflower algorithm according to claim 1, characterized in that: The sunflower includes M sub-items, and each sub-item is an integer. Initializing the resource scheduling state of each sunflower includes: Within the upper and lower bounds of the numbers that the machine can obtain, the initial state of each sunflower is generated using a formula containing random parameters.

3. The virtual machine resource scheduling method based on the improved sunflower algorithm according to claim 1, characterized in that: The maximum total task completion time is determined as follows: Calculate the processing time of the jth task on the i-th virtual machine and multiply it by the Boolean value that determines whether task j can be executed on virtual machine i; Sum the processing results of M tasks on a virtual machine; The maximum value of the time obtained in N virtual machines is taken as the maximum completion time of the total task.

4. The virtual machine resource scheduling method based on the improved sunflower algorithm according to claim 1, characterized in that: Total resource consumption versus CPU utilization is determined as follows: Sum the energy consumption of M tasks executed one by one on N virtual machines; The average of the summed energy consumption is taken as the total CPU energy consumption.

5. A virtual machine resource scheduling device based on an improved sunflower algorithm, characterized in that: Applicable to a resource scheduling environment including M tasks, N virtual machines and Z sunflowers, the sunflower including M sub-items, the sub-items are used to define the resource scheduling state of the task executed on the corresponding virtual machine, the virtual machine resource scheduling device includes a processor, and is configured to: Initialize the resource scheduling state of each sunflower, and determine the sunflower where the sun is located in the initial state based on the calculated total energy; Determining the energy absorbed by the sunflower, wherein the energy absorbed by the sunflower is defined as being determined by minimizing the maximum completion time and maximizing the CPU utilization; Determine the maximum total task completion time and CPU utilization; With the goal of minimizing the maximum completion time of the total task and maximizing the CPU utilization of the total resource consumption, pollination is performed based on the global state update and local state update of the sunflower to produce new sunflowers, and the position of the sunflower is updated; Based on the updated sunflowers, update the sunflowers where the sun is located; Determine whether the iteration meets the conditions. If the conditions are met, perform resource scheduling based on the sunflower where the output sun's position is located; Pollination is performed based on the global state update and local state update of the sunflower to produce a new sunflower, and the position of the sunflower is updated, including: Perform pollination operation with probability p and update the position of sunflower; Pollination operation meets: in, For the new sunflower produced after pollination, α t To control the balance factor between global pollination and local pollination, A is the pollination strategy under local conditions. A is used to define the pollination operation between the current sunflower and the sunflower closest to it to generate new position information to meet B is the pollination strategy under global conditions. B is used to define the pollination operation between the current sunflower and the current sun position, calculate and generate new position information, and satisfy S * represents the current position of the sun, and β is the parameter that controls the amplitude of the search direction; Pollination based on the global state update and local state update of the sunflower produces a new sunflower, and the position update of the sunflower also includes: Perform movement operation with a probability of 1-p, move in the direction of the vector indicated by the line connecting the sun's position with a preset range of random steps, and update the position of the sunflower; Based on the updated sunflowers, the sunflowers where the sun's position is updated include: Calculate the energy absorbed by all sunflowers after the update; The sunflower with the best energy absorption is used as the sun position.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the virtual machine resource scheduling method based on the improved Sunflower algorithm as described in any one of claims 1 to 4 are implemented.

Images