A network bandwidth allocation optimization method, system, storage medium and computing device

By screening and grouping network slices in 5G networks, combining the strategies of hard isolation and soft isolation, optimizing the QoS parameters of slices, the contradiction between slice resource isolation and utilization rate is solved, and efficient utilization of network resources is achieved.

CN114423047BActive Publication Date: 2024-11-05STATE GRID ELECTRIC POWER RES INST +4
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210131962.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-02-14
Publication Date
2024-11-05
Estimated Expiration
2042-02-14

AI Technical Summary

Technical Problem

In 5G networks, the resource isolation requirements of network slicing are inconsistent with the network utilization rate, resulting in the hard isolation scheme occupying a large amount of resources and the dynamic slicing service brings resource waste, and soft isolation is difficult to meet the isolation needs.

Method used

The predicted traffic and actual traffic of network slices in historical periods are filtered to predict trusted slices, and the predicted trusted slices are grouped using a hybrid strategy of hard isolation and soft isolation, and optimized according to the QoS parameters of the slices to ensure the isolation of slices. and resource utilization.

Benefits of technology

The isolation of network slices and resource utilization rate is achieved, the contradiction between slice resource isolation and utilization rate is solved, and the utilization efficiency of network resources is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114423047B_ABST
    Figure CN114423047B_ABST
Patent Text Reader

Abstract

The present invention discloses a method, system, storage medium and computing device for optimizing network bandwidth allocation. The present invention divides network slices into predicted credibility, adopts a soft and hard hybrid isolation strategy, groups predicted credible network slices, ensures the isolation of network slices, optimizes the QoS parameters of network slices within the group, ensures the utilization rate of network resources, and thus solves the problem of the contradiction between the isolation requirement of slice network resources and the network utilization requirement.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a network bandwidth allocation optimization method, system, storage medium and computing equipment, belonging to the field of bandwidth allocation. Background Art

[0002] Since the deployment and application of 5G networks, network slicing, as one of the key technologies, has become a carrier bridge for business and transmission resources. Network slicing is an on-demand networking method that allows operators to separate multiple virtual end-to-end networks on a unified infrastructure. Each network slice is logically isolated from the wireless access network, bearer network to the core network to adapt to various types of applications.

[0003] According to the characteristics of 5G networks, a large number of network slice services need to be carried in 5G transport networks, because network slicing technology requires the use of isolation solutions, which are divided into hard isolation and soft isolation. Hard isolation refers to the isolation of physical resources, but for transport networks, hard isolation will occupy a large number of network resources. At the same time, dynamic slice services will cause a large amount of network resource waste in hard pipes. Therefore, there are also a large number of soft isolation solutions in the transport network to realize the carrying of slices. Soft isolation is to add some identifiers in the service flow to distinguish the QoS parameters of the slice service and realize logical QoS isolation.

[0004] Since the frequency of changes in the transport network slice settings is relatively low, and the network slicing services are relatively dynamic, there is a contradiction between the slice network resource isolation requirements and the network utilization requirements. Summary of the invention

[0005] The present invention provides a network bandwidth allocation optimization method, system, storage medium and computing device, which solve the problem of the contradiction between the slice network resource isolation requirement and the network utilization requirement.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0007] A network bandwidth allocation optimization method, comprising:

[0008] Based on the predicted and actual traffic of network slices in historical periods, select network slices with reliable predictions;

[0009] The predicted reliable network slices are grouped; the hard isolation method is used to isolate the groups, and the soft isolation method is used to isolate the groups;

[0010] According to the guaranteed bandwidth rate and maximum bandwidth rate of the network slices in the group, the QoS parameters of the network slices in the group are optimized to obtain the QoS parameters of the network slices and the QoS parameters of the group;

[0011] Network bandwidth is allocated based on the QoS parameters of the network slice and the QoS parameters of the group.

[0012] Based on the network slice predicted traffic and actual traffic in the historical period, filter out the network slices with reliable predictions, including:

[0013] According to the predicted traffic and actual traffic of the network slices in the historical period, the error ratio of the predicted traffic of the network slices in the historical period is calculated;

[0014] Based on the network slice prediction traffic error ratio in the historical period and preset rules, filter out network slices with reliable predictions.

[0015] The default rules are:

[0016] If the network slice prediction traffic error ratios in the most recent n consecutive historical time periods are all less than the threshold, then the network slice is a predicted credible network slice; where n is a preset value.

[0017] Group the predicted trusted network slices, including:

[0018] According to the standard deviation of the network slice prediction traffic error, the network slices with reliable predictions are ranked;

[0019] The sorted predicted trusted network slices are grouped according to the preset maximum fusion degree of network slices.

[0020] According to the guaranteed bandwidth rate and maximum bandwidth rate of the network slices in the group, the QoS parameters of the network slices in the group are optimized to obtain the QoS parameters of the network slices and the QoS parameters of the group, including:

[0021] 1) Calculate the total shared bandwidth value within the group based on the guaranteed bandwidth rate and maximum bandwidth rate of the network slices within the group;

[0022] 2) Calculate the availability of each network slice in the group based on the total shared bandwidth value in the group;

[0023] 3) If the availability of each network slice in the group meets the preset conditions, obtain the QoS parameters of the network slice and the QoS parameters of the group; if the availability of any network slice in the group does not meet the preset conditions, adjust the total shared bandwidth value in the group and go to 2).

[0024] The formula for calculating the availability of each network slice in the group is:

[0025]

[0026] Among them, S is a predicted reliable network slice group, A is a subset of S, and s i is the i-th network slice in the set SA, P ub (s i) is the network slice s i Availability, g i For network slices i The guaranteed rate ratio, For network slices i The guaranteed bandwidth rate, x i For network slices i The business flow, b s is the total shared bandwidth value within S.

[0027] The default conditions are:

[0028] |P ub -min(a i )|≤ε

[0029] Among them, P ub (s i ) is the network slice s i Availability, a i For network slices i availability requirement, and ε is the threshold.

[0030] The formula for adjusting the total shared bandwidth value within a group is:

[0031]

[0032] Among them, b s ′ is the adjusted total shared bandwidth value within the group, b s is the total shared bandwidth value within the group before adjustment, a i For network slices i Availability requirements, g i For network slices i The guaranteed rate ratio, P ub (s i ) is the network slice s i availability.

[0033] A network bandwidth allocation optimization system, comprising:

[0034] Screening module: Screens out reliable network slices based on the predicted and actual traffic of network slices in historical periods;

[0035] Grouping module: groups the predicted reliable network slices; the hard isolation method is used to isolate the groups, and the soft isolation method is used to isolate the groups;

[0036] Optimization module: optimizes the QoS parameters of the network slices in the group according to the guaranteed bandwidth rate and maximum bandwidth rate of the network slices in the group, and obtains the QoS parameters of the network slices and the QoS parameters of the group;

[0037] Allocation module: Allocate network bandwidth according to the QoS parameters of the network slice and the QoS parameters of the group.

[0038] The optimization modules include:

[0039] Total shared bandwidth value calculation module: calculates the total shared bandwidth value within the group according to the guaranteed bandwidth rate and maximum bandwidth rate of the network slices within the group;

[0040] Availability calculation module: calculates the availability of each network slice in the group based on the total shared bandwidth value in the group;

[0041] Judgment module: If the availability of each network slice in the group meets the preset conditions, obtain the QoS parameters of the network slice and the QoS parameters of the group; if the availability of any network slice in the group does not meet the preset conditions, adjust the total shared bandwidth value in the group and transfer to the availability calculation module.

[0042] A computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions, which when executed by a computing device, cause the computing device to perform a network bandwidth allocation optimization method.

[0043] A computing device includes one or more processors, one or more memories, and one or more programs, wherein the one or more programs are stored in the one or more memories and are configured to be executed by the one or more processors, and the one or more programs include instructions for executing a network bandwidth allocation optimization method.

[0044] The beneficial effects achieved by the present invention are as follows: the present invention divides the predicted credibility of network slices, adopts a soft and hard mixed isolation strategy, groups the predicted credible network slices, ensures the isolation of network slices, optimizes the QoS parameters of the network slices within the group, ensures the network resource utilization, and thus solves the problem of the contradiction between the slice network resource isolation requirements and the network utilization requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A flowchart of a method for optimizing network bandwidth allocation;

[0046] Figure 2 Predicting trusted and untrusted transition graphs for network slices;

[0047] Figure 3 Flowchart for QoS parameter optimization. DETAILED DESCRIPTION

[0048] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0049] like Figure 1 As shown, a method for optimizing network bandwidth allocation includes the following steps:

[0050] Step 1: Based on the predicted traffic and actual traffic of network slices in the historical period, select network slices with reliable predictions;

[0051] Step 2: group the predicted reliable network slices; wherein the hard isolation method is used to isolate the groups, and the soft isolation method is used to isolate the groups;

[0052] Step 3: Optimize the QoS parameters of the network slices in the group according to the guaranteed bandwidth rate and the maximum bandwidth rate of the network slices in the group to obtain the QoS parameters of the network slices and the QoS parameters of the group;

[0053] Step 4: Allocate network bandwidth based on the QoS parameters of the network slice and the QoS parameters of the group.

[0054] The above method divides the prediction credibility of network slices, adopts a soft and hard mixed isolation strategy, groups the predicted reliable network slices, ensures the isolation of network slices, optimizes the QoS parameters of network slices within the group, and ensures the utilization of network resources, thereby solving the contradiction between the slice network resource isolation requirements and the network utilization requirements.

[0055] The above method is mainly for 5G transmission network. In 5G transmission network, each network slice can set guaranteed bandwidth rate (GBR) and maximum bandwidth rate (MBR). The network slice is expressed as s i =(a i ,g i ), a i For network slices i Availability requirements, g i ∈[0,1] is the network slice s i The guaranteed rate ratio of network slice s i To have g i Probability operates within the guaranteed bandwidth.

[0056] Traffic prediction has always been one of the key tasks in network optimization. Through the prediction results, network resources and related states can be adjusted dynamically in advance, so as to achieve active matching and real-time linkage between resources and network business needs. In addition to traditional prediction models, with the development of artificial intelligence technology, more and more researchers are applying artificial intelligence technology to the network field. Among them, traffic prediction based on artificial intelligence technology is favored by the majority of relevant personnel, which makes the accuracy and precision of data-based predictions continue to improve.

[0057] Therefore, based on traffic prediction, the predicted traffic of network slices in historical periods can be obtained. Based on the predicted traffic and actual traffic of network slices in historical periods, network slices with reliable predictions and network slices with unreliable predictions can be screened out. The specific process can be as follows:

[0058] 1) Calculate the network slice predicted traffic error ratio for the historical period based on the predicted traffic and actual traffic of the network slice in the historical period;

[0059] The network slice prediction traffic error ratio is the ratio between the network slice prediction traffic error and the actual traffic. Taking the t-1 period as an example, it can be expressed as Among them, t represents the current period, t-1 represents the previous period, t-2 period predicts the traffic of t-1 period, and the network slice predicts the traffic error represents the predicted traffic of the network slice at time t-1, Indicates the actual traffic of the network slice in period t-1.

[0060] 2) Based on the network slice prediction traffic error ratio and preset rules in the historical period, select the network slices that are reliable in prediction;

[0061] like Figure 2 If the network slice predicted traffic error ratio is greater than or equal to the threshold Th Acc , then the prediction result of the network slice is set to be unreliable, that is, the prediction is unreliable network slice; if the network slice prediction traffic error ratio in the most recent n consecutive historical periods is less than the threshold Th Acc , then the prediction result of the network slice is credible, that is, the network slice is a predicted credible network slice; where n is a preset value.

[0062] For network slices that are predicted to be unreliable, their default or preset QoS parameters are returned and the transmission network is isolated and transmitted separately. For network slices that are predicted to be reliable, they are grouped; the grouping is mainly based on the normal distribution of prediction errors and the requirements for network slice isolation. The soft isolation method is used within each group for logical isolation, and the hard isolation method is used for isolation between groups; the specific grouping process can be as follows:

[0063] 21) Sort the network slices with reliable predictions according to the standard deviation of the network slice prediction traffic error;

[0064] Through hypothesis testing, the predicted flow mean μ is calculated i and standard deviation σ i :

[0065]

[0066]

[0067] Among them, R i For network slices i The historical forecast error value, |R i | indicates R i The number of elements in ;

[0068] According to the standard deviation, the predicted reliable network slices are sorted from high to low;

[0069] 22) Grouping the sorted predicted credible network slices according to the preset maximum fusion degree of the network slices;

[0070] Calculate the number of groups based on the preset maximum fusion degree of network slices Among them, S is a predicted reliable network slice group, r is the maximum fusion degree of network slices, which is used to indicate the maximum number of network slices allowed in the group;

[0071] Find the number of groups N that can be fully allocated f = rN G ·r+|S|;

[0072] Take r elements from S in order and put them into a group In total, take N f Then take r-1 network slices and put them into In total, it is divided into N G There are groups, and each group is separated by hard isolation so that they do not affect each other;

[0073] For example: there are 10 network slices in total, and each group has a maximum of 3 network slices, so there are 4 groups in total. First, take 3 network slices in turn to divide into a group, take two times in total, and then take 2 network slices in turn to divide into a group, take two times in total. In this way, there are 3, 3, 2, and 2 network slices in the 4 groups respectively.

[0074] After the grouping is completed, the QoS parameters of each group of network slices are optimized. The principle of parameter optimization is to create a prediction error model based on the prediction results and prediction errors, and to achieve theoretically probability-guaranteed network slice resource allocation based on the cumulative distribution probability analysis of the multivariate normal distribution. The present invention mainly obtains the guaranteed bandwidth, peak bandwidth and peak bandwidth of the network slice. In the process of optimization, the following factors need to be considered:

[0075] 1) Network slice prediction accuracy, that is, the credibility of the prediction result; if the network slice prediction error is continuously less than the set threshold Th, the prediction result is considered to be credible;

[0076] 2) The maximum number of network slices in the group; this part directly affects the grouping situation and the final QoS bandwidth parameter allocation result.

[0077] The specific process of parameter optimization can be as follows Figure 3 As shown:

[0078] 31) Calculate the total shared bandwidth value within the group based on the guaranteed bandwidth rate and maximum bandwidth rate of the network slices within the group;

[0079] According to g i Determine the GBR rate of the network slice According to a i Determine the MBR rate of the network slice Calculate the shareable bandwidth value of network slices Then the total shared bandwidth value within the group can be expressed as

[0080] 32) Calculate the availability of each network slice in the group based on the total shared bandwidth value in the group;

[0081]

[0082] Among them, S is a predicted reliable network slice group, A is a subset of S, and s i is the i-th network slice in the set SA, P ub (s i ) is the network slice s i Availability, g i For network slices i The guaranteed rate ratio, For network slices i The guaranteed bandwidth rate, x i For network slices i The business flow, b s is the total shared bandwidth value within S;

[0083] Indicates that the service flow of each network slice in the set SA is greater than But the total flow is less than The probability of this value obeys the law of multidimensional normal distribution and can be calculated using computing software.

[0084] 33) If the availability of each network slice in the group meets the preset conditions, obtain the QoS parameters of the network slice and the QoS parameters of the group; if the availability of any network slice in the group does not meet the preset conditions, adjust the total shared bandwidth value in the group and go to 2);

[0085] If P ub (s i ) meets the following prerequisites:

[0086] |P ub (s i )-min(a i)|≤ε

[0087] Then the QoS parameters of the network slice and the QoS parameters of the group are directly obtained, that is, the MBR of the group is GBR is The MBR of each network slice is GBR is Among them, ε is the threshold.

[0088] If P ub (s i ) does not meet the above preset conditions, the total shared bandwidth value in the group is adjusted according to the following formula:

[0089]

[0090] Among them, b s ′ is the adjusted total shared bandwidth value within the group, b s is the total shared bandwidth value within the group before adjustment, a i For network slices i Availability requirements, g i For network slices i The guaranteed rate ratio, P ub (s i ) is the network slice s i availability.

[0091] The above method uses the traffic prediction results of network slicing to design a credibility scheme for the traffic prediction results, and gives a scheme for soft isolation and hard isolation grouping of network slices based on the credible prediction results, and then optimizes the QoS parameters for the grouping; among them, by designing the guaranteed rate ratio and maximum fusion degree parameters of the soft isolation slices, it is ensured that the network slices have the exclusivity and isolation of slice resources to a certain extent, ensuring a certain isolation effect; through QoS parameter optimization, on the basis of theoretically achieving the required availability of the slices, the value of the shared bandwidth is greatly reduced, and the utilization efficiency of network resources is effectively improved.

[0092] Based on the same technical solution, the present invention also discloses a software system of the above method, namely a network bandwidth allocation optimization system, comprising:

[0093] Screening module: Screen and predict reliable network slices based on the predicted traffic and actual traffic of network slices in historical periods.

[0094] Grouping module: Group the predicted reliable network slices; the hard isolation method is used to isolate the groups, and the soft isolation method is used to isolate the groups.

[0095] Optimization module: According to the guaranteed bandwidth rate and maximum bandwidth rate of the network slices in the group, the QoS parameters of the network slices in the group are optimized to obtain the QoS parameters of the network slices and the QoS parameters of the group.

[0096] The optimization modules include:

[0097] Total shared bandwidth value calculation module: calculates the total shared bandwidth value within the group according to the guaranteed bandwidth rate and maximum bandwidth rate of the network slices within the group;

[0098] Availability calculation module: calculates the availability of each network slice in the group based on the total shared bandwidth value in the group;

[0099] Judgment module: If the availability of each network slice in the group meets the preset conditions, obtain the QoS parameters of the network slice and the QoS parameters of the group; if the availability of any network slice in the group does not meet the preset conditions, adjust the total shared bandwidth value in the group and transfer to the availability calculation module.

[0100] Allocation module: Allocate network bandwidth according to the QoS parameters of the network slice and the QoS parameters of the group.

[0101] The data processing flow and method of each module in the above system are consistent and will not be repeated here.

[0102] Based on the same technical solution, the present invention also discloses a computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions, and when the instructions are executed by a computing device, the computing device executes a network bandwidth allocation optimization method.

[0103] Based on the same technical solution, the present invention also discloses a computing device, including one or more processors, one or more memories and one or more programs, wherein the one or more programs are stored in the one or more memories and are configured to be executed by the one or more processors, and the one or more programs include instructions for executing a network bandwidth allocation optimization method.

[0104] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0105] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0106] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0107] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0108] The above are merely embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are included in the scope of the claims of the present invention to be approved.

Claims

1. A method for optimizing network bandwidth allocation, characterized in that: include: Based on the predicted and actual traffic of network slices in historical periods, select network slices with reliable predictions; The predicted reliable network slices are grouped; the hard isolation method is used to isolate the groups, and the soft isolation method is used to isolate the groups; According to the guaranteed bandwidth rate and maximum bandwidth rate of the network slices in the group, the QoS parameters of the network slices in the group are optimized to obtain the QoS parameters of the network slices and the QoS parameters of the group; Network bandwidth is allocated based on the QoS parameters of the network slice and the QoS parameters of the group.

2. A method for optimizing network bandwidth allocation according to claim 1, characterized in that: Based on the network slice predicted traffic and actual traffic in the historical period, filter out the network slices with reliable predictions, including: According to the predicted traffic and actual traffic of the network slices in the historical period, the error ratio of the predicted traffic of the network slices in the historical period is calculated; Based on the network slice prediction traffic error ratio in the historical period and preset rules, filter out network slices that are credible in prediction.

3. A method for optimizing network bandwidth allocation according to claim 2, characterized in that: The default rules are: If the network slice prediction traffic error ratios in the most recent n consecutive historical time periods are all less than the threshold, then the network slice is a predicted reliable network slice; where n is a preset value.

4. A method for optimizing network bandwidth allocation according to claim 1, characterized in that: Group the predicted trusted network slices, including: According to the standard deviation of the network slice prediction traffic error, the network slices with reliable predictions are ranked; The sorted predicted trusted network slices are grouped according to the preset maximum fusion degree of the network slices.

5. A method for optimizing network bandwidth allocation according to claim 1, characterized in that: According to the guaranteed bandwidth rate and the maximum bandwidth rate of the network slices in the group, the QoS parameters of the network slices in the group are optimized to obtain the QoS parameters of the network slices and the QoS parameters of the group, including: 1) Calculate the total shared bandwidth value within the group based on the guaranteed bandwidth rate and maximum bandwidth rate of the network slices within the group; 2) Calculate the availability of each network slice in the group based on the total shared bandwidth value in the group; 3) If the availability of each network slice in the group meets the preset conditions, obtain the QoS parameters of the network slice and the QoS parameters of the group; if the availability of any network slice in the group does not meet the preset conditions, adjust the total shared bandwidth value in the group and go to 2).

6. A method for optimizing network bandwidth allocation according to claim 5, characterized in that: The formula for calculating the availability of each network slice in the group is: Where S is a predicted reliable network slice group, A is a subset of S, and s i is the i-th network slice in the set SA, P ub (s i ) is the network slice s i Availability, g i For network slices i The guaranteed rate ratio, B i G For network slices i The guaranteed bandwidth rate, x i For network slices i The business flow, b s is the total shared bandwidth value within S.

7. A method for optimizing network bandwidth allocation according to claim 5, characterized in that: The default conditions are: |P ub (s i )-min(a i )|≤ε Among them, P ub (s i ) is the network slice s i Availability, a i For network slices i availability requirement, and ε is the threshold.

8. A method for optimizing network bandwidth allocation according to claim 5, characterized in that: The formula for adjusting the total shared bandwidth value within a group is: Among them, b s ′ is the adjusted total shared bandwidth value within the group, b s is the total shared bandwidth value within the group before adjustment, a i For network slices i Availability requirements, g i For network slices i The guaranteed rate ratio, P ub (s i ) is the network slice s i availability.

9. A network bandwidth allocation optimization system, characterized in that: include: Screening module: Screens out network slices that are credible based on the predicted traffic and actual traffic of network slices in historical periods; Grouping module: Groups the predicted reliable network slices; hard isolation is used to isolate the groups, and soft isolation is used to isolate the groups; Optimization module: optimizes the QoS parameters of the network slices in the group according to the guaranteed bandwidth rate and maximum bandwidth rate of the network slices in the group, and obtains the QoS parameters of the network slices and the QoS parameters of the group; Allocation module: Allocate network bandwidth according to the QoS parameters of the network slice and the QoS parameters of the group.

10. A network bandwidth allocation optimization system according to claim 9, characterized in that: The optimization modules include: Total shared bandwidth value calculation module: calculates the total shared bandwidth value within the group according to the guaranteed bandwidth rate and maximum bandwidth rate of the network slices within the group; Availability calculation module: calculates the availability of each network slice in the group based on the total shared bandwidth value in the group; Judgment module: If the availability of each network slice in the group meets the preset conditions, obtain the QoS parameters of the network slice and the QoS parameters of the group; if the availability of any network slice in the group does not meet the preset conditions, adjust the total shared bandwidth value in the group and transfer to the availability calculation module.

11. A computer-readable storage medium storing one or more programs, characterized in that: The one or more programs include instructions which, when executed by a computing device, cause the computing device to perform any one of the methods according to claims 1 to 8.

12. A computing device, characterized in that: include: One or more processors, one or more memories, and one or more programs, wherein the one or more programs are stored in the one or more memories and are configured to be executed by the one or more processors, and the one or more programs include instructions for executing any of the methods according to claims 1 to 8.