Network resource pre-allocation methods, terminals, electronic devices and media

Abstract

The application provides a network resource pre-allocation method, a terminal, an electronic device and a medium, wherein the method is applied to the terminal and includes the following steps: S1, determining first parameters, second parameters and third parameters; S2, running a service deployment algorithm based on the second parameters to obtain a target function value; S3, running a derivative-free optimization algorithm based on the target function value, the first parameters and the third parameters to determine a network resource pre-allocation scheme of the next round of iteration; and S4, iteratively executing S2 and S3 until the derivative-free optimization algorithm converges to obtain a target resource pre-allocation scheme, wherein the service deployment algorithm is run based on the network resource pre-allocation scheme of the current round of iteration in each iteration. In the application, the relationship between various heterogeneous network resources can be established, and then the derivative-free optimization is performed through a network resource pre-allocation strategy, so that the pre-allocation of the network resources is realized.

Description

Technical Field

[0001] This application relates to the field of network resource allocation, and provides a method for pre-allocating network resources, a terminal, an electronic device, and a medium. Background Technology

[0002] The service deployment problem in a network can be viewed as the process of allocating various network resources (computing, storage, bandwidth, etc.) to user service requests; that is, the network resource allocation problem. In Software Defined Networking (SDN), Internet Service Providers (ISPs) typically need to apply to operators that own network resources to lease them. Given a network topology, the ISP needs to decide how much network resource to lease on each node and link. This process occurs before the ISP deploys network services and can be considered as the pre-allocation of network resources.

[0003] Currently, network service deployment typically employs an Integer Linear Programming (ILP) model, with total revenue or service quality as the optimization objective. The optimal allocation of network resources is determined either by directly solving the optimization problem or by designing heuristic algorithms. In this process, existing research usually assumes that the total amount of network resources (such as node computing resources and link bandwidth) is known to determine how to allocate network resources for different service requests. However, current technologies do not address the pre-allocation process of network resources by service providers. Summary of the Invention

[0004] This application provides a method for pre-allocating network resources, a terminal, an electronic device and a medium, and an electronic device and a medium to solve the problem in the prior art that network resources can only be allocated to different service requests based on a given total amount of network resources, but cannot enable service providers to pre-allocate network resources.

[0005] To solve the above-mentioned technical problems, this application is implemented as follows:

[0006] In a first aspect, embodiments of this application provide a method for pre-allocating network resources, applied to a terminal. The method includes: S1, determining the following first parameter, second parameter, and third parameter, wherein the first parameter is used to characterize the range of pre-allocated resources on each node and each link in the target network; the second parameter is used to characterize the initial solution of the network resource pre-allocation scheme, the initial solution including the resource rental amount of the service provider on each node and the resource rental amount of the service provider on each link; the third parameter is used to characterize the constraints, the constraints including constraints on the nodes and the links; S2, running a service deployment algorithm based on the second parameter to obtain an objective function value; S3, running a derivative-free optimization algorithm based on the objective function value, the first parameter, and the third parameter to determine the network resource pre-allocation scheme for the next iteration; S4, iteratively executing S2 and S3 until the derivative-free optimization algorithm converges to obtain the target resource pre-allocation scheme, wherein each iteration runs the service deployment algorithm based on the network resource pre-allocation scheme of the current iteration.

[0007] Secondly, embodiments of this application provide a terminal, including: a first determining module, configured to execute step S1, wherein step S1 is to determine the following first parameter, second parameter, and third parameter, wherein the first parameter is used to characterize the range of pre-allocated resources on each node and each link in the target network; the second parameter is used to characterize the initial solution of the network resource pre-allocation scheme, the initial solution including the resource rental amount of the service provider on each node and the resource rental amount of the service provider on each link; the third parameter is used to characterize constraints, the constraints including constraints on the nodes and the links; and a first running... The first module is used to execute step S2, which is to run a service deployment algorithm based on the first parameter and the second parameter to obtain the objective function value; the second running module is used to execute step S3, which is to run a derivativeless optimization algorithm based on the objective function value, the first parameter and the third parameter to determine the network resource pre-allocation scheme for the next iteration; the iterative module is used to execute step S4, which is to iteratively execute S2 and S3 until the derivativeless optimization algorithm converges to obtain the target resource pre-allocation scheme, wherein each iteration runs the service deployment algorithm based on the network resource pre-allocation scheme of the current iteration.

[0008] Thirdly, embodiments of this application provide an electronic device, including: a processor, a memory, and a program stored in the memory and executable on the processor, wherein when the program is executed by the processor, it implements the steps of the network resource pre-allocation method as described in the first aspect above.

[0009] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the network resource pre-allocation method described in the first aspect above.

[0010] In this embodiment, based on the second parameter of the initial solution used to characterize the network resource pre-allocation scheme, a service deployment algorithm is run to obtain the objective function value under the current resource configuration. Then, based on the objective function value, the first parameter used to characterize the range of pre-allocated resources on each node and each link in the target network, and the third parameter used to characterize the constraints, a derivative-free optimization algorithm is run to determine the network resource pre-allocation scheme for the next iteration. The above process is repeated iteratively until a set of optimal resource pre-allocation schemes, i.e., the target resource pre-allocation scheme, can be converged, thereby realizing the pre-allocation of network resources. This solves the problem in the prior art that network resources can only be allocated to different service requests based on the total amount of given network resources (such as node computing resources and link bandwidth), but cannot realize the pre-allocation of network resources by service providers. Attached Figure Description

[0011] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a flowchart illustrating a method for pre-allocating network resources provided in an embodiment of this application;

[0013] Figure 2 This is a flowchart of the network resource pre-allocation algorithm based on derivative-free optimization provided in the embodiments of this application;

[0014] Figure 3 This is a schematic diagram of a network topology provided in an embodiment of this application;

[0015] Figure 4 This is a schematic diagram of the derivative-free optimization iteration results based on method 1 provided in the embodiments of this application;

[0016] Figure 5 This is a schematic diagram of the derivative-free optimization iteration results based on method 2 provided in the embodiments of this application;

[0017] Figure 6 This is a schematic diagram of the terminal structure provided in the embodiments of this application. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] Because most existing service deployment algorithms primarily focus on matching network resources with service requests—that is, the allocation process of network resources—the total amount of network resources (such as node computing resources and link bandwidth) is known. Moreover, due to the complexity of problem modeling and solving, existing network resource allocation algorithms are usually simplified during modeling or solving, focusing only on the allocation of a single type of network resource, making it difficult to uniformly optimize the configuration of various heterogeneous network resources such as computing, storage, and bandwidth.

[0020] In this embodiment, the service provider's optimization variables are various heterogeneous network resource rental amounts, such as the computing resource rental amounts of each node and the bandwidth rental amounts of each link. The optimization objectives can be set as performance parameters such as user request acceptance rate, total revenue, average latency, and service quality, according to the actual needs of the service provider. This embodiment also provides a resource pre-allocation strategy based on derivative-free optimization (also known as black-box optimization), treating the service deployment process as a black box. This facilitates the unified establishment of the complex relationships implicit between various heterogeneous network resources, and the problem description is concise and convenient, enabling unified optimization of various heterogeneous network resources such as computing, storage, and bandwidth.

[0021] See Figure 1 , Figure 1 This is a flowchart illustrating a method for pre-allocating network resources according to an embodiment of this application, used by a terminal, such as... Figure 1 As shown, the method includes the following steps:

[0022] S1, determine the following first parameter, second parameter and third parameter, wherein the first parameter is used to characterize the range of pre-allocated resources on each node and each link in the target network, the second parameter is used to characterize the initial solution of the network resource pre-allocation scheme, the initial solution includes the resource rental amount of the service provider on each node and the resource rental amount of the service provider on each link; the third parameter is used to characterize the constraints, including constraints on nodes and links.

[0023] It should be noted that the target network in the embodiments of this application can be a heterogeneous network.

[0024] S2, run the service deployment algorithm based on the second parameter to obtain the objective function value;

[0025] It should be noted that the service deployment algorithm in this application embodiment refers to an algorithm used to calculate the deployment strategy of network services, that is, an algorithm used to calculate network resources, node allocation, etc. Furthermore, the objective function value in this application embodiment is a performance metric of concern to the service provider (ISP), and is determined according to the specific needs of the ISP. For example, if the ISP focuses on availability, it can use the service acceptance rate as the objective function value; if the ISP focuses on total revenue, it can use the total revenue value as the objective function value. In other words, the objective function value is used to evaluate the performance of the resource allocation scheme in each iteration.

[0026] S3, based on the objective function value, the first parameter and the third parameter, runs a derivative-free optimization algorithm to determine the network resource pre-allocation scheme for the next iteration;

[0027] It should be noted that the derivative-free optimization algorithm in this application embodiment can be implemented using black-box optimization toolkits such as OpenBox (an open-source system designed for black-box optimization, based on Bayesian optimization, to efficiently solve black-box optimization problems), NOMAD (a flexible service orchestration tool that uses a simple and unified workflow to easily deploy and manage any containerized application or traditional application), and Hyperopt (hyperparameter optimization).

[0028] S4 iterates through S2 and S3 until the derivative-free optimization algorithm converges to obtain the target resource pre-allocation scheme. In each iteration, the service deployment algorithm is run based on the network resource pre-allocation scheme of the current iteration.

[0029] Through steps S1 to S4 above, based on the second parameter used to characterize the initial solution of the network resource pre-allocation scheme, the service deployment algorithm is run to obtain the objective function value under the current resource configuration. Then, based on the objective function value, the first parameter used to characterize the range of pre-allocated resources on each node and each link in the target network, and the third parameter used to characterize the constraints, a derivative-free optimization algorithm is run to determine the network resource pre-allocation scheme for the next iteration. The above process is repeated iteratively until a set of optimal resource pre-allocation schemes, i.e. the target resource pre-allocation scheme, can be converged, thereby realizing the pre-allocation of network resources. This solves the problem in the prior art that network resources can only be allocated to different service requests based on the total amount of given network resources (such as node computing resources and link bandwidth), but cannot realize the pre-allocation of network resources by service providers.

[0030] In optional embodiments of this application, the method for determining the first parameter, second parameter, and third parameter involved in step S1 above may further include:

[0031] Step 11: Obtain the total number of nodes and links in the target network;

[0032] In this specific example, if the distribution of user requests is known, for a given target network G(V, E), the total number of nodes is V and the total number of links is E.

[0033] Step 12: Based on the total number of nodes and the total number of links, predefine the range of pre-allocated resources on each node in the target network and the range of pre-allocated resources on each link in the target network to obtain the first parameter;

[0034] Based on this, we first define the search space (corresponding to the first parameter) S∈R. E+V This refers to the upper and lower bounds of the pre-allocated resources on each node and each link in the network, i.e., the range of pre-allocated resources on each node and each link.

[0035] Step 13: Based on the first resource lease amount for each node in the target network and the second resource lease amount for each link provided by the service provider, set the initial solution of the network resource pre-allocation scheme to obtain the second parameter;

[0036] In a specific example, let c∈R V ,g∈R E This refers to the resource rental amount for the service provider on each node, i.e., each link, and a set of network resource pre-allocation schemes are set [c] (0) ,g (0) [This is] used as the initial solution.

[0037] Step 14: Set constraints based on the service provider's demand for network resources to obtain the third parameter.

[0038] In a specific example, constraints `cons` are defined based on the service provider's actual needs. In this case, where the service provider's demand for network resources is price, the method of setting constraints based on the service provider's demand for network resources in step 14 can further include:

[0039] Step 21: Obtain the first price of the computing source for renting a unit of resource on each node in the target network, and the second price of renting a unit of bandwidth resource on each link in the target network.

[0040] Step 22: Determine the constraints based on the first price and the first resource rental quantity, and the second price and the second resource rental quantity.

[0041] Furthermore, the method for determining the constraint conditions involved in step 22 can be to first obtain the first product result of the first price and the first resource rental quantity, and obtain the second product result of the second price and the second resource rental quantity, and then set the sum of the first product result and the second product result to be less than and equal to the constraint value, thereby determining the constraint condition, wherein the constraint value is the upper limit of the total expenditure of the service provider for renting network resources.

[0042] For steps 21 to 22 above, in a specific example, it can be assumed that the price of renting a unit of computing source on each node is α∈R. V The price for leasing a unit of bandwidth on each link is β∈R. E If the total overhead of a service provider renting network resources is C, then the constraint const can be expressed as:

[0043] α T c+β T g≤C

[0044] Based on this, the iterative process in step S4 above can, in a specific example, be based on the network resource pre-allocation scheme in each iteration [c]. (n) ,g (n) Run the service deployment algorithm once to obtain the objective function value under the current resource configuration, i.e., obj. (n) =PRB(c (n) ,g (n) Then, based on the selected derivative-free optimization algorithm and constraints, the resource pre-allocation scheme for the next iteration is determined. As the number of iterations increases, the black-box optimization algorithm can converge to obtain a set of optimal resource pre-allocation schemes. Therefore, the entire network resource pre-allocation algorithm process based on derivative-free optimization in this embodiment is illustrated in the specific example as follows: Figure 2 As shown.

[0045] As can be seen, in this embodiment of the application, the entire service deployment process can be regarded as a black box. The input is the service provider's resource pre-allocation scheme, that is, the amount of resources leased on each node and link in the target network. Based on the converged optimal resource pre-allocation scheme, the output can be the performance indicators that the service provider cares about, such as user request acceptance rate, total revenue, service quality and other parameters, so that the allocation of network resources is more in line with actual needs.

[0046] Furthermore, the derivative-free optimization iterative process in this embodiment may suffer from slow convergence speed because each iteration requires running the service deployment algorithm once. When the number of service requests to be deployed is large or the service deployment algorithm itself is complex, the time required for each iteration is long. To further improve the iterative convergence speed of the derivative-free optimization process in this embodiment, more prior information can be provided, such as adding constraints (cons) or reducing the search space S.

[0047] In the embodiments of this application, the first and second parameters can be optimized in the following ways to improve the iteration convergence speed:

[0048] Step 31: Based on historical user requests and service deployment results, determine the first correlation between user requests, service deployments and resource consumption on each node and link in the target network. Here, user requests are used to characterize the resource consumption on each node and link where user requests are distributed, and service deployment results are used to characterize the resource consumption on each node and link requested by the service provider.

[0049] Step 32: Determine the first parameter and the second parameter based on the first association relationship.

[0050] Steps 31 and 32 above are obtained by analyzing historical user requests and service deployment results to obtain the relationship between the distribution of user service requests and the resource occupancy of each node and link in the network. This guides and adjusts the search space (first parameter) and initial solution (second parameter) of derivative-free optimization, thereby obtaining more optimized first and second parameters, which can improve the iterative convergence speed in the embodiments of this application.

[0051] In this embodiment, the third parameter can also be optimized in the following way to improve the iteration convergence speed:

[0052] Step 41: Based on the load of the target network, determine the second correlation between the resource usage of each node and each link in the target network;

[0053] Step 42: Determine the third parameter based on the second association relationship.

[0054] As can be seen, in steps 41 and 42 of the above embodiments of this application, the correlation between resource usage of each node and link in the network is obtained by analyzing the network topology and load conditions, such as the positive / negative correlation between the bandwidth usage of certain links. Furthermore, the second correlation in the embodiments of this application is determined by the geometric structure of the network topology and can be transformed into a constraint condition (third parameter) for derivative-free optimization.

[0055] Based on this, in a specific example of the embodiments of this application, machine learning methods can be used to learn the prior information including the optimized search space, initial solution, and constraints. In a specific example, this prior information may be:

[0056] 1) Dynamic prior information: By analyzing historical user requests and service deployment results, the relationship between the distribution of user service requests and the resource consumption of each node and link in the network is obtained. This prior information is used to guide and adjust the search space and initial solution of derivative-free optimization.

[0057] 2) Static prior information: By analyzing the network topology and load conditions, the correlation between resource usage on each node and link in the network is obtained. For example, the bandwidth usage of certain links may show a positive / negative correlation. This prior information is determined by the geometric structure of the network topology and can be transformed into constraints for derivative-free optimization.

[0058] In this embodiment, machine learning methods can be used to guide the derivativeless optimization process, transforming the learned prior information into optimized constraints in the derivativeless optimization algorithm, thus reducing the number of iterations required for convergence and improving the overall running speed of the resource pre-allocation algorithm in this embodiment.

[0059] The present application will now be explained in detail with reference to specific embodiments of the present application. These specific embodiments provide a method for pre-allocating network resources, such as... Figure 3 The network topology shown has 15 nodes and 30 links. Figure 3 The computing nodes in the network have computing capabilities and can deploy services; forwarding nodes have no computing capabilities and can only forward service data. Nodes with solid boxes represent possible origin points of requests, and nodes with dashed boxes represent possible end points of requests. The service provider plans to deploy 200 service requests in this network, distributed as follows: 100 requests have their origin and end points randomly selected from the aforementioned origin and end points, and the other 100 requests have their origin and end points randomly selected throughout the network. For example, if the network has sufficient resources at each node but limited network bandwidth, and the price for renting a unit of network resource on each node and link is 1, then the specific settings for no-derivative optimization are as follows:

[0060] 1) Optimization variable: The amount of bandwidth leased by the service provider on each link.

[0061] 2) Optimization goal: Minimize the service request rejection rate, i.e., the number of failed deployment requests / total number of requests.

[0062] 3) Constraints: The total cost of the service provider renting network resources is ≤80,000.

[0063] For the network topology and user service request distribution described above, two sets of simulation experiments were conducted in this specific implementation using different search spaces (upper and lower limits of leased bandwidth for each link) and initial solutions to verify the effect of using prior information to guide the derivativeless optimization process. The search spaces (upper and lower bandwidth limits) and initial solutions corresponding to the two sets of experiments are shown in Table 1.

[0064]

[0065]

[0066] Table 1

[0067] In this study, the sampling interval of the search space was set to 50. Method 1 used a search space without prior information, with the same upper and lower limits and initial values ​​for the leased bandwidth on each link. Method 2 differentiated the search space for different links in the network by learning the relationship between the distribution of user service requests and the resource occupancy of each node and link in the network. Both methods iterated for 150 rounds, recording the number of service request rejections in each iteration. The derivative-free optimization iteration results of Method 1 and Method 2 are shown below. Figure 4 and Figure 5 The curve represents how the minimum number of request rejections changes with the number of iterations.

[0068] based on Figure 4 and Figure 5 It can be seen that as the number of iterations increases, the number of service request rejections generally decreases, and at the end of the iteration, the minimum number of request rejections for both Method 1 and Method 2 is the same, thus verifying the convergence of the derivativeless optimization algorithm. Furthermore, during the iteration process until convergence, with a fixed number of iterations, the minimum request rejection rate for Method 2 is lower than that for Method 1. And with a fixed minimum number of request rejections, Method 2 requires fewer iterations than Method 1. Therefore, in this embodiment, using machine learning methods to guide the derivativeless optimization process, transforming learned prior information into optimized constraints and a reduced search space, can reduce the number of iterations required to reach convergence, thereby improving the overall running speed of the resource pre-allocation algorithm.

[0069] See Figure 6 , Figure 6 This is a schematic diagram of the structure of a terminal provided in an embodiment of this application, such as... Figure 6 As shown, the terminal includes:

[0070] The first determining module 602 is used to execute step S1, which determines the following first parameter, second parameter and third parameter. The first parameter is used to characterize the range of pre-allocated resources on each node and each link in the target network. The second parameter is used to characterize the initial solution of the network resource pre-allocation scheme, which includes the resource rental amount of the service provider on each node and the resource rental amount of the service provider on each link. The third parameter is used to characterize the constraints, which include constraints on nodes and links.

[0071] The first running module 604 is used to execute step S2, which is to run the service deployment algorithm based on the second parameter to obtain the objective function value;

[0072] The second running module 606 is used to execute step S3, which is to run a derivative-free optimization algorithm based on the objective function value, the first parameter and the third parameter to determine the network resource pre-allocation scheme for the next iteration.

[0073] The iteration module 608 is used to execute step S4, which is to iteratively execute S2 and S3 until the derivative-free optimization algorithm converges to obtain the target resource pre-allocation scheme. In each iteration, the service deployment algorithm is run based on the network resource pre-allocation scheme of the current iteration.

[0074] Through the terminal of this application embodiment, based on the second parameter used to characterize the initial solution of the network resource pre-allocation scheme, a service deployment algorithm is run to obtain the objective function value under the current resource configuration. Then, based on the objective function value, the first parameter used to characterize the range of pre-allocated resources on each node and each link in the target network, and the third parameter used to characterize the constraints, a derivative-free optimization algorithm is run to determine the network resource pre-allocation scheme for the next iteration. The above process is repeated iteratively until a set of optimal resource pre-allocation schemes, i.e., the target resource pre-allocation scheme, can be converged, thereby realizing the pre-allocation of network resources. This solves the problem in the prior art that network resources can only be allocated to different service requests based on the total amount of given network resources (such as node computing resources and link bandwidth), but cannot realize the pre-allocation of network resources by service providers.

[0075] In an optional embodiment of this application, the device may further include: a second determining module, configured to determine a first correlation between user requests, service deployments, and resource occupancy on each node and each link in the target network based on historical user requests and service deployment results, wherein user requests are used to characterize the resource occupancy on each node and each link where user requests are distributed, and service deployment results are used to characterize the resource occupancy on each node and each link requested by the service provider; and a third determining module, configured to determine a first parameter and a second parameter based on the first correlation.

[0076] In an optional embodiment of this application, the device may further include: a fourth determining module, configured to determine a second correlation between the resource usage of each node and each link in the target network based on the load of the target network; and a fifth determining module, configured to determine a third parameter based on the second correlation.

[0077] In an optional embodiment of this application, the first determining module may further include: an acquisition unit, configured to acquire the total number of nodes and the total number of links in the target network; a first predefinition unit, configured to predefine the range of pre-allocated resources on each node in the target network and the range of pre-allocated resources on each link in the target network based on the total number of nodes and the total number of links, to obtain a first parameter; a setting unit, configured to set an initial solution of the network resource pre-allocation scheme based on the first resource rental amount on each node in the target network and the second resource rental amount on each link provided by the service provider, to obtain a second parameter; and a second setting unit, configured to set constraints based on the service provider's demand for network resources, to obtain a third parameter.

[0078] In an optional embodiment of this application, when the service provider's demand for network resources is based on price, the second setting unit in this application embodiment may include: an acquisition subunit, used to acquire a first price for a unit of computing source rented on each node in the target network, and a second price for a unit of bandwidth resource rented on each link in the target network; and a determination subunit, used to determine constraints based on the first price and the first resource rental amount, and the second price and the second resource rental amount.

[0079] In an optional embodiment of this application, the determining subunit in this application embodiment is further configured to obtain a first product result of a first price and a first resource rental amount, and to obtain a second product result of a second price and a second resource rental amount; and to set the sum of the first product result and the second product result to be less than and equal to a constraint value, thereby determining a constraint condition, wherein the constraint value is the upper limit of the total cost of the service provider renting network resources.

[0080] It should be noted that the terminal provided in this application embodiment is capable of performing the above-described... Figure 1 If the apparatus of the network resource pre-allocation method in the embodiments is applicable to the terminal, then all implementations of the network resource pre-allocation method in the above embodiments are applicable to the terminal and can achieve the same or similar beneficial effects.

[0081] This application also provides an electronic device, including: a processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the above-described functionality. Figure 1The various processes of the network resource pre-allocation method embodiment shown can achieve the same technical effect, and will not be described again here to avoid repetition.

[0082] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, this computer program implements the various processes of the network resource pre-allocation method embodiment shown in example 1 above, and achieves the same technical effect. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0083] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0084] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0085] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for pre-allocation of network resources, applied to a terminal, characterized in that, The method comprises: S1, determining the following first parameter, second parameter and third parameter, wherein the first parameter is used to represent the range of pre-allocated resources on each node and each link in the target network, the second parameter is used to represent the initial solution of the network resource pre-allocation scheme, and the initial solution includes the resource rental amount of the service provider on each node and the resource rental amount of the service provider on each link; and the third parameter is used to represent the constraint condition, which includes the constraint condition of the node and the link; S2, running a service deployment algorithm based on the second parameter to obtain a target function value; S3, running a derivative-free optimization algorithm based on the target function value, the first parameter and the third parameter to determine the network resource pre-allocation scheme of the next round of iteration; S4, iteratively performing the S2 and the S3 until the derivative-free optimization algorithm converges to obtain a target resource pre-allocation scheme, wherein the service deployment algorithm is run based on the network resource pre-allocation scheme of the current round of iteration in each iteration.

2. The method of claim 1, wherein, The method further comprises: determining a first correlation between the user request, the service deployment and the resource occupation amount on each node and each link in the target network based on historical user requests and service deployment results, wherein the resource occupation amount on each node and each link represents the user request distribution, and the resource occupation amount on each node and each link represents the service provider requested deployment; determining the first parameter and the second parameter based on the first correlation.

3. The method of claim 1, wherein, The method further comprises: determining a second correlation between the resource usage amount on each node and each link in the target network based on the load condition of the target network; determining the third parameter based on the second correlation.

4. The method of claim 1, wherein, The determination of the following first parameter, second parameter and third parameter comprises: obtaining the total number of nodes and the total number of links in the target network; predefining the range of pre-allocated resources on each node in the target network and the range of pre-allocated resources on each link in the target network based on the total number of nodes and the total number of links to obtain the first parameter; setting the initial solution of the network resource pre-allocation scheme based on the first resource rental amount of the service provider on each node in the target network and the second resource rental amount on each link to obtain the second parameter; setting the constraint condition based on the demand of the service provider for the network resources to obtain the third parameter.

5. The method of claim 4, wherein, In the case that the demand of the service provider for the network resources is the price, setting the constraint condition based on the demand of the service provider for the network resources comprises: obtaining the first price of a computing source renting a unit resource on each node in the target network and the second price of renting a unit bandwidth resource on each link in the target network; determining the constraint condition based on the first price and the first resource rental amount, and the second price and the second resource rental amount.

6. The method of claim 5, wherein, The constraint condition is determined based on the first price and the first resource lease amount, and the second price and the second resource lease amount, comprising: a first product result of the first price and the first resource lease amount is obtained, and a second product result of the second price and the second resource lease amount is obtained; a sum of the first product result and the second product result is set to be less than and equal to a constraint value, and the constraint condition is determined, wherein the constraint value is an upper limit of a total cost of the service provider leasing network resources.

7. A terminal, characterized by comprising: comprising: a first determining module configured to perform step S1, wherein the step S1 is to determine a first parameter, a second parameter and a third parameter, wherein the first parameter is used to represent a range of pre-allocated resources on each node and each link in a target network, the second parameter is used to represent an initial solution of a network resource pre-allocation scheme, and the initial solution includes resource lease amounts of a service provider on each node and resource lease amounts of the service provider on each link; and the third parameter is used to represent a constraint condition, and the constraint condition includes constraints on the nodes and the links; a first running module configured to perform step S2, wherein the step S2 is to run a service deployment algorithm based on the second parameter to obtain a target function value; a second running module configured to perform step S3, wherein the step S3 is to run a derivative-free optimization algorithm based on the target function value, the first parameter and the third parameter to determine a network resource pre-allocation scheme of a next round of iteration; an iteration module configured to perform step S4, wherein the step S4 is to iteratively perform the steps S2 and S3 until the derivative-free optimization algorithm converges to obtain a target resource pre-allocation scheme, and each iteration is based on a network resource pre-allocation scheme of a current round of iteration to run the service deployment algorithm.

8. The terminal according to claim 7, characterized by Further comprising: a second determining module configured to determine a first correlation between a user request, a service deployment and resource occupation amounts on nodes and links in a target network based on historical user requests and service deployment results, wherein the user request is used to represent resource occupation amounts on the nodes and the links in a user request distribution, and the service deployment result is used to represent resource occupation amounts on the nodes and the links requested to be deployed by a service provider; a third determining module configured to determine the first parameter and the second parameter based on the first correlation.

9. An electronic device, comprising: comprising: a processor, a memory, and a program stored on the memory and executable on the processor, and the program, when executed by the processor, implements steps of the network resource pre-allocation method according to any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the computer readable storage medium, and the computer program, when executed by a processor, implements steps of the network resource pre-allocation method according to any one of claims 1 to 6.

Images