Method, apparatus and medium for deploying service function chain of joint user access

Abstract

The application provides a joint user access service function chain deployment method, device, equipment and medium, comprising: determining a target deployment network based on a pre-constructed space-air-ground integrated network model of joint user access and service function chain deployment; determining a target deployment algorithm based on the importance of each virtualized network function called by a service function chain in a service function chain set; deploying the virtualized network function according to the target deployment algorithm to determine the target network node of the virtualized network function in the target deployment network, with the goal of minimizing end-to-end delay; and determining the resource amount allocated to the virtualized network function based on the task amount of different virtualized network functions carrying service function chains, to complete the deployment of the service function chain set. The application can further shorten the end-to-end delay in SDN-SAGIN.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of service function chain deployment, in particular to a service function chain deployment method and device for joint user access, equipment and medium. BACKGROUND

[0002] With the rapid development of mobile Internet, people's demand for network coverage is increasing, and the traditional ground network is difficult to meet the growing user demand due to its limited coverage. In order to cope with this challenge, Space-Air-Ground Integrated Network (SAGIN) has emerged. SAGIN combines unmanned aerial vehicles, satellite networks and ground networks to provide more extensive coverage of communication services. However, SAGIN faces a variety of protocols coexist, multi-dimensional resources and multiple network interfaces, how to carry out high dynamic heterogeneous network integration is a hot research issue in SAGIN.

[0003] Software Defined Network (SDN) combined with Virtual Network Functions (VNF) is considered to be the preferred technology for heterogeneous network integration. This perfect combination not only realizes the separation of control signaling and data, but also realizes the separation of software and hardware facilities. SDN controller can deploy VNF at a reasonable location in the network according to different user needs. These ordered VNFs form a service function chain (SFC) that can complete the user's specific needs. Therefore, SDN-SAGIN not only solves the problem of heterogeneous network integration, but also improves the flexibility and efficiency of the network. However, due to the long distance between space-based networks, air-based networks and ground users, shortening the end-to-end delay in SDN-SAGIN becomes an important research target, especially for delay-sensitive services.

[0004] The main method for shortening the end-to-end delay in existing SDN-SAGIN is through SFC deployment. Because SFC is composed of some ordered VNF chain, the deployment location of different VNFs will directly affect the propagation delay. The existing SFC deployment methods mainly include the following: the first method is to extend the SFC deployment from the ground network to the air-based network and space-based network; the second method is based on the theory of Timeline Expansion Network (TEN); the third method is based on heuristic or meta-heuristic algorithm; the fourth method is based on mathematical model; the fifth method is based on machine learning and reinforcement learning. However, the above methods for shortening the end-to-end delay still have optimization space. SUMMARY

[0005] Therefore, the present application aims to provide a service function chain deployment method, device, equipment and medium combined with user access, which can further shorten the end-to-end delay in SDN-SAGIN.

[0006] In a first aspect, the embodiments of the present application provide a service function chain deployment method combined with user access, comprising:

[0007] Based on the pre-constructed space-air-ground integrated network model of joint user access and service function chain deployment, a target deployment network is determined; wherein the space-air-ground integrated network model comprises a space-based network, a space-based network and a ground network, and the space-based network, the space-based network and the ground network are all configured with network nodes, the associated terminal of the user is accessed from the ground network to the target deployment network based on the service function chain set, and the target deployment network is the space-based network or the space-based network;

[0008] Based on the importance of the virtualized network function called by each service function chain in the service function chain set, a target deployment algorithm is determined; wherein the target deployment algorithm is a front-end deployment algorithm or a back-end deployment algorithm;

[0009] With the goal of minimizing end-to-end delay, the virtualized network function is deployed according to the target deployment algorithm to determine the target network node of the virtualized network function deployed in the target deployment network;

[0010] Based on the task amount of different virtualized network functions carrying service function chains, the target network node of the virtualized network function is determined to allocate the resource amount of the virtualized network function to complete the deployment of the service function chain set.

[0011] In an embodiment, based on the pre-constructed space-air-ground integrated network model of joint user access and service function chain deployment, the target deployment network is determined, comprising:

[0012] Based on the pre-constructed space-air-ground integrated network model of joint user access and service function chain deployment, it is judged whether the associated terminal of the user is covered by the space-based network, and whether the source network node and the destination network node corresponding to the service function chain in the service function chain set are covered by the space-based network;

[0013] If yes, the space-based network is determined as the target deployment network;

[0014] If not, the space-based network is determined as the target deployment network.

[0015] In an embodiment, based on the importance of the virtualized network function called by each service function chain in the service function chain set, the target deployment algorithm is determined, comprising:

[0016] According to the virtualized network function called by each service function chain in the service function chain set, the service function chain combination is divided to obtain a service function chain sub-set corresponding to each virtualized network function;

[0017] Each service function chain sub-set is counted to determine the number of calls of each virtualized network function, and the number of calls is used to describe the importance of each virtualized network function;

[0018] Based on the position of each virtualized network function in the pre-constructed virtualized network function set, the front-end virtualized network function and the back-end virtualized network function are determined.

[0019] Based on the importance of the front-end virtualized network function and the importance of the back-end virtualized network function, a target deployment algorithm is determined.

[0020] In an embodiment, the virtualized network function is deployed according to the target deployment algorithm to determine the target network node of the virtualized network function in the target deployment network, with the goal of minimizing end-to-end delay, including:

[0021] Based on the target deployment algorithm, the deployment order of the virtualized network function is determined to deploy the virtualized network function according to the deployment order method;

[0022] In the process of deploying the current virtualized network function, the current virtualized network function is pre-deployed in the network node of the target deployment network, and based on the service function chain sub-set corresponding to the current virtualized network function, the candidate network node corresponding to the current virtualized network function is determined from the pre-deployed network node, with the goal of minimizing end-to-end delay.

[0023] Determine whether the candidate network node meets the preset node and link constraint condition;

[0024] If not, the candidate network node is removed from the target deployment network, and the candidate network node corresponding to the current virtualized network function is re-determined until the new candidate network node meets the node and link constraint condition, the candidate network node is used as the target network node corresponding to the current virtualized network function, and the current virtualized network function is deployed to the target network node.

[0025] In an embodiment, based on the service function chain sub-set corresponding to the current virtualized network function, the candidate network node corresponding to the current virtualized network function is determined from the pre-deployed network node, with the goal of minimizing end-to-end delay, including:

[0026] If the current virtualized network function is the first deployed virtualized network function, link delays between the source network node or the destination network node corresponding to each service function chain in the service function chain sub-set corresponding to the current virtualized network function and the network node pre-deployed for the current virtualized network function are determined, and the network node corresponding to the shortest link delay is taken as the candidate network node corresponding to the current virtualized network function.

[0027] If the current virtualized network function is not the first deployed virtualized network function, link delays between the target network node corresponding to the previous virtualized network function or the target network node corresponding to the next virtualized network function and the network node pre-deployed for the current virtualized network function are determined, and the network node corresponding to the shortest link delay is taken as the candidate network node corresponding to the current virtualized network function.

[0028] In an implementation, judging whether the candidate network node satisfies the preset node and link constraint condition comprises:

[0029] judging whether total occupied resources of all virtualized network functions deployed on the candidate network node are less than CPU resources corresponding to the candidate network node, and whether total traffic of service function chains flowing through the candidate network node is less than link bandwidth;

[0030] If yes, it is determined that the preset node and link constraint condition is satisfied.

[0031] In an implementation, based on the task amount of service function chains borne by different virtualized network functions, determining the resource amount allocated to the virtualized network function by the target network node on which the virtualized network function is deployed comprises:

[0032] taking a sum value of data transmission amounts corresponding to each service function chain calling the virtualized network function as the task amount of service function chains borne by the virtualized network function;

[0033] If there are multiple virtualized network functions deployed on the target network node on which the virtualized network function is deployed, taking a sum value of task amounts of service function chains borne by each virtualized network function deployed on the target network node as total task amount corresponding to the target network node;

[0034] based on a ratio between the task amount and the total task amount, determining the resource amount allocated to the virtualized network function by the target network node on which the virtualized network function is deployed.

[0035] In a second aspect, an embodiment of the present application further provides a service function chain deployment device combined with user access, comprising:

[0036] The network determining module is configured to determine a target deployment network based on a pre-constructed integrated space-air-ground network model of joint user access and service function chain deployment, wherein the integrated space-air-ground network model comprises a space-based network, a space-based network and a ground network, and each of the space-based network, the space-based network and the ground network is configured with a network node, an associated terminal of a user accesses to the target deployment network from the ground network based on a service function chain set, and the target deployment network is the space-based network or the space-based network.

[0037] The algorithm determining module is configured to determine a target deployment algorithm based on an importance degree of a virtualized network function called by each service function chain in the service function chain set, wherein the target deployment algorithm is a front-end deployment algorithm or a back-end deployment algorithm.

[0038] The deployment module is configured to deploy the virtualized network function according to the target deployment algorithm with the minimum end-to-end delay as a target, to determine a target network node of the virtualized network function deployed in the target deployment network.

[0039] The resource allocation module is configured to determine a resource amount allocated to the virtualized network function by the target network node of the virtualized network function deployed based on a task amount of the service function chain borne by different virtualized network functions, to complete deployment of the service function chain set.

[0040] In a third aspect, an electronic device is provided, including a processor and a memory, the memory storing computer executable instructions capable of being executed by the processor, and the processor executes the computer executable instructions to implement the method of any one of the first aspect.

[0041] In a fourth aspect, a computer readable storage medium is provided, the computer readable storage medium storing computer executable instructions, and when the computer executable instructions are called and executed by a processor, the computer executable instructions cause the processor to implement the method of any one of the first aspect.

[0042] The embodiment of the application provides a service function chain deployment method, device and equipment combined with user access and a medium. First, a space-air-ground integrated network model of combined user access and service function chain deployment is constructed in advance to determine a target deployment network. The space-air-ground integrated network model comprises a space-based network, a space-based network and a ground network. Network nodes are configured in the space-based network, the space-based network and the ground network. An associated terminal of a user is accessed to the target deployment network from the ground network based on a service function chain set. The target deployment network is the space-based network or the space-based network. Then, the importance of a virtualized network function called by each service function chain in the service function chain set is determined to determine a target deployment algorithm. The target deployment algorithm is a front-end deployment algorithm or a back-end deployment algorithm. The virtualized network function is deployed according to the target deployment algorithm to determine a target network node of the virtualized network function in the target deployment network, with the minimum end-to-end delay as the target. Finally, the target network node of the virtualized network function is determined based on the task amount of the virtualized network function carrying the service function chain to allocate the resource amount of the virtualized network function to the target network node to complete the deployment of the service function chain set. The above method introduces the user access condition, dynamically adjusts the network to be accessed and the algorithm to be used, deploys the virtualized network function based on the above, allocates the resource amount of the virtualized network function, optimizes the use of network resources, and minimizes the end-to-end delay.

[0043] Other features and advantages of the present application will be set forth in the descriptions below, and in part will become apparent to those skilled in the art upon examination of the following or can be learned by practice of the present application. The objects and other advantages of the present application can be realized and attained by the structure particularly pointed out in the description, claims and drawings.

[0044] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the following preferred embodiments are specifically described below, and the accompanying drawings are described in detail as follows. BRIEF DESCRIPTION OF DRAWINGS

[0045] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings needed to be used in the specific embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0046] Figure 1 A flowchart of a service function chain deployment method combined with user access provided by the embodiment of the present application is shown in the figure.

[0047] Figure 2 An example diagram of combined user access and VNF deployment provided by the embodiment of the present application is shown in the figure.

[0048] Figure 3 A space-air-ground integrated network model diagram provided for an embodiment of the present application;

[0049] Figure 4 A joint user access SFC two-end deployment algorithm flowchart provided for an embodiment of the present application;

[0050] Figure 5 A resource allocation flowchart provided for an embodiment of the present application;

[0051] Figure 6 A structure schematic diagram of a joint user access service function chain deployment device provided for an embodiment of the present application;

[0052] Figure 7 A structure schematic diagram of an electronic device provided for an embodiment of the present application. DETAILED DESCRIPTION

[0053] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions of the present application will be described below in connection with the embodiments. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application.

[0054] The existing SFC deployment method mainly includes the following:

[0055] The first SFC deployment method: the SFC deployment algorithm commonly used in ground networks is extended to space-based networks and space-based networks. Although this method can realize the SFC deployment problem in SAGIN network, the effect is not very good, because this method ignores the characteristics of heterogeneous networks, ignores the user access network, and does not consider the end-to-end delay difference caused by user access to different satellites, and all of them consider the SFC deployment situation in the quasi-static topology.

[0056] The second SFC deployment method: the SFC deployment problem is realized by time-expanded network theory, which can effectively analyze and represent dynamic network structure. By expanding the nodes and connection relationships of the network on the time axis, the state and change at different time points are displayed. However, the problem of splitting time into multiple granularity has always been unsolvable, and SFC needs to be re-modeled and deployed for each time node. Not only a large amount of system resources is occupied, but also the computational complexity is high.

[0057] The third SFC deployment method is to deploy SFC by using a heuristic or meta-heuristic algorithm, which can find a high-quality approximate solution and has strong flexibility, but cannot guarantee to find a global optimal solution, and is highly dependent on the problem, complex in parameter tuning, and easy to fall into a local optimal solution search space.

[0058] The fourth SFC deployment method is to deploy SFC by using a mathematical model, which usually uses linear programming, integer linear programming, mixed integer linear programming, nonlinear programming, and dynamic programming to solve. This method can usually provide high precision and even find an optimal solution, but has high computational complexity, is difficult to model, and is sensitive to parameters.

[0059] The fifth SFC deployment method is to deploy SFC by using deep learning and reinforcement learning algorithms, which can usually get a good deployment result, but usually needs a large number of experiments to train to get an optimal solution. Especially when the network size is large, the network needs to be monitored in time to prune the overfitting phenomenon.

[0060] Conclusion: The existing SAGIN service chain deployment method is mostly based on ground network, and expands space-based and air-based network nodes and links. The research focuses on SFC deployment and resource allocation. The existing SFC deployment algorithm assumes that the user selects the nearest satellite access network, and shortens the inter-satellite link delay and VNF processing delay by deploying SFC, but ignores the user access situation. Different access network nodes will affect the end-to-end delay of SFC. Therefore, the embodiment of the present application proposes a joint user access service function chain deployment method, device, equipment and medium, which not only considers the user access network problem, but also focuses on the problem of affecting the end-to-end delay caused by accessing different network nodes. The proposed method not only reduces the algorithm complexity, but also shortens the end-to-end delay.

[0061] Embodiment one:

[0062] In order to facilitate the understanding of the present embodiment, first, a joint user access service function chain deployment method disclosed by the embodiment of the present application is introduced in detail, referring to the flowchart of a joint user access service function chain deployment method shown in Figure 1 The method mainly includes the following steps S102 to S108:

[0063] Step S102, based on the pre-constructed integrated space-air-ground network model of joint user access and service function chain deployment, determine the target deployment network.

[0064] The SAGIN model, an integrated air-space-ground network model, comprises air-based, space-based, and terrestrial networks. Each network contains network nodes. A user's associated terminal accesses the target deployment network from the terrestrial network based on a Service Function Chain (SFC) set, which can be either an air-based or space-based network. In one example, if the user's associated terminal is covered by airspace, and both the source and destination network nodes of the corresponding service function chain are covered by airspace, then the air-based network is identified as the target deployment network; otherwise, the space-based network is identified as the target deployment network.

[0065] Step S104: Determine the target deployment algorithm based on the importance of the virtualized network functions called by each service function chain in the service function chain set.

[0066] The target deployment algorithm can be either a front-end deployment algorithm or a back-end deployment algorithm. In one example, this embodiment proposes a two-end deployment algorithm for the joint user access service function chain deployment, and performs service function chain deployment based on this algorithm to minimize end-to-end latency. Specifically, a virtualized network function set (VNF set) is pre-constructed. Based on the position of the virtualized network functions in this set, the front-end and back-end virtualized network functions are determined, and their importance is compared to determine the target deployment algorithm from the front-end and back-end deployment algorithms.

[0067] Step S106: With the goal of minimizing end-to-end latency, deploy the virtualized network function according to the target deployment algorithm to determine the target network node in which the virtualized network function is deployed in the target deployment network.

[0068] In one example, the service function chain deployment problem for joint user access is described as a mixed-integer programming problem. An end-to-end latency expression is defined, and the virtualized network function is deployed according to a target deployment algorithm with the goal of minimizing end-to-end latency. Specifically, during the deployment of the current virtualized network function, it is pre-deployed in the network nodes of the target deployment network, and candidate network nodes corresponding to the current virtualized network function are determined from the pre-deployed network nodes. If a candidate network node does not meet the preset node and link constraints, a new candidate network node is determined until the new candidate network node meets the node and link constraints. This new candidate network node is the target network node to which the current virtualized network function will be deployed.

[0069] Step S108: Based on the workload of the service function chain carried by different virtualized network functions, determine the resource amount allocated to the target network node deployed by the virtualized network function, and complete the deployment of the service function chain set.

[0070] In one example, the workload of the service function chain carried by the virtualized network function is first determined, as well as the total workload corresponding to the target network node deployed by the virtualized network function. Finally, based on the ratio between the workload and the total workload, the amount of resources allocated to the target network node deployed by the virtualized network function for the virtualized network function can be determined.

[0071] The service function chain deployment method for joint user access provided in this embodiment of the invention dynamically adjusts the network to be accessed and the algorithm used for deployment by introducing user access information, and deploys virtualized network functions and allocates resources to virtualized network functions on this basis, thereby achieving optimal utilization of network resources and minimization of end-to-end latency.

[0072] Example 2:

[0073] With the development of mobile communication technology and Network Function Virtualization (VNF) technology, the Space-Air-Ground Integrated Network (SAGIN) is gradually becoming an important architecture for future communication networks. In this architecture, the efficient deployment of Service Function Chains (SFCs) is crucial for improving network performance and service quality. However, existing SFC deployment research mainly focuses on resource allocation and service chain optimization, often neglecting user access scenarios. This can lead to uneven resource utilization and degraded service quality in practical applications. Therefore, this invention proposes a joint user access SFC deployment scheme. By incorporating user access scenarios and analyzing user location characteristics, it dynamically adjusts user access policies and SFC deployment schemes to achieve optimal utilization of network resources and minimize end-to-end latency.

[0074] To more clearly illustrate the importance of the issues addressed in the embodiments of the present invention, see [link to relevant documentation]. Figure 2The diagram illustrates a combined user access and VNF deployment. User UE1 is simultaneously covered by satellites S4 and S5, and can choose either S4 or S5 to access the satellite network. Assume user UE1 has two services transmitted to users UE2 and UE3 respectively. If user UE1 sends data to user UE2, it should choose satellite S5 for access, with an end-to-end latency of 27ms, while access via S4 would require 40ms. If user UE1 sends data to user UE3, it should choose satellite S4 for access, with an end-to-end latency of 30ms, while access via S5 would require 39ms. Therefore, different access schemes should be selected based on the receiving end of different services; simply choosing the shortest distance or maximum received power access is not the optimal choice. Let's assume only user UE1 needs to transmit service data to user UE2 via the virtualized network function VNF1, and VNF1 can be deployed on any of the satellite nodes S1-S6. Assuming user UE1 accesses the satellite network via S4, the end-to-end latencies caused by VNF1s deployed at S1-S6 are 64ms, 64ms, 64ms, 40ms, 40ms, and 40ms, respectively. Assuming user UE1 accesses the satellite network via S5, the end-to-end latencies caused by VNF1s deployed at S1-S6 are 73ms, 51ms, 54ms, 49ms, 27ms, and 27ms, respectively. Similarly, if user UE1 needs to transmit service data to user UE3 via virtualized network function VNF1, the optimal access scheme and optimal deployment location are: access node S4, VNF1 deployed at network node S1 or S4, with a minimum end-to-end latency of 30ms. Therefore, the SFC deployment scheme for joint user access is a noteworthy issue.

[0075] Existing SAGIN service chain deployment methods are mostly based on terrestrial networks, expanding them with space-based and airborne network nodes and links. Research focuses on SFC deployment and resource allocation, neglecting user access and failing to consider the impact of network node access on end-to-end latency. Therefore, this invention addresses these issues by constructing a joint optimization mathematical model for user access, SFC deployment, and network node resource allocation. Furthermore, it proposes user access methods, VNF two-end deployment methods, and CPU allocation methods to shorten end-to-end latency.

[0076] Specifically, the core points of the service function chain deployment method for joint user access provided in this embodiment of the invention are as follows: 1) In the SAGIN user access introduction SFC deployment problem, this embodiment of the invention proposes an optimization problem for joint user access, SFC deployment, and resource allocation with the goal of minimizing end-to-end latency. 2) The joint optimization problem is described as a mixed integer programming problem, and constraints are given. 3) Based on all SFC request sets, and given the constraints and the solution objective, a two-end deployment method for SFC is proposed, which reduces the end-to-end latency of SFC. 4) Based on the data volume of SFC, a CPU resource allocation strategy is proposed, which reduces the end-to-end latency.

[0077] Specifically, the main idea of ​​the service function chain deployment method for joint user access provided in this embodiment of the invention is as follows: 1) Introducing user access into the SFC deployment problem of SAGIN, and through the analysis of different access points, obtaining the SFC end-to-end latency expression, and proposing optimization problems for joint user access, SFC deployment, and resource allocation. 2) Describing the above problem as a mixed integer programming problem and giving constraints. 3) Based on all SFC request sets, and given the constraints and solution objectives, proposing a two-end deployment method for SFC, reducing the SFC end-to-end latency. 4) Based on the data volume of SFC, proposing a CPU resource allocation strategy, reducing the end-to-end latency. 5) Obtaining user access selection, VNF deployment method, and CPU allocation scheme.

[0078] Based on this, this embodiment of the invention provides a specific implementation method for deploying a service function chain for joint user access.

[0079] Before performing the aforementioned step S102, it is necessary to pre-build an integrated air-space-ground network model for joint user access and SFC deployment, as well as define the expression of end-to-end latency, node and link constraints.

[0080] (a) Establish the SAGIN model, define the VNF set, construct the SFC set, obtain the CPU resource capabilities of different network nodes, and obtain the latency and bandwidth resources between different links. Construct an integrated air-space-ground network model that combines user access and SFC deployment. See steps A1 to A3 below for details:

[0081] Step A1, see Figure 3 The diagram shown is an integrated air-space-ground network model, which can be represented as follows: , , ,in This represents the set of underlying physical nodes (i.e., network nodes). Represents a set of links between network nodes, for example include and , include and , include and SAG stands for space-based network, air-based network, and terrestrial network, respectively. , , These represent the number of nodes in each network. , , Let represent the sets of links between air-space, ground-to-ground, and air-to-ground nodes, respectively. The CPU resources of each network node are represented as follows: , The bandwidth resources and link latency between network nodes are respectively expressed as... , , , That is, network nodes and Inter-bandwidth resources, That is, network nodes and Link latency between nodes. Assume the set of all VNF instances in the system can be represented as... ,in Defined as the first Each VNF consumes a certain amount of CPU resources during operation. Here, VNFs represent network functions required by the network, such as firewalls, encryption, decryption, and deep packet compression. The set of service function chains in a network can be represented as... , coexist A service function chain, in which Indicates the first A service function chain. Each service function chain can be represented as... ,in , Represents the service function chain The source network node and the destination network node, Represents the service function chain The VNF set that needs to be traversed. , Represents the service function chain The first in A VNF, Represents the service function chain The total number of VNFs that need to be traversed. Represents the service function chain The amount of data transmitted. To reduce network deployment costs, it is assumed that each VNF can only be deployed on one network node in each network (space-based network, air-based network).

[0082] Step A2, Deployment model between VNF and network nodes: Define coupled variables , ,in express Deployed on network nodes Otherwise, there is no deployment. , Indicating in space-based and space-based networks It can only be deployed on one network node. Define a coupled variable. Represents the service function chain Occupied Define variables Represents the service function chain The first in Is the VNF the nth VNF in the VNF set? A VNF, Represents the service function chain The first in The VNF is the th VNF in the VNF set. VNF.

[0083] Step A3, User Access Model: Define Coupled Variables , , This indicates that the user is covered by airspace, and it applies to the service function chain. In other words, , Both are covered by airspace; users access the air-based network if they are not, and otherwise access the space-based network.

[0084] (ii) The SFC deployment problem for joint user access is described as a mixed-integer programming problem, defining the expression for end-to-end latency with the goal of minimizing end-to-end latency. See steps B1 to B2 below for details:

[0085] Step B1, End-to-end latency model: The latency of each service function chain includes propagation latency, transmission latency, and processing latency.

[0086] Define propagation delay ,in express Link length between , This indicates the speed at which light travels.

[0087] Define transmission delay ,in For data transmission volume, For link Transmission rate.

[0088] Define processing delay ,in Represented as nodes for Allocated computing resources. This represents the number of CPU cycles required to process each bit.

[0089] Therefore, the service function chain The total link latency is .

[0090] Step B2, Node and Link Constraints:

[0091] For network nodes in space-based and air-based networks, the resources occupied by deployed VNFs cannot exceed the total capacity (i.e., CPU resources) of that network node. The expression for this constraint is as follows:

[0092] ;

[0093] Define link bandwidth constraint couple variables When the service function chain Link occupied , Otherwise, it is 0. The expression for this constraint is as follows:

[0094] ;

[0095] Indicates all nodes that flow through The traffic between service function chains must not exceed the link bandwidth.

[0096] (III) A two-end deployment algorithm for SFC deployment with joint user access is proposed. Based on this algorithm, SFC deployment is performed to minimize end-to-end latency. The basic idea of ​​this part is to first sort the importance of VNFs, and then deploy VNFs sequentially according to the two-end deployment method with minimum latency, specifically involving the aforementioned steps S102 to S106.

[0097] Regarding the aforementioned step S102, this embodiment of the invention provides a specific implementation method for determining the target deployment network based on a pre-constructed integrated air-space-ground network model deployed with a joint user access and service function chain: Based on the pre-constructed integrated air-space-ground network model deployed with a joint user access and service function chain, it is determined whether the user's associated terminal is covered by the air-based network, and whether the source network node and destination network node corresponding to the service function chain in the service function chain set are covered by the air-based network; if so, the air-based network is determined as the target deployment network; if not, the space-based network is determined as the target deployment network.

[0098] In the specific implementation, in the collection In the middle, first determine whether it meets the requirements. ,Right now , All are covered by airspace. Users access the air-based network or the space-based network. Let the set of service function chains accessing the air-based network be denoted as . The service function chain set for accessing the space-based network is as follows: ,in Because the deployment scheme of VNF is in and The VNF deployment scheme is consistent across different regions, therefore a unified overview will be provided. The following adopts... Indicates, that is or .

[0099] Regarding the aforementioned step S104, this embodiment of the invention provides an implementation method for determining the target deployment algorithm based on the importance of virtualized network functions called by each service function chain in the service function chain set, specifically including the following steps C1 to C4:

[0100] Step C1: Based on the VNFs called by each service function chain in the service function chain set, divide the service function chains into subsets corresponding to each VNF.

[0101] In specific implementation, according to The service function chain set in the middle VNF combines and breaks down all possible service chains deployed in the network into A collection of service function chains Each service function chain subset represents a collection of all service function chains that call this VNF.

[0102] Step C2 involves statistically analyzing each service function chain subset to determine the number of calls to each VNF, using the call count to describe the importance of each VNF.

[0103] In the specific implementation, they are respectively in the collection The number of calls to each VNF is calculated. Definition Indicates the first The number of times a VNF is called. The higher the value, the more important the VNF is.

[0104] Step C3: Based on the position of each VNF in the pre-built VNF ​​set, determine the front-end VNF and the back-end VNF.

[0105] Step C4: Determine the target deployment algorithm based on the importance of the front-end VNF and the back-end VNF.

[0106] In one example, if the importance of the front-end VNF is greater than that of the back-end VNF, then the front-end deployment algorithm is determined as the target deployment algorithm; otherwise, the back-end deployment algorithm is determined as the target deployment algorithm.

[0107] In practical implementation, comparison and Size, prioritize deployment of important components. , Define for each service function chain. , .

[0108] In one example, the target deployment algorithm is as follows , It is dynamically adjusted according to changes. For example, suppose... The importance is higher than The importance of this factor dictates that a front-end deployment algorithm should be used for deployment at this stage; when... After deployment, the new front-end VNF (i.e., ...) will be compared. )and The importance of each, if Its importance is still higher than Depending on the importance, the front-end deployment algorithm will continue to be used. Its importance is still lower than The importance of each VNF will be determined by the backend deployment algorithm, and so on, until all VNFs are deployed.

[0109] Regarding the aforementioned step S106, this embodiment of the invention provides an implementation method that aims to minimize end-to-end latency by deploying virtualized network functions according to a target deployment algorithm to determine the target network nodes in the target deployment network where the virtualized network functions are deployed. Specifically, see steps D1 to D4 below:

[0110] Step D1: Determine the deployment order of VNFs based on the target deployment algorithm, so as to deploy VNFs according to the deployment order method.

[0111] In one example, if the front-end deployment algorithm is used, the deployment starts from the front-end VNF and continues until the back-end VNF is deployed; if the back-end deployment algorithm is used, the deployment starts from the back-end VNF and continues until the front-end VNF is deployed.

[0112] Step D2: During the deployment of the current VNF, the current VNF ​​is pre-deployed in the network nodes of the target deployment network. With the goal of minimizing end-to-end latency, candidate network nodes corresponding to the current VNF ​​are determined from the pre-deployed network nodes based on the service function chain subset corresponding to the current VNF. For specific implementation details, please refer to Case 1 and Case 2 below:

[0113] Scenario 1: If the current VNF ​​is the first deployed VNF, then determine the link latency between the source network node or destination network node of each service function chain in the service function chain subset corresponding to the current VNF ​​and the network node pre-deployed in the current VNF, and take the network node corresponding to the shortest link latency as the candidate network node corresponding to the current VNF.

[0114] Specifically, if a front-end deployment algorithm is used, it is necessary to determine the link latency between the source network node of each service function chain in the service function chain subset corresponding to the current VNF ​​and the pre-deployed network node of the current VNF; if a back-end deployment algorithm is used, it is necessary to determine the link latency between the destination network node of each service function chain in the service function chain subset corresponding to the current VNF ​​and the pre-deployed network node of the current VNF.

[0115] The embodiments of the present invention will explain and illustrate Case 1 using front-end deployment algorithm and back-end deployment algorithm as examples respectively:

[0116] Front-end deployment algorithm: If Then deploy It is necessary to Calculate all source network nodes to The link latency of pre-deployed network nodes is considered, and among all pre-deployed network nodes, the network node with the shortest latency is selected. The deployment. Among them... From all source network nodes to any network node in the network The time delay can be expressed as:

[0117] ;

[0118] The shortest deployment latency across all networks can be expressed as: ,or Once the shortest delay is determined, the problem can be solved. The deployment location is to be determined. middle The deployment location, that is, the determination The corresponding candidate network nodes.

[0119] Backend deployment algorithm: If Then deploy It is necessary to Calculate all destination network nodes to The link latency is considered, and among all pre-deployed network nodes, the network node with the shortest latency is selected for [the task]. The deployment. Among them... From all destination network nodes to any network node in the network The time delay can be expressed as:

[0120] ;

[0121] The shortest deployment latency across all networks can be expressed as: ,or Once the shortest delay is determined, the problem can be solved. The deployment location is to be determined. middle The deployment location, that is, the determination The corresponding candidate network nodes.

[0122] Scenario 2: If the current VNF ​​is not the first deployed VNF, determine the link latency between the target network node corresponding to the previous VNF or the target network node corresponding to the next VNF ​​and the pre-deployed network node of the current VNF, and take the network node corresponding to the shortest link latency as the candidate network node corresponding to the current VNF.

[0123] Specifically, if a front-end deployment algorithm is used, it is necessary to determine the link latency between the target network node corresponding to the previous VNF and the pre-deployed network node of the current VNF; if a back-end deployment algorithm is used, it is necessary to determine the link latency between the target network node corresponding to the next VNF ​​and the pre-deployed network node of the current VNF.

[0124] The embodiments of the present invention will explain and illustrate Case 2 using front-end deployment algorithm and back-end deployment algorithm as examples respectively:

[0125] Front-end deployment method: If Then calculate the link delay between VNFs. ,in or .at this time All the previous ones Deployment complete. You can confirm now. or Once the shortest delay is determined, the problem can be solved. The deployment location. Among them... express middle The deployed network nodes and middle The shortest distance between network nodes previously deployed in VNF. express The first in The VNF, that is, the VNF set of the th VNF. The network nodes deployed in each VNF are , express The first in The network nodes deployed in the VNF are: Calculate network nodes using Dijkstra's algorithm. The shortest distance between them, the link propagation delay can be expressed as To be determined The deployment location, that is, the determination The corresponding candidate network nodes.

[0126] Backend deployment method: If Calculate the link delay between VNFs Dijkstra's algorithm is used to calculate network nodes. The shortest distance between them, the link propagation delay can be expressed as To be determined The deployment location, that is, the determination The corresponding candidate network nodes.

[0127] Step D3: Determine whether the candidate network nodes meet the preset node and link constraints.

[0128] In practical implementation, it can be determined whether the total resources occupied by all VNFs deployed on the candidate network node are less than the CPU resources corresponding to the candidate network node, and whether the total traffic flowing through the service function chain of the candidate network node is less than the link bandwidth; if so, it is determined that the preset node and link constraints are met. For details, please refer to the explanation of the node and link constraints in the foregoing embodiments, which will not be repeated in this embodiment.

[0129] Step D4: If not, remove the candidate network node from the target deployment network and re-determine the candidate network node corresponding to the current VNF ​​until the new candidate network node meets the node and link constraints. Then, use the candidate network node as the target network node corresponding to the current VNF ​​and deploy the current VNF ​​to the target network node.

[0130] Continuing with the example of front-end deployment algorithms, if If the node resources and link resources of the corresponding candidate network nodes meet the constraints, then the network is determined to be a candidate network node. The deployment location (i.e., the location of the deployment) (corresponding target network node), determine The access scheme for the source network node. Otherwise, remove the candidate network node from the network. Repeat step D2. Until... Deployment location and user access scheme.

[0131] Continuing with the example of backend deployment algorithms, if If the node resources and link resources of the corresponding candidate network nodes meet the constraints, then the network is determined to be a candidate network node. The deployment location (i.e., the location of the deployment) (corresponding target network node), determine The access scheme for the target network node. Otherwise, remove the candidate network node from the network. Repeat step D2. Until... Deployment location and Access scheme for the target network node.

[0132] In summary, through steps S102 to S106, the embodiments of the present invention provide the following: Figure 4 The flowchart shown is a deployment algorithm for SFC at both ends of a joint user access system, including the following steps S402 to S444:

[0133] Step S402, Network initialization, given service function chain combination VNF set .

[0134] Step S404: Determine whether the access is a combination of an empty set network or a space-based network service chain, and split it into... and And will and Separately into Each subset represents a set of all service function chains that call this VNF.

[0135] Step S406, Calculate / In the middle, the number of times each VNF is called .

[0136] Step S408: Determine the importance of the VNFs at both ends. If yes, proceed to step S410; otherwise, proceed to step S426.

[0137] Step S410, determine If yes, proceed to step S412; otherwise, proceed to step S418.

[0138] Step S412, calculate Select distance Deploy the shortest network node in the internal network. .

[0139] Step S414: Determine whether the node resources and link resources meet the conditions. If yes, proceed to step S424; otherwise, proceed to step S416.

[0140] Step S416: Remove the network node from the network.

[0141] Step S418, calculate Select distance middle Deploy the network node with the shortest distance from the previous VNF deployment node. .

[0142] Step S420: Determine whether the node resources and link resources meet the conditions. If yes, proceed to step S424; otherwise, proceed to step S422.

[0143] Step S422: Remove the network node from the network.

[0144] Step S424, .

[0145] Step S426, determine If yes, proceed to step S428; if no, proceed to step S434.

[0146] Step S428, calculate Select distance Deploy the shortest network node for the internal destination network node. .

[0147] Step S430: Determine whether the node resources and link resources meet the conditions. If not, proceed to step S432. If yes, proceed to step S440.

[0148] Step S432: Remove the network node from the network.

[0149] Step S434, calculate Select distance Inside Deploy the next VNF ​​deployment node to the shortest distance network node. .

[0150] Step S436: Determine whether the node resources and link resources meet the conditions. If not, proceed to step S438. If yes, proceed to step S440.

[0151] Step S438: Remove the network node from the network.

[0152] Step S440, .

[0153] Step S442, determine If yes, proceed to step S408; if no, proceed to step S444.

[0154] Step S444 yields all user access schemes and VNF deployment schemes.

[0155] (iv) Based on the workload of different VNFs carrying SFCs, allocate CPU resources for VNFs on the network nodes where VNFs are deployed to reduce end-to-end latency. This part corresponds to step S108 above, which can be found in detail below. Figure 5 The resource allocation flowchart shown includes the following steps E1 to E4:

[0156] Step E1 involves taking the sum of the data transfer volumes corresponding to each service function chain that invokes the virtualized network function as the workload of the service function chain carried by the virtualized network function.

[0157] In practical implementation, the amount of data flowing through each service function chain of a VNF is summed. That is... ,in Indicates flow through The workload.

[0158] Step E2: If multiple virtualized network functions are deployed on the target network node where the virtualized network function is deployed, then the sum of the task volume of the service function chain of each virtualized network function deployed on the target network node is taken as the total task volume corresponding to the target network node.

[0159] In practical implementation, if multiple VNFs are deployed on the same network node, then the data traffic (i.e., the total workload) of that network node can be expressed as... .

[0160] Step E3: Based on the ratio between the task volume and the total task volume, determine the amount of resources allocated to the target network nodes deployed by the virtualization network function.

[0161] In practical implementation, the CPU resources allocated to this network node are... The traffic is allocated according to a certain proportion, that is... .

[0162] Step E4: CPU resources have been allocated to all nodes.

[0163] In summary, the service function chain deployment method for joint user access provided in this embodiment of the invention has at least the following characteristics:

[0164] (1) Most existing SAGIN service chain deployment methods are based on terrestrial networks, expanding space-based and airborne network nodes and links. The research focuses on SFC deployment and resource allocation, but neglects the user access situation and does not consider the impact of access network nodes on end-to-end latency. Therefore, this embodiment of the invention addresses the above problems by constructing a mathematical model for joint optimization of user access, SFC deployment and node resource allocation.

[0165] (2) Traditional user access methods and SFC deployment schemes cannot solve this problem. Therefore, this embodiment of the invention proposes a two-end deployment method that combines user access, SFC deployment, and node resource allocation. Based on the source and destination nodes of different service chains, the shortest distance from different nodes to the VNF is determined. Based on the shortest end-to-end latency, the deployment scheme of the user access network node and VNF is selected. According to the service chain's traffic volume, the CPU resources of the nodes are allocated to minimize the end-to-end latency.

[0166] Based on the foregoing embodiments, this invention provides a service function chain deployment device for joint user access, see [link to previous document]. Figure 6 The diagram shows a structural schematic of a service function chain deployment device for joint user access. The device mainly includes the following parts:

[0167] The network determination module 602 is used to determine the target deployment network based on a pre-built integrated air-space-ground network model deployed with a joint user access and service function chain. The integrated air-space-ground network model includes an air-based network, a space-based network, and a ground network. Each of the air-based network, space-based network, and ground network is equipped with network nodes. The user's associated terminal accesses the target deployment network from the ground network based on the service function chain set. The target deployment network is either an air-based network or a space-based network.

[0168] The algorithm determination module 604 is used to determine the target deployment algorithm based on the importance of the virtualized network functions called by each service function chain in the service function chain set; wherein, the target deployment algorithm is a front-end deployment algorithm or a back-end deployment algorithm;

[0169] Deployment module 606 is used to deploy virtualized network functions according to a target deployment algorithm with the goal of minimizing end-to-end latency, so as to determine the target network nodes in which the virtualized network functions are deployed in the target deployment network;

[0170] The resource allocation module 608 is used to determine the resource allocation amount for the target network nodes deployed by the virtualized network functions based on the task volume of the service function chain under different virtualized network functions, and to complete the deployment of the service function chain set.

[0171] The service function chain deployment device for joint user access provided in this embodiment of the invention dynamically adjusts the network to be accessed and the algorithm used for deployment by introducing user access information, and deploys virtualized network functions and allocates resources to virtualized network functions on this basis, thereby achieving optimal utilization of network resources and minimization of end-to-end latency.

[0172] In one implementation, the network determination module 602 is specifically used for:

[0173] Based on the pre-built air-ground integrated network model deployed by the joint user access and service function chain, it is determined whether the user's associated terminal is covered by the air-based network, and whether the source network node and destination network node corresponding to the service function chain in the service function chain set are covered by the air-based network.

[0174] If so, the airborne network will be identified as the target deployment network;

[0175] If not, then the space-based network will be identified as the target deployment network.

[0176] In one implementation, the algorithm determination module 604 is specifically used for:

[0177] Based on the virtualized network functions called by each service function chain in the service function chain set, the service function chains are divided to obtain a service function chain subset corresponding to each virtualized network function.

[0178] Statistical analysis is performed on each service function chain subset to determine the number of calls to each virtualized network function. The number of calls is used to describe the importance of each virtualized network function.

[0179] Based on the location of each virtualized network function in the pre-built set of virtualized network functions, the front-end virtualized network functions and the back-end virtualized network functions are determined.

[0180] The target deployment algorithm is determined based on the importance of the front-end virtualization network function and the back-end virtualization network function.

[0181] In one implementation, the deployment module 606 is specifically used for:

[0182] The deployment order of virtualized network functions is determined based on the target deployment algorithm, so that the virtualized network functions can be deployed according to the deployment order method;

[0183] During the deployment of the current virtualized network function, the current virtualized network function is pre-deployed in the network nodes of the target deployment network. With the goal of minimizing end-to-end latency, candidate network nodes corresponding to the current virtualized network function are determined from the pre-deployed network nodes based on the service function chain subset corresponding to the current virtualized network function.

[0184] Determine whether candidate network nodes meet the preset node and link constraints;

[0185] If not, the candidate network node is removed from the target deployment network, and a new candidate network node corresponding to the current virtualization network function is determined until the new candidate network node meets the node and link constraints. The candidate network node is then used as the target network node corresponding to the current virtualization network function, and the current virtualization network function is deployed to the target network node.

[0186] In one implementation, the deployment module 606 is specifically used for:

[0187] If the current virtualized network function is the first virtualized network function deployed, then determine the link latency between the source network node or destination network node of each service function chain in the service function chain subset corresponding to the current virtualized network function and the network node pre-deployed by the current virtualized network function, and take the network node corresponding to the shortest link latency as the candidate network node corresponding to the current virtualized network function.

[0188] If the current virtualized network function is not the first virtualized network function deployed, then determine the link latency between the target network node corresponding to the previous virtualized network function or the target network node corresponding to the next virtualized network function and the network node pre-deployed by the current virtualized network function, and take the network node corresponding to the shortest link latency as the candidate network node corresponding to the current virtualized network function.

[0189] In one implementation, the deployment module 606 is specifically used for:

[0190] Determine whether the total resources occupied by all virtualized network functions deployed on the candidate network node are less than the CPU resources corresponding to the candidate network node, and whether the total traffic of the service function chain flowing through the candidate network node is less than the link bandwidth.

[0191] If so, then the preset node and link constraints are satisfied.

[0192] In one implementation, the resource allocation module 608 is specifically used for:

[0193] The sum of the data transfer volume corresponding to each service function chain that calls the virtualized network function will be used as the task volume of the service function chain carried by the virtualized network function.

[0194] If multiple virtualized network functions are deployed on the target network node where the virtualized network function is deployed, then the sum of the task volume of the service function chain of each virtualized network function deployed on the target network node is taken as the total task volume corresponding to the target network node.

[0195] Based on the ratio between the task volume and the total task volume, the resource volume allocated to the target network nodes deployed by the virtualization network function is determined.

[0196] The device provided in this embodiment of the invention has the same implementation principle and technical effect as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0197] This invention provides an electronic device, specifically, the electronic device includes a processor and a storage device; the storage device stores a computer program, and the computer program, when run by the processor, executes the method described in any of the above embodiments.

[0198] Figure 7 The present invention provides a schematic diagram of the structure of an electronic device 100, which includes a processor 70, a memory 71, a bus 72 and a communication interface 73. The processor 70, the communication interface 73 and the memory 71 are connected through the bus 72. The processor 70 is used to execute executable modules, such as computer programs, stored in the memory 71.

[0199] The memory 71 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 73 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc.

[0200] Bus 72 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 7 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0201] The memory 71 is used to store programs. After receiving an execution instruction, the processor 70 executes the programs. The method executed by the device for defining the flow process disclosed in any of the foregoing embodiments of the present invention can be applied to the processor 70 or implemented by the processor 70.

[0202] The processor 70 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 70 or by instructions in software form. The processor 70 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules may reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 71. Processor 70 reads the information in memory 71 and, in conjunction with its hardware, completes the steps of the above method.

[0203] The computer program product of the readable storage medium provided in the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the foregoing method embodiments. For specific implementation, please refer to the foregoing method embodiments, which will not be repeated here.

[0204] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0205] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for deploying a service function chain for joint user access, characterized in that, include: Based on a pre-built integrated air-space-ground network model with a joint user access and service function chain deployment, a target deployment network is determined. The integrated air-space-ground network model includes an air-based network, a space-based network, and a ground network. Each of the air-based network, the space-based network, and the ground network is equipped with network nodes. The user's associated terminal accesses the target deployment network from the ground network based on the service function chain set. The target deployment network is either an air-based network or a space-based network. The target deployment algorithm is determined based on the importance of the virtualized network functions invoked by each service function chain in the service function chain set; wherein, the target deployment algorithm is a front-end deployment algorithm or a back-end deployment algorithm; With the goal of minimizing end-to-end latency, the virtualized network function is deployed according to the target deployment algorithm to determine the target network node in which the virtualized network function is deployed in the target deployment network; Based on the workload of the service function chain carried by different virtualized network functions, the resource amount allocated to the target network node deployed by the virtualized network function is determined, and the deployment of the service function chain set is completed. Based on a pre-built integrated air-space-ground network model with a joint user access and service function chain deployment, the target deployment network is determined, including: based on the pre-built integrated air-space-ground network model with a joint user access and service function chain deployment, determining whether the user's associated terminal is covered by the air-based network, and whether the source network node and destination network node corresponding to the service function chain in the service function chain set are covered by the air-based network; if yes, the air-based network is determined as the target deployment network; if no, the space-based network is determined as the target deployment network. The target deployment algorithm is determined based on the importance of the virtualized network functions invoked by each service function chain in the service function chain set. This includes: dividing the service function chain set according to the virtualized network functions invoked by each service function chain in the set, obtaining a service function chain subset corresponding to each virtualized network function; statistically analyzing each service function chain subset to determine the number of times each virtualized network function is invoked, where the number of invocations describes the importance of each virtualized network function; determining front-end and back-end virtualized network functions based on the position of each virtualized network function in a pre-constructed set of virtualized network functions; and determining the target deployment algorithm based on the importance of the front-end and back-end virtualized network functions.

2. The service function chain deployment method for joint user access according to claim 1, characterized in that, With the goal of minimizing end-to-end latency, the virtualized network functions are deployed according to the target deployment algorithm to determine the target network nodes where the virtualized network functions are deployed in the target deployment network, including: The deployment order of the virtualized network functions is determined based on the target deployment algorithm, and the virtualized network functions are deployed in accordance with the deployment order; During the deployment of the current virtualized network function, the current virtualized network function is pre-deployed in the network nodes of the target deployment network, and with the goal of minimizing end-to-end latency, candidate network nodes corresponding to the current virtualized network function are determined from the pre-deployed network nodes based on the service function chain subset corresponding to the current virtualized network function. Determine whether the candidate network nodes meet the preset node and link constraints; If not, the candidate network node is removed from the target deployment network, and a new candidate network node corresponding to the current virtualization network function is determined until the new candidate network node satisfies the node and link constraints. The candidate network node is then used as the target network node corresponding to the current virtualization network function, and the current virtualization network function is deployed to the target network node.

3. The service function chain deployment method for joint user access according to claim 2, characterized in that, With the goal of minimizing end-to-end latency, candidate network nodes corresponding to the current virtualized network function are determined from the pre-deployed network nodes based on the service function chain subset corresponding to the current virtualized network function, including: If the current virtualized network function is the first virtualized network function deployed, then determine the link latency between the source network node or destination network node of each service function chain in the service function chain subset corresponding to the current virtualized network function and the network node pre-deployed by the current virtualized network function, and take the network node corresponding to the shortest link latency as the candidate network node corresponding to the current virtualized network function. If the current virtualized network function is not the first deployed virtualized network function, then determine the link latency between the target network node corresponding to the previous virtualized network function or the target network node corresponding to the next virtualized network function and the network node pre-deployed by the current virtualized network function, and take the network node corresponding to the shortest link latency as the candidate network node corresponding to the current virtualized network function.

4. The service function chain deployment method for joint user access according to claim 2, characterized in that, Determining whether the candidate network nodes meet preset node and link constraints includes: Determine whether the total resource usage of all the virtualized network functions deployed on the candidate network node is less than the CPU resources corresponding to the candidate network node, and whether the total traffic of the service function chain flowing through the candidate network node is less than the link bandwidth; If so, then the preset node and link constraints are satisfied.

5. The service function chain deployment method for joint user access according to claim 1, characterized in that, Based on the workload of the service function chain carried by different virtualized network functions, the amount of resources allocated to the target network nodes deployed by the virtualized network function is determined, including: The sum of the data transmission volume corresponding to each service function chain that invokes the virtualized network function is used as the task volume of the virtualized network function carrying the service function chain. If multiple virtualized network functions are deployed on the target network node where the virtualized network function is deployed, then the sum of the task volume of the service function chain carried by each virtualized network function deployed on the target network node is taken as the total task volume corresponding to the target network node. Based on the ratio between the task volume and the total task volume, the resource volume allocated to the virtualization network function is determined for the target network node deployed by the virtualization network function.

6. A service function chain deployment device for joint user access, characterized in that, include: The network determination module is used to determine the target deployment network based on a pre-built integrated air-space-ground network model deployed with a joint user access and service function chain. The integrated air-space-ground network model includes an air-based network, a space-based network, and a ground network. Each of the air-based network, the space-based network, and the ground network is equipped with network nodes. The user's associated terminal accesses the target deployment network from the ground network based on the service function chain set. The target deployment network is either an air-based network or a space-based network. An algorithm determination module is used to determine a target deployment algorithm based on the importance of the virtualized network functions invoked by each service function chain in the service function chain set; wherein, the target deployment algorithm is a front-end deployment algorithm or a back-end deployment algorithm; The deployment module is used to deploy the virtualized network function according to the target deployment algorithm with the goal of minimizing end-to-end latency, so as to determine the target network node where the virtualized network function is deployed in the target deployment network; The resource allocation module is used to determine the resource allocation amount of the target network node deployed by the virtualization network function as the amount of tasks carried by the service function chain based on the different virtualization network functions, and to complete the deployment of the service function chain set. The network determination module is specifically used to: determine whether the user's associated terminal is covered by the air-based network based on the pre-built air-space-ground integrated network model of the joint user access and service function chain deployment, and whether the source network node and destination network node corresponding to the service function chain in the service function chain set are covered by the air-based network; if so, the air-based network is determined as the target deployment network; if not, the space-based network is determined as the target deployment network. The algorithm determination module is specifically used for: dividing the service function chain set according to the virtualized network functions called by each service function chain in the service function chain set, obtaining a service function chain subset corresponding to each virtualized network function; statistically analyzing each service function chain subset to determine the number of times each virtualized network function is called, the number of calls being used to describe the importance of each virtualized network function; determining the front-end virtualized network function and the back-end virtualized network function based on the position of each virtualized network function in the pre-constructed virtualized network function set; and determining the target deployment algorithm based on the importance of the front-end virtualized network function and the importance of the back-end virtualized network function.

7. An electronic device, characterized in that, The method includes a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement the method of any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when invoked and executed by a processor, cause the processor to perform the method described in any one of claims 1 to 5.

Images