Satellite-ground convergence network routing method and system based on serial and parallel hybrid service function chain constraint

Abstract

The invention discloses a satellite-ground convergence network routing method and system based on serial-parallel hybrid service function chain constraint, and belongs to the field of satellite communication, and the method comprises the steps: S1, initializing network parameters, and dividing network nodes; s2, performing initialization representation on the task request; s3, based on the represented task request, performing functional decomposition on nodes in the initialized network; s4, based on the represented task request, dividing a network planning period according to connectivity of nodes in the network; s5, constructing a multifunctional time expansion graph in a network planning period based on a functional decomposition result of the nodes; s6, for the represented task request, setting a network task constraint in the multifunctional time extension graph; and S7, adjusting the data volume of the virtual flow to meet the constraint condition so as to guarantee the optimal network maximum flow performance in the plan period. According to the method, the network flow can be maximized, and the utilization efficiency of the multi-dimensional resources of the satellite-ground fusion network is improved.

Description

Technical Field

[0001] This invention relates to the field of satellite communications, and more specifically, to a satellite-ground converged network routing method and system based on serial-parallel hybrid service function chain constraints. Background Technology

[0002] The inherent heterogeneity and dynamism of space-ground converged networks lead to low network resource utilization. Furthermore, these networks face highly dynamic network topologies due to the high-speed movement of network nodes. Relative motion between satellites results in inter-satellite and satellite-to-ground link connections being intermittent, and the availability of communication resources changing over time. Traditional static diagrams for modeling network topology are insufficient to represent the heterogeneous resources in dynamic space-ground converged networks, cannot depict time-varying network topologies, and are unsuitable for time-varying space-ground converged networks. Moreover, a large amount of data in the network typically needs to be transmitted to the destination node via a store-and-forward method. Therefore, applying virtualization technology in space-ground converged networks decouples network functions from their physical operating entities, allowing for convenient and flexible deployment of network functions anywhere in the network using software.

[0003] In a software-defined space-ground converged network, each task can be characterized by a pre-defined sequence of service functions, called a service function chain. To fulfill a task request, the software-defined space-ground converged network must satisfy the constraint of sequentially traversing the pre-defined sequence of service functions, i.e., the service function chain. The path traversed by the nodes deploying the service functions forms the corresponding service function chain route. Among these, maximum flow routing solves the routing decision problem of maximizing traffic within the corresponding service function chain route.

[0004] Currently, typical maximum flow routing algorithms in satellite-ground converged networks, such as those for networks with limited transceiver numbers and energy-constrained networks, do not consider serial or hybrid serial-parallel service function chain constraints and cannot be directly applied to software-defined satellite-ground converged networks. Furthermore, current routing algorithms for time-varying satellite-ground converged networks primarily focus on maximum flow routing algorithms that satisfy serial service function chain constraints, only considering the serial service function chain constraint and neglecting the hybrid serial-parallel service function chain constraint. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a routing method and system for satellite-ground converged networks based on serial-parallel hybrid service function chain constraints, which can maximize network flow and improve the utilization efficiency of multi-dimensional resources of satellite-ground converged networks.

[0006] The objective of this invention is achieved through the following solution: A satellite-to-ground converged network routing method based on serial-parallel hybrid service function chain constraints includes the following steps: S1, initialize network parameters and divide network nodes; S2, initializes the task request representation; S3, based on the represented task request, performs functional decomposition on the nodes in the initialized network; S4, based on the represented task requests, divide the network planning cycle according to the connectivity of nodes in the network; S5. Based on the functional decomposition results of the nodes, construct a multi-functional time extension graph within the network planning cycle; S6, For the represented task request, set network task constraints in the multifunctional time extension graph; S7. Adjust the data volume of the virtual stream to meet the constraints in step S6, so as to ensure the optimal maximum network flow performance within the planning period.

[0007] Further, in step S1, the initialization of network parameters and the partitioning of network nodes specifically include the following sub-steps: Initialize the satellite-ground fusion network node set as follows The number of network nodes is The network nodes are divided into source nodes and destination nodes. Nodes that only provide communication and storage and do not provide computational functions are classified as non-functional nodes, while nodes that provide both communication, storage, and computational functions are classified as functional nodes. Based on this division, the network node set is created. Represented as: ; in, and These represent the source node and the destination node of the task, respectively. A collection of functional nodes. It is a set of non-functional nodes. .

[0008] Furthermore, in step S2, the initialization of the task request specifically includes the following sub-steps: S2a uses a service function chain with a pre-determined functional sequence. Characterize the task request, where, Indicates the number of tasks received in the task flow Each service function This indicates the number of service functions deployed in the network; S2b divides the service function chain into stages and defines... Indicates the service function in the service chain The execution phase; the first step in the service function chain The set of service functions involved in a stage is represented as ,in, Indicates the first The number of service functions involved in the phase; The service functions involved in the initial stage are denoted as those flowing out from the source node, i.e. , ; S2c, using virtual streams This indicates that the first... Phase service functions Virtual stream This indicates a virtual stream that has just flowed out of the source node and has not yet received any service functions.

[0009] Furthermore, in step S3, the functional decomposition of nodes in the initialized network specifically includes the following sub-steps: S3a, functional nodes The set of service functions that can be deployed is represented as follows: ; in, Represents functional nodes The total number of service functions that can be deployed; S3b will be the functional node The virtual decomposition is divided into three types of components: a virtual child node. and Virtual functional nodes and virtual transmission links and ,in, Represented as a functional node Virtual functional nodes in the virtual decomposition can provide service functions for tasks. ; This represents a virtual child node. To virtual function nodes Communication links, This refers to virtual functional nodes. To virtual child nodes Communication links; S3c decomposes all functional nodes to obtain the decomposed set of network nodes: The set of virtual child nodes from which all functional nodes are decomposed is represented as: ; The set of virtual functional nodes decomposed from all functional nodes is represented as: ; The set of virtual functional nodes obtained by virtual decomposition of functional nodes is represented as: ; It will be able to support deployment service functions. The set of virtual functional nodes is represented as: .

[0010] Furthermore, in step S4, the step of dividing the network planning period according to the connectivity of nodes in the network specifically includes the following sub-steps: Set up satellite-ground integrated network nodes Import the data into Satellite Tool Kit software. Based on the connectivity between network nodes obtained from the ephemeris table in the software, and based on the connectivity of the network nodes, adjust the network planning cycle. Divided into A continuous time slot, i.e. ,in, each Indicates the first A time interval.

[0011] Further, in step S5, a multi-functional time-spreading graph is constructed within the network planning period, specifically including the following sub-steps: S5a, initialize a blank card. A directed graph of layers, where the first layer is... The time interval of the layered directed graph is ; S5b adds the source node of the task in the network at each time interval of the directed graph. , destination node Virtual child nodes decomposed from all functional nodes Virtual sub-functional nodes decomposed from all functional nodes This yields four types of node sets, which together form a multifunctional time-extended node graph. S5c adds transmission links, storage links, and virtual transmission links to the multi-functional time extension node graph to obtain the multi-functional time extension graph.

[0012] Furthermore, in step S6, the network task constraints include communication capacity constraints, storage capacity constraints, computing capacity constraints, serial-parallel service function chain constraints, traffic conservation constraints, and task requirement constraints.

[0013] Furthermore, in step S6, setting network task constraints, when setting communication capacity constraints, specifically includes the following sub-steps: The total amount of data transmitted by all virtual streams on the transmission link or virtual transmission link is limited to not exceeding the communication capacity of the transmission link or virtual transmission link.

[0014] Furthermore, in step S6, setting network task constraints, when setting storage capacity constraints, specifically includes the following sub-steps: The total amount of data stored on a storage link by all virtual streams is limited to not exceeding the storage capacity of that storage link.

[0015] Further, in step S6, setting network task constraints, when setting computational capacity constraints, specifically includes the following sub-steps: Limited to time interval Internal network functional nodes Decomposition Virtual functional nodes The computing capacity consumed does not exceed that of network functional nodes. The computing power provided.

[0016] Further, in step S6, setting network task constraints, when setting serial-parallel service function chain constraints and traffic conservation constraints, specifically includes the following sub-steps: For virtual child nodes Limited virtual stream The amount of data flowing into a virtual child node is equal to the amount of data flowing out of it. For virtual functional nodes According to the first step of the serial-parallel service function chain and the Number of service functions involved in the stage and There are four possible scenarios: Scenario 1, Single Inbound, Single Outbound: If the service function chain is... The number of service function types involved in this phase is 1, that is , No. The number of service function types involved in this phase is 1, that is If the virtual function node provides a service function chain for the task flow, then... Service functions involved in the phase Then virtual stream The stream flowing into this virtual function node will be processed and transformed into a virtual stream. Flow out from this node; in addition, define For virtual streams and virtual stream The scaling factor flows into the virtual functional nodes. virtual stream Data volume multiplied by This is equivalent to the outflow of this virtual functional node. virtual stream The amount of data; Scenario 2, Multiple Inbounds, Single Outbound: If the service function chain... The number of service function types involved in a phase is greater than 1, that is... , No. The number of service function types involved in this phase is 1, that is If the virtual function node provides a service function chain for the task flow, then... Service functions involved in the phase Then, after receiving the first service function chain... Virtual flow of service functions involved in the phase After being merged proportionally, the streams flowing into this virtual functional node will be processed and transformed into virtual flows. Flowing out from this node; multiple virtual flows flowing into this virtual function node. After being merged according to the appropriate ratio, the virtual streams flow into the virtual function nodes, and the merged virtual streams flow into these virtual function nodes to receive service functions; Define For virtual streams Inflow into virtual functional nodes The ratio represents the virtual stream. proportionally Inflow into virtual function nodes; for any two inflowing virtual function nodes virtual stream and virtual stream The amount of data should meet the following requirements: virtual stream Data volume multiplied by equal to virtual stream Data volume multiplied by ;definition For the merged virtual stream and virtual stream The scaling factor flows into the virtual functional nodes proportionally. virtual stream Data volume multiplied by This is equivalent to the virtual flow flowing out of the virtual functional node. The amount of data; Scenario 3, Single Input Multiple Output: If the service function chain... The number of service functions involved in this phase is 1, that is , No. The number of service function types involved in a phase is greater than 1, that is... If the virtual function node provides a service function chain for the task flow, then... Service functions involved in the phase Then virtual stream The data flowing into this virtual functional node will be processed and transformed. The same virtual stream The stream flowing out from this node; the virtual stream flowing out from the virtual function node. copy Share; definition For virtual streams and virtual stream The scaling factor flows into the virtual functional nodes. virtual stream Data volume multiplied by This is equivalent to the virtual flow flowing out of the virtual functional node. The amount of data; Scenario 4, Multiple Inputs and Multiple Outputs: If the service function chain... The number of service function types involved in a phase is greater than 1, that is... , No. The number of service function types involved in a phase is greater than 1, that is... If the virtual function node provides a service function chain for the task flow, then... Service functions involved in the phase Then all service function chains have received the first... Virtual flow of service functions involved in the phase The stream flowing into this virtual function node will be processed and transformed into a virtual stream. Flowing out from this node; multiple virtual flows flowing into this virtual function node. After being merged proportionally, the service functions flow into this virtual functional node; definition For virtual streams Inflow into virtual functional nodes The ratio represents the virtual stream. proportionally Inflow into virtual function nodes; for any two inflowing virtual function nodes virtual stream and virtual stream The amount of data satisfies: virtual stream Data volume multiplied by equal to virtual stream Data volume multiplied by ;definition For the merged virtual stream and virtual stream The scaling factor; at the same time, the SFC's first If the number of service functions involved in a stage is greater than 1, the virtual stream flowing out from the virtual function node will be... copy Shares; flowing into virtual functional nodes proportionally. virtual stream Data volume multiplied by This is equivalent to the virtual flow flowing out of the virtual functional node. The amount of data; For the source node Limited virtual stream Outflow source node The amount of data is equal to 0; For the destination node Limited virtual stream Outgoing destination node The amount of data is equal to 0.

[0017] Further, in step S6, setting network task constraints, when setting task requirement constraints, specifically includes the following sub-steps: During the planning period Internally, from all source node replicas Outgoing virtual stream The virtual stream will be based on the number of service functions involved in Phase 1. copy A copy, that is, a copy flowing out from all source node replicas. The same virtual stream If SFC's first The service functions involved in the phase are greater than 1, that is Then, before flowing into the destination node replica, all SFCs that have been received will be... The virtual flows of service functions involved in the phase flow proportionally into the destination node replica; define For virtual streams The proportion of virtual flows flowing into the destination node replica, and the total virtual flows flowing out of all source node replicas. The total amount of data is multiplied by the amount of data from phase 1 to phase 2. All scaling factors of the stages, the scaling factor of the parallel stages, and This equals all virtual flows flowing into all destination node replicas. Total data volume; If the service function chain is... The number of service functions involved in a phase is greater than 1, that is During the planning period Internally, it flows into all support. Virtual stream of virtual function node replicas for service functions The amount of data is equal.

[0018] A satellite-ground converged network routing system based on serial-parallel hybrid service function chain constraints includes a computer device, which includes a processor and a memory. The memory stores a computer program that executes the method described in any of the preceding claims when the computer program is loaded by the processor.

[0019] The beneficial effects of this invention include: (1) In a software-defined satellite-ground converged network, this invention characterizes the flow conservation condition satisfying the serial-parallel hybrid service function chain based on a multi-functional time spread graph. Specifically, the multi-functional time spread graph is used to model the time-varying network, which can characterize multiple parallel and multiple serial computing functions of the task flow within the same node. Based on the multi-functional time spread graph, this invention characterizes the flow conservation under the constraint of the serial-parallel hybrid service function chain and classifies it into four types: single-input single-output, multiple-input single-output, single-input multiple-output, multiple-input single-output, and multiple-input multiple-output.

[0020] (2) This invention considers the multi-dimensional resources of communication, storage and computing in the space-ground integrated network. By setting constraints such as serial-parallel hybrid service function chain, communication capacity, storage capacity, computing capacity, flow conservation and task requirement, the problem of joint management of communication, storage and computing resources is transformed into a flow routing problem in a multi-functional time extension graph. This allows tasks to complete task requests while satisfying the communication, storage and computing resource constraints of nodes, and can effectively guarantee different service quality requirements of tasks. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram illustrating the usage scenario; Figure 2 This is a flowchart illustrating the steps of the method in an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the connectivity between nodes in the network during the planning period in an embodiment of the present invention. Figure 4 This is a schematic diagram of the virtual child nodes and virtual sub-functional nodes obtained by decomposing functional nodes in an embodiment of the present invention; Figure 5 This is the initial blank directed graph in the embodiments of the present invention; Figure 6 This is a multifunctional time extension graph constructed in an embodiment of the present invention; Figure 7 This is a simulation comparison chart of storage capacity and maximum network flow in an embodiment of the present invention; Figure 8 This is a simulation comparison chart of computational capacity and maximum network flow in an embodiment of the present invention. Detailed Implementation

[0023] All features disclosed in all embodiments of this specification, or steps in all methods or processes implied in the disclosure, may be combined and / or extended or replaced in any way, except for mutually exclusive features and / or steps.

[0024] The specific implementation process of this invention is as follows: Given the problems mentioned in the background, and considering the inherent resource heterogeneity and topology dynamics of converged space-ground networks, this invention introduces virtualization technology into converged space-ground networks to develop a routing method based on a hybrid serial-parallel service function chain constraint. In software-defined converged space-ground networks, existing algorithms have only studied routing algorithms that satisfy serial service function chain constraints, without considering hybrid serial-parallel service function chain constraints. To address the shortcomings of the aforementioned routing algorithms for software-defined space-ground networks, this invention proposes a routing method that satisfies the hybrid serial-parallel service function chain constraint, thereby maximizing network flow and improving the utilization efficiency of the multi-dimensional resources of the converged space-ground network.

[0025] In a preferred embodiment of the present invention, a satellite-ground converged network routing method based on serial-parallel hybrid service function chain constraints is specifically proposed to maximize network flow while satisfying the serial-parallel hybrid service function chain constraints, thereby improving the efficient utilization of multi-dimensional resources of the satellite-ground converged network. The method specifically includes the following steps: Step S1: Initialize network parameters and divide network nodes. Specifically, initialize the set of satellite-ground fusion network nodes as follows: The number of network nodes is The network nodes are divided into source nodes and destination nodes. Nodes that only provide communication and storage and do not provide computational functions are classified as non-functional nodes, while nodes that provide both communication, storage, and computational functions are classified as functional nodes. Based on this division, the network node set is created. Represented as: ; in, and These represent the source node and the destination node of the task, respectively. A collection of functional nodes. It is a set of non-functional nodes. ; Step S2: Initialize the representation task request; S2a uses a service function chain with a pre-determined functional sequence. , representing a task request, where, Indicates the number of tasks received in the task flow Each service function This indicates the number of service functions deployed in the network; Step S2b: Divide the service function chain into stages and define... Indicates the service function in the service chain The execution phase; the first step in the service function chain The set of service functions involved in a stage is represented as ,in, Indicates the first The number of service functions involved in the phase. The service functions involved in the initial stage are denoted as those flowing out from the source node, i.e. ,therefore ; Step S2c, using virtual streams This indicates that the first... Phase service functions Virtual stream This indicates a virtual stream that has just flowed out of the source node and has not yet received any service functions.

[0026] Step S3: Decompose the network functional nodes: Step S3a, the functional nodes The set of service functions that can be deployed can be represented as follows: ; in, Represents functional nodes The total number of service functions that can be deployed; Step S3b, add functional nodes The virtual decomposition is divided into three types of components: a virtual child node. and Virtual functional nodes and virtual transmission links and ,in, Represented as a functional node Virtual functional nodes in the virtual decomposition can provide service functions for tasks. ; This represents a virtual child node. To virtual function nodes Communication links, This refers to virtual functional nodes. To virtual child nodes Communication links; Step S3c: Decompose all functional nodes to obtain the decomposed set of network nodes. The set of virtual child nodes from which all functional nodes are decomposed is represented as: ; The set of virtual functional nodes decomposed from all functional nodes is represented as: ; The set of virtual functional nodes obtained by virtual decomposition of functional nodes is represented as: ; It will be able to support deployment service functions. The set of virtual functional nodes is represented as: ; Step S4: Divide the network planning cycle according to the network node connectivity. Set up the satellite-ground fusion network nodes. Import the data into Satellite Tool Kit software. Based on the connectivity between network nodes obtained from the ephemeris table in the software, and based on the connectivity of the network nodes, adjust the network planning cycle. Divided into A continuous time slot, i.e. ,in, each Indicates the first A time interval. Assuming in a time slot The internal network topology remains constant, but the network topology across time slots may change.

[0027] Step S5, construct a multi-functional time extension graph: Step S5a: Initialize a blank sheet. A directed graph of layers, where the first layer is... The time interval of the layered directed graph is ; Step S5b: Add the source node of the task in the network during each time interval of the directed graph. , destination node Virtual child nodes decomposed from all functional nodes Virtual sub-functional nodes decomposed from all functional nodes This yields four types of node sets, forming a multifunctional time-extended node graph; Step S5c: Add transmission links, storage links, and virtual transmission links to the multi-functional time extension node graph to obtain the multi-functional time extension graph; Step S6: Set communication capacity constraints, storage capacity constraints, computing capacity constraints, traffic conservation constraints, serial-parallel hybrid service function chain constraints, and task requirement constraints; Step S7: Adjust the data volume of all virtual streams to meet all constraints in step S6, thereby ensuring optimal network maximum flow performance within the planning period.

[0028] More specifically, in other embodiments, a satellite-to-ground converged network routing method based on serial-parallel hybrid service function chain constraints is provided. (Refer to...) Figure 1 There is a task request in the network, which is provided by a service function chain, as follows: Figure 1 As shown. The space-ground integrated network scenario consists of 4 satellites. , , , and 1 ground station The mission consists of two parts: the source node and the destination node, which are the first satellite. and ground station The second satellite Provide service functions and The third satellite Provide service functions The fourth satellite Provide service functions and The network planning cycle is... During the planning period, the connectivity between satellites and between satellites and ground stations, such as Figure 3 As shown, Figure 3 Each horizontal and vertical axis corresponds to a pair of nodes in the graph, where the horizontal axis represents time and the vertical axis represents connectivity. State 1 indicates connectivity and state 0 indicates disconnection.

[0029] Reference Figure 2 The specific implementation steps of this embodiment in the above scenario are as follows: Step 1: Initialize network parameters and divide network nodes.

[0030] The space-ground integrated network scenario consists of 4 satellites , , , and 1 ground station Composition, i.e., the number of network nodes is Initialize the network node set as The source node and destination node of the mission are the first satellite. and ground station Network nodes , , This is a functional node. The second functional node. Provide service functions and The third functional node Provide service functions The fourth functional node Provide service functions and .

[0031] Based on the above division, the network node set Represented as ,in, It is a collection of functional nodes.

[0032] Step 2: Initialize the representation task request.

[0033] 2.1) A set of serial-parallel hybrid service function chains with a predetermined order represents the task request, referring to... Figure 1 The service function chain has a length of 3. The service function chain is divided into stages, where... , , This means service functionality and The request should be made in Then, the processes are executed in parallel, and the service functions are received in parallel. and Then, the service functions are executed serially. At the same time, the first step in the service function chain... The set of service functions involved in each stage is represented as follows: , , ,in, , , .

[0034] 2.2) After receiving the first... Phase service functions virtual stream This indicates that the first [item] has just been received. Phase service functions The task is to use a virtual stream that has not yet received any service functions. This indicates a task that has just flowed out from the source node and has not yet received any service functions. If the first [node] in the service function chain... If the number of service functions involved in a stage is greater than 1, then the virtual stream... Indicates the first The stage receives a set of tasks with different service functions.

[0035] 2.3) Transform the changes between virtual streams according to the service functions received by the virtual stream: If the first has been received Service functions involved in the phase virtual stream Inflow provides the first Service functions involved in the phase The functional node received the service function. Then the already received number Service functions involved in the phase virtual stream Converted to received number Service functions involved in the phase virtual stream Outflow; If the first has been received Service functions involved in the phase virtual stream Flowing into non-functional nodes, or virtual child nodes, or not providing the third Service functions involved in the phase Functional nodes that are not processed will still result in a virtual stream. .

[0036] Step 3, decompose the network functional nodes, such as Figure 4 As shown.

[0037] 3.1) Represent the functional set of functional nodes: The second functional node The set of service functions that can be deployed can be represented as follows: ; The third functional node The set of service functions that can be deployed can be represented as follows: ; The fourth functional node The set of service functions that can be deployed can be represented as follows: ; 3.2) Decompose each functional node: The second functional node Virtually decomposed into a virtual child node and two virtual functional nodes and two types of virtual transport links and The second functional node The first virtual functional node of virtual decomposition The service functions involved in the first phase of the service function chain The second virtual functional node The service functions involved in the third phase of the service function chain .

[0038] The third functional node Virtually decomposed into a virtual child node and a virtual functional node and two types of virtual transport links The third functional node Virtual functional nodes of virtual decomposition The service functions involved in the second phase of the service function chain .

[0039] The fourth functional node Virtually decomposed into a virtual child node and two virtual functional nodes and two types of virtual transport links and The fourth functional node The first virtual functional node of virtual decomposition The service functions involved in the second phase of the service function chain The second virtual functional node The service functions involved in the third phase of the service function chain .

[0040] 3.3) Decompose all functional nodes to obtain the decomposed set of network nodes. ,in It is represented as the set of virtual child nodes decomposed from all functional nodes in the network, that is, the set is composed of , , It consists of three virtual child nodes; It is represented as the set of virtual functional nodes decomposed from all functional nodes, that is, the set is composed of , , , , It consists of five virtual functional nodes; .

[0041] It will be able to support deployment service functions. The set of virtual functional nodes is represented as ; It will be able to support deployment service functions. The set of virtual functional nodes is represented as ; It will be able to support deployment service functions. The set of virtual functional nodes is represented as ; It will be able to support deployment service functions. The set of virtual functional nodes is represented as .

[0042] Step 4: Divide the network planning cycle according to the connectivity of network nodes.

[0043] 4.1) Import the satellite-ground fusion network node set into the Satellite Tool Kit software. Based on the connectivity between network nodes obtained from the ephemeris table in the software, and based on the connectivity of the above 5 network nodes, determine the network planning cycle. Divided into A series of unequal time intervals ,in , , ,like Figure 3 As shown.

[0044] 4.2) Based on the connectivity of network nodes, determine the transmission relationship between nodes within each time interval: First time interval Inside, source node A second functional node can be given Transmit data; Second time interval Inside, the second functional node A third functional node can be given and the fourth functional node Transmit data; The third time interval Inside, the second functional node A fourth functional node can be provided. Transmitting data, the third functional node A fourth functional node can be provided. Transmitting data, the fourth functional node Can give the destination node Transmitting data.

[0045] Step 5: Construct a multifunctional time extension graph.

[0046] 5.1) Initialize a blank sheet A layered directed graph, wherein the first layer is... The time interval of the layered directed graph is , ,like Figure 5 As shown; 5.2) In each time interval of the directed graph Add the source node, destination node, virtual child nodes decomposed from all functional nodes, and virtual functional nodes decomposed from all functional nodes to form a multi-functional time-extended node graph, as shown below. Figure 6 As shown, where: The source node set of the multi-functional time-extended node graph is That is, the set consists of , , It consists of three nodes, among which, Indicates the source node In the time interval The copy inside, , ; The target node set of the multi-functional time-extended node graph is That is, the set consists of , , It consists of three nodes, among which, Indicates the destination node In the time interval The copy inside, , ; The set of virtual child nodes in the multi-functional time-extended node graph is That is, the set consists of , , , , , , , , It consists of nine nodes, among which, Represents virtual child nodes In the time interval The copy inside, , ; The set of virtual functional nodes in the multi-functional time-extended node graph is as follows: That is, the set consists of , , , , , , , , , , , , , , It consists of fifteen nodes, among which, Represents virtual functional nodes In the time interval The copy inside, , .

[0047] 5.3) Add links to the multi-functional time-extended node graph, such as... Figure 6 As shown: 5.3.1) During the time interval Internally, transmission links are added based on the connectivity between network nodes: If in the time interval Inside, the source node in the network Able to give the target node Data transmission occurs in the source node of the multi-functional time-extended node graph. With the target node Add a directed line segment between them ; If in the time interval Inside, the source node in the network Able to give the first Functional nodes Data transmission occurs in the source node of the multi-functional time-extended node graph. With the virtual child nodes Add a directed line segment between them ; If in the time interval Inside, in the network Functional nodes Able to give the first Functional nodes Data transmission occurs in the multi-functional time-extended node diagram at the [number]th node. virtual child nodes With the virtual child nodes Add a directed line segment between them ; If in the time interval Inside, in the network Functional nodes Able to give the target node Data transmission occurs in the multi-functional time-extended node diagram at the [number]th node. virtual child nodes With the target node Add a directed line segment between them ; 5.3.2) Add storage links: Add a time interval from the first node to the next node in the multi-functional time-expanded node graph. Nodes in time intervals To the Nodes in time intervals directed line segments , , ; Add a line between adjacent time intervals of each destination node in the multi-functional time-extended node graph, starting from the first node. Nodes in time intervals To the Nodes in time intervals directed line segments , , ; Add a time interval from the first virtual child node to the adjacent time intervals of each virtual child node in the multi-functional time-extended node graph. Nodes in time intervals To the Nodes in time intervals directed line segments , , , , ; 5.3.3) Add a virtual transport link: Each virtual child node in the multi-functional time-extended node graph Its corresponding virtual sub-functional nodes Add two directed line segments between them and Thus, we obtain the following: Figure 6 The multi-functional time extension diagram is shown.

[0048] Step 6: Set network task constraints.

[0049] The constraints include five types: communication capacity constraints, storage capacity constraints, computing capacity constraints, serial-parallel service function chain constraints, flow conservation constraints, and task requirement constraints. The specific settings are as follows: 6.1) The communication capacity constraint is set, which limits the total amount of data transmitted by all virtual streams on the transmission link or virtual transmission link to not exceeding the communication capacity of the transmission link or virtual transmission link: 6.1.1) For communication links The total amount of data transmitted by all virtual streams on the communication link cannot exceed the communication capacity of that communication link, i.e.

[0050] in, Indicates the time interval Inner Virtual Stream In communication link The cumulative amount of data transmitted; Indicates communication link The communication capacity, which is expressed in time intervals. Internal task flow can be achieved through communication links The maximum cumulative amount of data transmitted. This is a set of communication links.

[0051] 6.1.2) For virtual transmission links Only virtual streams are allowed. Used, but not allowed in other stages The use of virtual streams, i.e.

[0052]

[0053] in, Represents virtual stream In virtual link The amount of data transmitted; It is a set of virtual transmission links from virtual child nodes to virtual sub-functional nodes.

[0054] 6.1.3) For virtual transmission links Only virtual streams are allowed. Used, but not allowed in other stages The use of virtual streams, i.e.

[0055]

[0056] in, Represents virtual stream In virtual link The amount of data transmitted; It is a set of virtual transmission links from virtual child nodes to virtual sub-functional nodes.

[0057] 6.2) Set storage capacity constraints, which limit the total amount of data stored on the storage link for all virtual streams to not exceed the storage capacity of the storage link.

[0058] in, Represents virtual stream In storage link The total amount of data stored on the upper part, Indicates storage link The communication capacity, which is expressed in time intervals. Internal task flow can be achieved in the storage link The maximum cumulative amount of data transmitted.

[0059] 6.3) Set computational capacity constraints, which are limited within a time interval. Internal network functional nodes Decomposition Virtual functional nodes The computing capacity consumed cannot exceed that of the network functional nodes. The computing power provided, i.e.

[0060] in, Represents virtual stream In virtual link The amount of data transmitted; when the first The service function involved in this phase is 1, namely... When, define The computation factor is a single-input, single-output factor, which represents the virtual stream processed per unit of data volume. And convert it into a virtual stream The number of computing units required. When the... The service functions involved in the phase are greater than 1, that is When, define The calculation factor is a multi-input single-output factor, which represents the virtual stream processed per unit of data after merging. And convert it into a virtual stream. The number of computing units required; Indicates the first time interval Internal functional nodes It has computing power; Functional nodes Decomposition A set of virtual functional nodes.

[0061] 6.4) Set constraints on the serial-parallel service function chain and traffic conservation constraints, which include four aspects: virtual child nodes, virtual function nodes, source nodes, and destination nodes. The specific implementation is as follows: 6.4.1) For virtual child nodes It is a limited virtual stream The amount of data flowing into a virtual child node is equal to the amount of data flowing out of it, that is:

[0062] 6.4.2) For virtual functional nodes According to the first step of the serial-parallel service function chain and the Number of service functions involved in the stage and There are four possible scenarios: Scenario 1, Single Inbound, Single Outbound: If the service function chain is... The number of service function types involved in this phase is 1, that is , No. The number of service function types involved in this phase is 1, that is If the virtual function node provides a service function chain for the task flow, then... Service functions involved in the phase Then virtual stream The stream flowing into this virtual function node will be processed and transformed into a virtual stream. Outflow from this node. Additionally, services are received by the virtual function node. virtual stream The data volume may be changed. Therefore, define... For virtual streams and virtual stream The scaling factor. Therefore, the flow into virtual functional nodes. virtual stream Data volume multiplied by This is equivalent to the outflow of this virtual functional node. virtual stream The amount of data, i.e.

[0063] Scenario 2, Multiple Inbounds, Single Outbound: If the service function chain... The number of service function types involved in a phase is greater than 1, that is... , No. The number of service function types involved in this phase is 1, that is If the virtual function node provides a service function chain for the task flow, then... Service functions involved in the phase Then, after receiving the first service function chain... Virtual flow of service functions involved in the phase After being merged proportionally, the streams flowing into this virtual functional node will be processed and transformed into virtual flows. Flowing out from this node. Multiple virtual flows flowing into this virtual function node. The virtual flows must be merged according to a certain ratio before flowing into the virtual function node. The merged virtual flows then flow into that virtual function node to receive service functions. Therefore, the following definition is provided. For virtual streams Inflow into virtual functional nodes The ratio represents the virtual stream. proportionally Flowing into virtual function nodes. Therefore, for any two flowing into virtual function nodes... virtual stream and virtual stream The amount of data should meet the following requirements: virtual stream Data volume multiplied by equal to virtual stream Data volume multiplied by ,Right now: In addition, the virtual streams after proportional merging Flowing into the virtual functional node to receive service functions The amount of data may be changed. Definition For the merged virtual stream and virtual stream The scaling factor. Therefore, the flow into virtual functional nodes is proportional. virtual stream Data volume multiplied by This is equivalent to the virtual flow flowing out of the virtual functional node. The amount of data, i.e. , Scenario 3, Single Input Multiple Output: If the service function chain... The number of service functions involved in this phase is 1, that is , No. The number of service function types involved in a phase is greater than 1, that is... If the virtual function node provides a service function chain for the task flow, then... Service functions involved in the phase Then virtual stream The data flowing into this virtual functional node will be processed and transformed. The same virtual stream Flow out from this node. Because SFC's... If the number of service functions involved in a phase is greater than 1, in order to achieve parallelism, the virtual streams flowing out of the virtual function nodes should be... copy In addition, services are received by the virtual functional node. virtual stream The data volume may be changed. Therefore, define... For virtual streams and virtual stream The scaling factor. Therefore, the flow into virtual functional nodes. virtual stream Data volume multiplied by This is equivalent to the virtual flow flowing out of the virtual functional node. The amount of data, i.e. ; Scenario 4, Multiple Inputs and Multiple Outputs: If the service function chain... The number of service function types involved in a phase is greater than 1, that is... , No. The number of service function types involved in a phase is greater than 1, that is... If the virtual function node provides a service function chain for the task flow, then... Service functions involved in the phase Then all service function chains have received the first... Virtual flow of service functions involved in the phase The stream flowing into this virtual function node will be processed and transformed into a virtual stream. Flowing out from this node. Multiple virtual flows flowing into this virtual function node. The services received by this virtual functional node must be proportionally merged. Therefore, the following definition is provided. For virtual streams Inflow into virtual functional nodes The ratio represents the virtual stream. proportionally Flowing into virtual function nodes. Therefore, for any two flowing into virtual function nodes... virtual stream and virtual stream The amount of data should meet the following requirements: virtual stream Data volume multiplied by equal to virtual stream Data volume multiplied by ,Right now In addition, the merged virtual stream Flowing into the virtual functional node to receive service functions The data volume may be changed. Therefore, define... For the merged virtual stream and virtual stream The scaling factor. Meanwhile, the SFC's... If the number of service functions involved in a phase is greater than 1, in order to achieve parallelism, the virtual streams flowing out of the virtual function nodes should be... copy Therefore, funds flow proportionally into virtual functional nodes. virtual stream Data volume multiplied by This is equivalent to the virtual flow flowing out of the virtual functional node. The amount of data, i.e. 6.4.3) For the source node It is a limited virtual stream Outflow source node The amount of data is equal to 0, that is:

[0064] 6.4.4) For the destination node It is a limited virtual stream Outgoing destination node The amount of data is equal to 0, that is

[0065] 6.5) Set task requirement constraints, specifically including the following two aspects: 6.5.1) During the planning period Internally, from all source node replicas Outgoing virtual stream The virtual stream will be based on the number of service functions involved in Phase 1. copy A copy, that is, a copy flowing out from all source node replicas. The same virtual stream If SFC's first The service functions involved in the phase are greater than 1, that is Therefore, before flowing into the destination node replica, all SFCs that have been received need to be processed. The virtual flows of service functions involved in the phase flow proportionally into the destination node replica. Therefore, the following is defined: For virtual streams The proportion of virtual flows flowing into the destination node replica. Therefore, all virtual flows flowing out of all source node replicas. The total amount of data is multiplied by the amount of data from phase 1 to phase 2. All scaling factors of the stages, the scaling factor of the parallel stages, and This equals all virtual flows flowing into all destination node replicas. The total amount of data, i.e.

[0066] Among them, the definition This represents the set of stages in a service function chain where the number of service functions involved is 1. According to... The previous stage The number of service functions involved in the stage can be further divided into Divided into and ,in, Indicates the first in the service function chain The number of service functions involved in the stage is greater than 1 and the first A set of stages in which the number of service functions involved in a stage is 1. Indicates the first in the service function chain The number of service functions involved in the stage is 1 and the first A set of stages in which the number of service functions involved is 1. Therefore, .definition This represents the set of stages in a service function chain where the number of service functions involved is greater than 1. Therefore, in the service function chain... Each stage can be represented as .

[0067] Reference Figure 1 The serial-parallel hybrid service function chain in the middle, , , .

[0068] 6.5.2) If the service function chain is... The number of service functions involved in a phase is greater than 1, that is To achieve parallelism, virtual streams with equal data volumes must be used. The services flow into the corresponding virtual functional nodes to receive functions. Therefore, during the planning period... Internally, it flows into all support. Virtual stream of virtual function node replicas for service functions The amount of data is equal, that is in, This indicates that it can support the deployment of service functions. A collection of virtual functional nodes.

[0069] Step 7: Adjust the data volume of all virtual flows to satisfy all constraints in Step 6, achieving the optimization objective of maximizing network flow within the planning period. Based on the data volume transmitted on the communication link and virtual transport link, set the following optimization objective function: .

[0070] The effectiveness of the embodiments of the present invention can be further illustrated by the following simulations.

[0071] 1. Simulation experimental conditions: The hardware platform for the simulation experiment in this embodiment of the invention is: an Intel Pentium CPU G3260 processor with a clock speed of 3.30 GHz and 4 GHz memory.

[0072] The software platform for the simulation experiment in this embodiment of the invention is: Windows 10 operating system, satellite simulation software (STK), and MATLAB R2021b.

[0073] The optimizer for the experimental simulation experiment in this embodiment of the invention is the CVX optimizer.

[0074] 2. Simulation Experiment Scenario: The simulation scenario of this invention, a spatial information network, consists of 18 satellites and 4 ground stations, i.e., N = 22. The 18 satellites are randomly selected from the six orbital planes of the Iridium satellite system. The four ground stations are located in Sanya, Hainan; Kashgar, Xinjiang; Beijing; and Xi'an, Shaanxi, respectively. Connectivity between the satellite-ground fusion network nodes is obtained using STK software. The planning period is 60 minutes. One satellite is randomly selected from the 18 satellites as the source node of the task flow, and the ground station in Xi'an, Shaanxi, is selected as the destination node. The remaining four satellites and four ground stations are randomly selected as functional nodes, and each functional node deploys two different service functions.

[0075] 3. Simulation content and result analysis: Figure 7 The impact of different storage capacities on the maximum network flow under different schemes was plotted. Figure 7 As can be seen, on the one hand, the maximum network flow of the method in this embodiment of the invention gradually increases with the increase of storage capacity. This is because with increased storage resources, more data can utilize the store-and-carry-forward mechanism during transmission. Data is stored on nodes awaiting a suitable opportunity to be transmitted to the next functional node to receive service functions, or a relay node relays and forwards it. Furthermore, it can be seen that when the storage capacity increases to a certain level, the maximum network flow of the method in this embodiment of the invention remains unchanged. This is because the network is limited by communication resources at this point. On the other hand, from... Figure 7 As can be seen, the maximum network flow of the snapshot sequence graph algorithm remains unchanged. This is because the snapshot sequence algorithm does not consider storage resources and only seeks paths that satisfy the hybrid serial-parallel service function chain constraints in each time slot. Therefore, the maximum network flow of the snapshot sequence graph algorithm does not change with changes in storage capacity.

[0076] Figure 8 The impact of different computational capacities on the network's maximum flow under various schemes is plotted. Figure 8 As can be seen, on the one hand, the maximum network flow gradually increases with the increase of computing capacity for different schemes. This is because when computing capacity increases, the available computing resources in the network increase. It can be observed that the maximum network flow of the proposed method is more than three times that of the snapshot sequence graph algorithm. This is because the proposed method uses a multi-functional time-spread graph to jointly represent the communication, storage, and computing resources in the network. When computing capacity increases, the amount of data that functional nodes can process also increases. However, the snapshot sequence graph algorithm does not utilize storage resources; therefore, the computing and communication resources cannot match the storage resources, resulting in poor maximum network flow performance. On the other hand, from... Figure 8As can be seen, when the computing capacity increases to a certain extent, the maximum network flow no longer increases. This is because the communication and storage resources in the method of this embodiment become the bottleneck for the maximum network flow. For the snapshot sequence graph algorithm, the communication resources in the network become the bottleneck for the maximum network flow at this point.

[0077] The units described in the embodiments of the present invention can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself.

[0078] According to one aspect of the present invention, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and executes the computer instructions, causing the computer device to perform the methods provided in the various optional implementations described above.

[0079] In another aspect, embodiments of the present invention also provide a computer-readable medium, which may be included in the electronic device described in the above embodiments; or it may exist independently and not assembled into the electronic device. The computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the methods described in the above embodiments.

Claims

1. A method for routing in a space-ground integrated network based on constraints of a series-parallel hybrid service function chain, characterized in that, The method comprises the following steps: S1, initializing network parameters and dividing network nodes; S2, initializing and representing task requests; S3, decomposing the functions of nodes in the initialized network based on the represented task requests; S4, dividing network planning periods based on the connectivity of nodes in the network; S5, constructing a multi-functional time expansion graph within the network planning period based on the function decomposition results of the nodes; S6, setting network task constraints in the multi-functional time expansion graph for the represented task requests; S7, adjusting the data volume of virtual flows to meet the constraint conditions in step S6 to ensure optimal network maximum flow performance within the planning period. 2.The method of claim 1, wherein, In step S1, the initialization of network parameters and the division of network nodes specifically include the following sub-steps: Initialize the set of satellite-ground fusion network nodes as , the number of network nodes is ; According to the network node division, the nodes in the network are divided into source nodes and destination nodes; nodes in the network that do not provide task computing functions and only serve as communication and storage are divided into non-functional nodes, and nodes in the network that can not only provide communication and storage functions but also provide computing functions are divided into functional nodes; according to this division, the network node set is divided into a source node set and a destination node set is represented as: ； wherein, and respectively represent the source node and the destination node of the task, is a set of functional nodes, is a set of non-functional nodes, . 3.The method of claim 1, wherein, In step S2, the initialization and representation of task requests specifically include the following sub-steps: S2a uses a service function chain with a pre-determined functional sequence. Characterize the task request, where, Indicates the number of tasks received in the task flow Each service function This indicates the number of service functions deployed in the network; S2b, divide the service function chain by stage, define the service function in the service function chain executed stage; the service function in the service function chain involved in the stage is expressed as , wherein the number of service functions involved in the stage is expressed as the service function involved in the initial stage is expressed as , ; S2c, virtual stream representing a virtual stream just received the first stage service function , virtual stream representing a virtual stream just flowed out of the source node and has not received any service function yet.

4. The method of claim 3, wherein, In step S3, the function decomposition of nodes in the initialized network specifically includes the following sub-steps: S3a, the functional node The set of service functions that can be supported by the deployment is represented as: ； wherein, representing a function node the total number of service functions that can be supported by the deployment; S3b, the functional node Virtual decomposition into three types of components: a virtual child node With a virtual functional node And a virtual transmission link With Wherein, Is represented as a functional node The virtual functional node of the virtual decomposition is able to provide a service function for the task ; Is represented as a virtual child node To a virtual functional node A communication link, Is represented as a virtual functional node To a virtual child node A communication link; S3c, decomposing all functional nodes to obtain a set of decomposed network nodes: The set of virtual sub-nodes decomposed from all functional nodes is represented as: ； The set of virtual functional nodes decomposed from all functional nodes is represented as: ； The set of virtual functional nodes obtained by virtual decomposition of functional nodes is represented as: ； A set of virtual function nodes capable of supporting deployment of service functions is represented as: 。 5.The method of claim 3, wherein, In step S4, the division of network planning periods based on the connectivity of nodes in the network specifically includes the following sub-steps: Satellite-ground fusion network node set Imported into the Satellite Tool Kit software, the connectivity between network nodes according to the ephemeris in the software, and according to the connectivity of the network nodes, the network planning period Divided into Continuous time slots, namely , wherein , each Indicates the Time interval. 6.The method of claim 5, wherein, In step S5, the construction of a multi-functional time expansion graph within the network planning period specifically includes the following sub-steps: S5a, initialize a blank layer directed graph, where the time interval of the layer directed graph is ; S5b, adding source nodes of tasks in the network in each time interval of the directed graph respectively , destination nodes , virtual sub-nodes of all function nodes , virtual sub-function nodes of all function nodes , obtaining four sets of nodes to constitute a multi-functional time-expanded node graph; S5c, adding transmission links, storage links, and virtual transmission links in the multi-functional time expansion node graph to obtain a multi-functional time expansion graph. 7.The method of claim 1, wherein, In step S6, the network task constraints include communication capacity constraints, storage capacity constraints, computing capacity constraints, serial-parallel service function chain constraints, and flow conservation constraints, task demand constraints. 8.The method of claim 7, wherein, In step S6, when setting the communication capacity constraints, the network task constraints specifically include the following sub-steps: Limit the total data volume of all virtual flows transmitted on the transmission link or virtual transmission link to not exceed the communication capacity of the transmission link or virtual transmission link. 9.The method of claim 7, wherein, In step S6, when setting the storage capacity constraints, the network task constraints specifically include the following sub-steps: Limit the total data volume of all virtual flows stored on the storage link to not exceed the storage capacity of the storage link. 10.The method of claim 7, wherein, In step S6, when setting the computing capacity constraints, the network task constraints specifically include the following sub-steps: Limited to time interval Internal network functional nodes Decomposition Virtual functional nodes The computing capacity consumed does not exceed that of network functional nodes. The computing power provided. 11.The method of claim 7, wherein, In step S6, when setting the serial-parallel service function chain constraints and the flow conservation constraints, the network task constraints specifically include the following sub-steps: For a virtual child node , the virtual flow The amount of data flowing into a virtual child node is equal to the amount of data flowing out of the virtual child node; For virtual functional nodes According to the first step of the serial-parallel service function chain and the Number of service functions involved in the stage and There are four possible scenarios: Scenario 1, Single Inbound, Single Outbound: If the service function chain is... The number of service function types involved in this phase is 1, that is , No. The number of service function types involved in this phase is 1, that is If the virtual function node provides a service function chain for the task flow, then... Service functions involved in the phase Then virtual stream The stream flowing into this virtual function node will be processed and transformed into a virtual stream. Flow out from this node; in addition, define For virtual streams and virtual stream The scaling factor flows into the virtual functional nodes. virtual stream Data volume multiplied by This is equivalent to the outflow of this virtual functional node. virtual stream The amount of data; Scenario 2, Multiple Inbounds, Single Outbound: If the service function chain... The number of service function types involved in a phase is greater than 1, that is... , No. The number of service function types involved in this phase is 1, that is If the virtual function node provides a service function chain for the task flow, then... Service functions involved in the phase Then, after receiving the first service function chain... Virtual flow of service functions involved in the phase After being merged proportionally, the streams flowing into this virtual functional node will be processed and transformed into virtual flows. Flowing out from this node; multiple virtual flows flowing into this virtual function node. After being merged according to the appropriate ratio, the virtual streams flow into the virtual function nodes, and the merged virtual streams flow into these virtual function nodes to receive service functions; Define For virtual streams Inflow into virtual functional nodes The ratio represents the virtual stream. proportionally Inflow into virtual function nodes; for any two inflowing virtual function nodes virtual stream and virtual stream The amount of data should meet the following requirements: virtual stream Data volume multiplied by equal to virtual stream Data volume multiplied by ;definition For the merged virtual stream and virtual stream The scaling factor flows into the virtual functional nodes proportionally. virtual stream Data volume multiplied by This is equivalent to the virtual flow flowing out of the virtual functional node. The amount of data; Scenario 3, Single Input Multiple Output: If the service function chain... The number of service functions involved in this phase is 1, that is , No. The number of service function types involved in a phase is greater than 1, that is... If the virtual function node provides a service function chain for the task flow, then... Service functions involved in the phase Then virtual stream The data flowing into this virtual functional node will be processed and transformed. The same virtual stream The stream flowing out from this node; the virtual stream flowing out from the virtual function node. copy Share; definition For virtual streams and virtual stream The scaling factor flows into the virtual functional nodes. virtual stream Data volume multiplied by This is equivalent to the virtual flow flowing out of the virtual functional node. The amount of data; Scenario 4, Multiple Inputs and Multiple Outputs: If the service function chain... The number of service function types involved in a phase is greater than 1, that is... , No. The number of service function types involved in a phase is greater than 1, that is... If the virtual function node provides a service function chain for the task flow, then... Service functions involved in the phase Then all service function chains have received the first... Virtual flow of service functions involved in the phase The stream flowing into this virtual function node will be processed and transformed into a virtual stream. Flowing out from this node; multiple virtual flows flowing into this virtual function node. After being merged proportionally, the service functions flow into this virtual functional node; definition For virtual streams Inflow into virtual functional nodes The ratio represents the virtual stream. proportionally Inflow into virtual function nodes; for any two inflowing virtual function nodes virtual stream and virtual stream The amount of data satisfies: virtual stream Data volume multiplied by equal to virtual stream Data volume multiplied by ;definition For the merged virtual stream and virtual stream The scaling factor; at the same time, the SFC's first If the number of service functions involved in a stage is greater than 1, the virtual stream flowing out from the virtual function node will be... copy share; Proportional flow into virtual function nodes virtual stream Data volume multiplied by This is equivalent to the virtual flow flowing out of the virtual functional node. The amount of data; For the source node , the virtual flow The amount of data flowing out of the source node equals 0; For the destination node , the virtual flow The amount of data flowing out of the destination node equals 0. 12.The method of claim 7, wherein, In step S6, when setting the task demand constraints, the network task constraints specifically include the following sub-steps: During the planning period Internally, from all source node replicas Outgoing virtual stream The virtual stream will be based on the number of service functions involved in Phase 1. copy A copy, that is, a copy flowing out from all source node replicas. The same virtual stream If SFC's first The service functions involved in the phase are greater than 1, that is Then, before flowing into the destination node replica, all SFCs that have been received will be... The virtual flow of the service functions involved in the phase flows into the destination node replica proportionally; Definition To virtual stream The proportion of the virtual stream flowing into the destination node copy from all source node copies The total data amount of all virtual streams flowing into the destination node copy from all source node copies at the same time multiplied by all proportion factors from the first stage to the The proportion of the parallel stage and The total data amount of all virtual streams flowing into all destination node copies is equal to The total data amount of all virtual streams flowing into all destination node copies If the service function chain is... The number of service functions involved in a phase is greater than 1, that is During the planning period Internally, it flows into all support. Virtual stream of virtual function node replicas for service functions The amount of data is equal. 13.A satellite-terrestrial converged network routing system based on constraints of a serial-parallel hybrid service function chain, characterized in that, A computer device comprising a processor and a memory, the memory storing a computer program which, when loaded by the processor, executes the method of any one of claims 1-12.

Images