Source load storage communication virtual network mapping method and device

Abstract

The invention provides a source-load-storage communication virtual network mapping method and device, and belongs to the field of communication. The method comprises the following steps: constructing a virtual network mapping model of source-load-storage communication; obtaining a region service request of a target region; wherein the regional service request comprises a plurality of virtual nodes and a plurality of virtual links, the virtual nodes comprise a starting node and a target node, and the virtual links are links between the corresponding starting node and the target node; according to the virtual network mapping model and the regional service request, establishing a target function for mapping the virtual nodes to the physical nodes by taking the minimum sum of propagation delays of all the physical nodes as a target; solving the objective function to obtain an optimal node mapping scheme; and based on the virtual network mapping model and the optimal node mapping scheme, mapping the plurality of virtual links to obtain a virtual network mapping scheme. The communication efficiency can be improved, and safe and economical operation of a power system is guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a source-load-storage communication virtual network mapping method and device. Background Art

[0002] With the construction of a new power system based on new energy, large-scale distributed source, load and storage resources are connected to the power grid for coordinated operation. Distributed source, load and storage resources are characterized by large quantity and wide dispersion, which will lead to fluctuations and instability in the operation of the power grid. Through the coordinated interactive communication network of distributed resources, the needs of efficient aggregation and unified coordination and optimization can be met, which can support the safe and stable operation of the new power system.

[0003] There is a distributed resource communication network architecture in the related technology, which can realize the automatic management of real-time energy by performing virtual network mapping. However, with the access of massive distributed resources, information interaction is not smooth, and dispatching control is more complicated. The effectiveness of the existing virtual network mapping method in the mapping process is not ideal. It is difficult to achieve business mapping while taking into account communication efficiency, which affects the safety and economic operation of the power system. Summary of the invention

[0004] The embodiment of the present invention provides a source-load-storage communication virtual network mapping method and device to improve communication efficiency and ensure the safe and economical operation of the power system.

[0005] In a first aspect, an embodiment of the present invention provides a source-load-storage communication virtual network mapping method, comprising:

[0006] Construct a virtual network mapping model for source-load-storage communication;

[0007] Obtaining a regional service request of a target area; wherein the regional service request includes a plurality of virtual nodes and a plurality of virtual links, the virtual nodes include a start node and a target node, and the virtual link is a link between the corresponding start node and the target node;

[0008] According to the virtual network mapping model and the regional service request, an objective function for mapping virtual nodes to physical nodes is established with the goal of minimizing the sum of propagation delays of all physical nodes;

[0009] Solving the objective function to obtain an optimal node mapping solution;

[0010] Based on the virtual network mapping model and the optimal node mapping solution, the multiple virtual links are mapped to obtain a virtual network mapping solution.

[0011] In a possible implementation, the algorithm for solving the objective function is an improved grey wolf algorithm;

[0012] Solving the objective function to obtain an optimal node mapping scheme includes:

[0013] Setting relevant parameters of the improved grey wolf algorithm; wherein, the relevant parameters include population size and maximum number of iterations;

[0014] Randomly generating grey wolf individuals of the population size; wherein, the first position of each grey wolf individual is used as a node mapping scheme;

[0015] According to the objective function and the first position of each grey wolf individual, calculating the first fitness of each grey wolf individual; determining the reference grey wolf individuals with the best first fitness, medium first fitness, and worst first fitness in the current population; updating the position of each grey wolf individual according to the first position, first speed of each grey wolf individual, and the first position of the reference grey wolf individuals to obtain the first population; performing mutation and crossover operations on the grey wolf individuals in the first population to obtain the second population; selecting grey wolf individuals of the population size from the first population and the second population to obtain a new population, completing one iteration;

[0016] Continuously iterating until the maximum number of iterations is reached, determining the grey wolf individual with the best fitness in the current population, and taking the first position of the grey wolf individual with the best fitness as the optimal node mapping scheme of the virtual node.

[0017] In a possible implementation manner, the updating the position of each grey wolf individual according to the first position, first speed of each grey wolf individual, and the position of the reference grey wolf individuals to obtain the first population includes:

[0018] Calculating the second position and second speed of each grey wolf individual according to the first position, first speed of each grey wolf individual, and the position of the reference grey wolf individuals;

[0019] Calculating the second fitness of each grey wolf individual according to the objective function and the second position of each grey wolf individual;

[0020] For each grey wolf individual, if the second fitness of the grey wolf individual is less than its first fitness, then taking the second position of the grey wolf individual as the new first position of the grey wolf individual;

[0021] Based on the current first positions of all grey wolf individuals, obtaining the first population.

[0022] In a possible implementation manner, the calculating the second position and second speed of each grey wolf individual according to the first position, first speed of each grey wolf individual, and the position of the reference grey wolf individuals includes:

[0023] According to Calculating the second position and second speed of each grey wolf individual;

[0024] Wherein, X i ′ represents the second position of the i-th gray wolf individual, and X i represents the first position of the i-th gray wolf individual, V′ represents the second velocity of the i-th gray wolf individual, and V i represents the first velocity of the i-th gray wolf individual, represents the addition operation of the binary operator, represents the subtraction operation of the binary operator, ω represents the weight coefficient, and X α represents the first position of the reference gray wolf individual with the best first fitness, and X β represents the first position of the reference gray wolf individual with medium first fitness, and X γ represents the first position of the reference gray wolf individual with the worst first fitness, C1, C2, and C3 represent the acceleration coefficients, r1 represents the distance between the i-th gray wolf individual and the reference gray wolf individual with the best first fitness, r2 represents the distance between the i-th gray wolf individual and the reference gray wolf individual with medium first fitness, and r3 represents the distance between the i-th gray wolf individual and the reference gray wolf individual with the worst first fitness.

[0025] In a possible implementation, before obtaining the regional service request of the target area, it further includes:

[0026] Obtain the regional service requests of each area and the waiting time of each regional service request;

[0027] Determine the request priorities of each area according to the regional service requests of each area and the corresponding waiting times;

[0028] Take the area with the highest request priority as the target area.

[0029] In a possible implementation, the determining the request priorities of each area according to the regional service requests of each area and the corresponding waiting times includes:

[0030] According to the expression: Calculate the request priorities of each area;

[0031] Wherein, A j (G V ) represents the request priority of the j-th area, R j (G V ) represents the resources required for the regional service request of the j-th area, T cj represents the waiting time of the regional service request of the j-th area, η j is a binary value, q m represents the number of service requests in the regional service request, and Q MRepresents the quantity threshold of service requests.

[0032] In a possible implementation manner, mapping the multiple virtual links based on the virtual network mapping model and the optimal node mapping scheme to obtain a virtual network mapping scheme includes:

[0033] Determine the set of candidate physical paths corresponding to each virtual link according to the physical nodes corresponding to the start node and the target node in the virtual network mapping model and the optimal node mapping scheme;

[0034] For each virtual link, retain the candidate physical paths that meet the bandwidth constraint in the set of candidate physical paths corresponding to this virtual link; calculate the path priority of each retained candidate physical path according to the bandwidth and delay of the candidate physical path; use the candidate physical path with the highest path priority as the link mapping scheme of this virtual link;

[0035] Based on all the link mapping schemes and the optimal node mapping scheme, obtain a virtual network mapping scheme.

[0036] In a possible implementation manner, after retaining the candidate physical paths that meet the bandwidth constraint in the set of candidate physical paths corresponding to this virtual link, it further includes:

[0037] If the number of retained candidate physical paths in the set of candidate physical paths corresponding to this virtual link is 0, remove the node mappings of the start node and the target node corresponding to this virtual link from the optimal node mapping scheme.

[0038] In a possible implementation manner, calculating the path priority of each retained candidate physical path according to the bandwidth and delay of the candidate physical path includes:

[0039] According to the expression: Calculate the path priority of each retained candidate physical path;

[0040] In the formula, PathPri(p) represents the path priority of candidate physical path p, b(p) represents the bandwidth of candidate physical path p, delay(p) represents the delay of candidate physical path p, Paths(l s ) represents the set of candidate physical paths corresponding to virtual link l s γ represents a weight factor, p represents a candidate physical path, and l s represents a virtual link.

[0041] In a second aspect, an embodiment of the present invention provides a source-load-storage communication virtual network mapping device, including:

[0042] A construction module, configured to construct a virtual network mapping model for source-load-storage communication;

[0043] An acquisition module, configured to acquire a regional service request for a target area; wherein, the regional service request includes a plurality of virtual nodes and a plurality of virtual links, the virtual nodes include a starting node and a target node, and the virtual link is a link between the corresponding starting node and the target node;

[0044] A node mapping module, configured to establish an objective function for mapping virtual nodes to physical nodes with the goal of minimizing the sum of propagation delays of all physical nodes according to the virtual network mapping model and the regional service request;

[0045] A solving module, configured to solve the objective function to obtain an optimal node mapping scheme;

[0046] A link mapping module, configured to map the plurality of virtual links based on the virtual network mapping model and the optimal node mapping scheme to obtain a virtual network mapping scheme.

[0047] The beneficial effects of the embodiments of the present invention compared with the prior art are as follows:

[0048] In the embodiments of the present invention, through the constructed virtual network mapping model of source-load-storage communication and the regional service request of the target area, with the goal of minimizing the sum of propagation delays of physical nodes, an objective function for mapping virtual nodes to physical nodes is established, so that the propagation delay of physical nodes can be considered during the virtual node mapping process, the average delay is reduced, the delay requirements of source-load-storage services are guaranteed, and the communication efficiency is improved; by solving the objective function, the optimal physical nodes for virtual node mapping can be determined, and the optimal node mapping scheme can be obtained quickly and accurately; then, through the virtual network mapping model and the above node mapping scheme, multiple virtual links are mapped, and the mapping of virtual links can be realized, so that the virtual network mapping scheme can be accurately obtained, and the mapping of service requests can be realized. This application can effectively and accurately realize the mapping of service requests, while taking into account the communication efficiency, and ensure the safe and economic operation of the power system. Description of the Drawings

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0050] Figure 1 It is a schematic diagram of the source-load-storage communication networking system architecture provided by the embodiments of the present invention;

[0051] Figure 2It is the implementation flowchart of the source-load-storage communication virtual network mapping method provided by the embodiments of the present invention;

[0052] Figure 3 It is the structural schematic diagram of the virtual network mapping model provided by the embodiments of the present invention;

[0053] Figure 4 It is the implementation flowchart of the improved grey wolf algorithm provided by the embodiments of the present invention;

[0054] Figure 5 It is the implementation flowchart of the virtual link mapping provided by the embodiments of the present invention;

[0055] Figure 6 It is the schematic diagram of the hierarchical multi-region resource management architecture provided by the embodiments of the present invention;

[0056] Figure 7 It is the structural schematic diagram of the source-load-storage communication virtual network mapping device provided by the embodiments of the present invention. Detailed implementation manners

[0057] In the following description, specific details such as specific system structures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present invention. However, those skilled in the art should clearly understand that the present invention can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present invention.

[0058] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will be described through specific embodiments with reference to the accompanying drawings.

[0059] Figure 1 It is the schematic diagram of the source-load-storage communication networking system architecture provided by the embodiments of the present invention. As Figure 1 shown, this architecture combines cloud-edge-end collaborative technology to achieve hierarchical coordinated control of the source-load-storage networking architecture, integrates original services such as demand response and resource regulation, and constructs a new communication networking system using public network communication technologies represented by 4G / 5G and local communication networks of various communication technologies such as EPON optical communication, High-Speed Power Line Communication (HPLC), and High Radio Frequency (HRF). This architecture helps to achieve hierarchical and zonal balance of "source-load-storage", ensure the safe and economic operation of the power system, and improve the consumption of intermittent renewable energy power generation.

[0060] The cloud layer consists of a power dispatching control center, a regional centralized control system, and a power trading market. The power dispatching control center is responsible for monitoring and optimizing the operation of the power grid, and formulating a load dispatching plan according to the actual situation of the power grid. The regional centralized control system allows multiple regions to access and achieve regional interaction, meeting the needs of overall management of differentiated services. The power trading market is a place for the exchange of power commodities, which allows market participants such as power generation companies, suppliers, and consumers to buy and sell electric energy. It can realize the optimal allocation of power resources and promote the clean and low-carbon transformation of energy.

[0061] The edge layer consists of access terminals, gateways, edge servers, centralized control devices, and local communication networks. The edge server provides edge real-time analysis and processing for the power system. The centralized control device aggregates distributed source-load-storage resources at the edge layer and enables them to participate in power grid dispatching control. The local communication network can transmit the information collected by the centralized control device to remote transmission devices.

[0062] The end layer includes multiple regions, covering terminal devices such as V2G charging piles, new energy vehicles, large power plants, photovoltaics, wind power, small wind turbines, as well as related devices such as community distribution network systems and regional energy centers. The V2G charging pile and the energy storage of new energy vehicles form an "integrated load and energy storage" energy storage method, which is both a "flexible controllable load" and an energy storage system at the same time, and can fully absorb the renewable energy of (wind) light in the community and the valley electricity of the "large power grid". The controllable renewable energy in the community makes full use of the roofs and open spaces in the community, and adopts a renewable energy power generation method mainly based on photovoltaics and supplemented by wind power, providing a safe and stable power guarantee for the "green and low-carbon" operation of community services. At the same time, it provides an energy basis for power exchange for participating in the peak shaving and auxiliary services of the large power grid.

[0063] Due to different regional geographical locations, different communication transmission methods are adopted.

[0064] Distributed source-load-storage resources in residential and industrial areas in remote areas are connected to the power grid dispatching center through optical fiber communication, as Figure 1 shown. Among them, the devices at the end layer are directly connected to cloud servers such as the power grid dispatching control center, the regional centralized control system, and the power trading market through a remote communication network composed of an EPON optical communication network and 4G / 5G to complete peak shaving and frequency modulation instructions.

[0065] Business terminals in the city center have the characteristic of a large number. If directly connected to the power grid dispatching control center, it will lead to low efficiency of multi-service interaction and instability of the power grid. Therefore, a local communication network between power terminals and centralized control devices and a remote communication network between centralized control devices and the cloud layer can be constructed. Among them, a single-network dual-mode communication method of HPLC+HRF can be adopted between the end layer and the centralized control device to give play to the high stability of HPLC and the high-speed transmission advantage of HRF to complete efficient local communication networking.

[0066] This embodiment performs virtual network mapping for the above architecture. Figure 2 The following is the implementation flowchart of the source-load-storage communication virtual network mapping method provided by the embodiment of the present invention:

[0067] Step 201: Construct a virtual network mapping model for source-load-storage communication.

[0068] In this embodiment, the virtual network mapping model can include three layers, namely the infrastructure layer, the virtual network provision layer, and the service request layer. As Figure 3 shown, the service request layer can be divided into multiple regions, and each region makes service requests; the network provision layer is virtual resources, including multiple virtual base stations; the infrastructure layer is physical nodes and physical paths, which are used to provide resources for the network provision layer.

[0069] Among them, the underlying physical network can be represented by a weighted undirected graph G s =(N s , E s ). N s represents the set of physical nodes, and E s represents the set of physical paths. G v =(N v , E v ) can be used to represent the virtual network with an undirected graph. N v represents the set of virtual nodes, and E v represents the set of virtual links. The computing power requirement of a virtual node can be expressed as cpu(n v ), and the bandwidth capacity requirement of a virtual link can be expressed as B(l v ).

[0070] In the above mapping model, the mapping of nodes and links needs to meet the following constraint conditions: (1) Node resource constraint condition: cpu(n v ) ≤ cpu(n s ); Link resource constraint condition: Among them, cpu(n v ) represents the computing power requirement of virtual node n v , cpu(n s ) represents the computing power requirement that physical node n s can meet, represents the bandwidth capacity requirement of the virtual link formed by virtual node to virtual node , represents the bandwidth capacity requirement that the physical path formed by physical node to physical node can meet.

[0071] Step 202: Obtain the regional service request of the target area; wherein, the regional service request includes multiple virtual nodes and multiple virtual links, the virtual nodes include a start node and a target node, and the virtual link is the link between the corresponding start node and target node.

[0072] In this embodiment, the service request layer of the virtual network mapping model includes multiple regions. Taking one of the regions as the target area, obtain the regional service request of this target area, and map this regional service request.

[0073] Here, the regional service request includes multiple service requests of this region. Each service request corresponds to a start node, a target node, and the virtual link between this start node and this target node. Among them, the start node and the target node are virtual nodes.

[0074] Virtual network mapping needs to map virtual nodes and virtual links.

[0075] Step 203: According to the virtual network mapping model and the regional service request, with the goal of minimizing the sum of the propagation delays of all physical nodes, establish an objective function for mapping virtual nodes to physical nodes.

[0076] In this embodiment, considering ensuring the delay requirements of different services and ensuring communication efficiency, establish an objective function according to the power service delay sensitivity, that is, establish an objective function for the virtual node mapping stage based on the propagation delay of physical nodes. Since the regional service request includes multiple service requests, therefore, the objective function can be determined with the goal of minimizing the sum of the propagation delays of the physical nodes.

[0077] Here, the propagation delay of a physical node can be composed of the processing delay of this physical node and the sum of the average delays of all adjacent links connected to this physical node. If the propagation delay of a physical node is smaller, then the ability of this physical node to meet the virtual node delay requirement is stronger.

[0078] Optionally, the calculation formula for the propagation delay of a physical node can be:

[0079]

[0080] In the formula, represents the propagation delay of physical node , represents the processing delay of physical node , represents physical node 's adjacent link e p 's delay, N represents the total number of adjacent links of physical node , represents physical node The set of adjacent links, represents the physical node The sum of the average delays of all adjacent links connected.

[0081] Step 204, solve the objective function to obtain the optimal node mapping scheme.

[0082] In this embodiment, by solving the above objective function, an optimal mapping scheme for each virtual node to be mapped to a physical node can be obtained, thus completing the mapping of virtual nodes.

[0083] Step 205, based on the virtual network mapping model and the optimal node mapping scheme, map multiple virtual links to obtain a virtual network mapping scheme.

[0084] In this embodiment, after completing the mapping of virtual nodes to physical nodes, the virtual links are also mapped. Since a virtual link is a link between a start node and a target node, therefore, the mapping of the virtual link is to map the virtual link to the physical path between the physical node corresponding to the start node and the physical node corresponding to the target node. By mapping multiple virtual links in the regional service request, the mapping of the virtual link is completed, thereby obtaining the final virtual network mapping scheme and realizing the virtual network mapping.

[0085] In the embodiment of the present invention, through the constructed virtual network mapping model for source-load-storage communication and the regional service request of the target area, with the goal of minimizing the sum of the propagation delays of physical nodes, an objective function for mapping virtual nodes to physical nodes is established, which can consider the propagation delays of physical nodes during the virtual node mapping process, reduce the average delay, ensure the delay requirements of source-load-storage services, and improve communication efficiency; by solving the objective function, the optimal physical nodes for virtual node mapping can be determined, and the optimal node mapping scheme can be obtained quickly and accurately; then, through the above virtual network mapping model and the above node mapping scheme, multiple virtual links can be mapped, and the mapping of the virtual link can be realized, so that the virtual network mapping scheme can be accurately obtained and the mapping of the service request can be realized. This application can effectively and accurately realize the mapping of service requests, while taking into account communication efficiency and ensuring the safe and economic operation of the power system.

[0086] In some embodiments, the algorithm for solving the objective function is an improved gray wolf algorithm.

[0087] As Figure 4 shown, in this embodiment, solving the objective function to obtain the optimal node mapping scheme can be:

[0088] Step 1, set the relevant parameters of the improved gray wolf algorithm; among them, the relevant parameters include the population size and the maximum number of iterations.

[0089] Step 2: Randomly generate gray wolf individuals of the population size; among them, the first position of each gray wolf individual serves as a node mapping scheme.

[0090] Here, the position of each gray wolf individual represents a possible node mapping scheme for virtual nodes. The position vector of a gray wolf individual can be expressed as X i =(x i1 ,x i2 ,…,x ip ), where the subscript p represents the number of nodes in the service requests of this region and is also the search space dimension of the gray wolf individual, and x ij takes positive values (1 < j < p), representing the candidate boundary node number selected by the j-th virtual node in its matching set.

[0091] Step 3: According to the objective function and the first position of each gray wolf individual, calculate the first fitness of each gray wolf individual; determine the reference gray wolf individuals with the best first fitness, medium first fitness, and worst first fitness in the current population; update the position of each gray wolf individual according to the first position, first velocity of each gray wolf individual, and the first position of the reference gray wolf individuals to obtain the first population; perform mutation and crossover operations on the gray wolf individuals in the first population to obtain the second population; select gray wolf individuals of the population size from the first population and the second population to obtain a new population, completing one iteration.

[0092] In this embodiment, to prevent the algorithm from falling into a locally optimal mapping scheme during iteration, the gray wolves after position update do not directly enter the next round of iteration, but instead perform mutation operations, crossover operations, and selection operations on the position vectors of the gray wolf individuals in the population and then enter the next round of iteration.

[0093] Here, taking the objective function as the fitness function, the position of the gray wolf individual can be substituted into the objective function to obtain the corresponding fitness.

[0094] Step 4: Continuously iterate until the maximum number of iterations is reached, determine the gray wolf individual with the best fitness in the current population, and take the first position of the gray wolf individual with the best fitness as the optimal node mapping scheme for virtual nodes.

[0095] In this embodiment, considering that the gray wolf optimization algorithm is mainly used to solve continuous problems, the update formulas for the position vector and velocity vector of individuals cannot be directly used in the process of solving discrete problems. Moreover, the gray wolf optimization algorithm is prone to falling into local optimality, resulting in the fact that the feasible solution finally obtained is not necessarily the optimal solution. Therefore, the basic gray wolf optimization algorithm cannot be directly applied to virtual node mapping.

[0096] Therefore, in this embodiment, the basic variables and related operations in the Grey Wolf Optimization algorithm are redefined. Mutation, crossover, and selection operations are added to the Grey Wolf Optimization algorithm to make full use of the differences between individuals to enrich the diversity of the population, so that it is not easy to fall into the local optimum during the process of solving the optimization problem.

[0097] Optionally, in this embodiment, according to the first position and the first speed of each grey wolf individual and the position of the reference grey wolf individual, the position of each grey wolf individual is updated to obtain the first population, which may be:

[0098] Step 1: According to the first position and the first speed of each grey wolf individual and the position of the reference grey wolf individual, calculate the second position and the second speed of each grey wolf individual.

[0099] In this embodiment, the speed of the grey wolf individual is used to adjust the position of the grey wolf individual. The dimension of the speed vector is the same as that of the position vector, and it can be represented by V i =(v i1 , v i2 , …, v ip ), where v ij represents the index increment of the boundary node in the jth virtual node matching set, and this increment can be positive, negative, or zero. Since the index is an integer, for the convenience of subsequent operations, when the speed vector is updated, if there is a decimal, it needs to be rounded to an integer.

[0100] In this embodiment, according to the first position and the first speed of each grey wolf individual and the position of the reference grey wolf individual, calculating the second position and the second speed of each grey wolf individual may be:

[0101] According to calculate the second position and the second speed of each grey wolf individual;

[0102] In the formula, X i ′ represents the second position of the ith grey wolf individual, X i represents the first position of the ith grey wolf individual, V′ represents the second speed of the ith grey wolf individual, V i represents the first speed of the ith grey wolf individual, represents the addition operation of the binary operator, represents the subtraction operation of the binary operator, ω represents the weight coefficient, X α represents the first position of the reference grey wolf individual with the best first fitness, X β represents the first position of the reference grey wolf individual with medium first fitness, X γRepresents the first position of the reference grey wolf individual with the worst first fitness. C1, C2, and C3 represent acceleration coefficients. r1 represents the distance between the i-th grey wolf individual and the reference grey wolf individual with the best first fitness. r2 represents the distance between the i-th grey wolf individual and the reference grey wolf individual with medium first fitness. r3 represents the distance between the i-th grey wolf individual and the reference grey wolf individual with the worst first fitness.

[0103] Subtraction operation Used to calculate the difference magnitude between two partitioning schemes. Subtraction Is a binary operator. The subtraction operation of two vectors with the same dimension means that the subtraction operation is performed on each dimension component respectively. Among them The result is represented by the index difference between the boundary node x ij And the boundary node x ij In the matching set of the j-th virtual node.

[0104] Addition operation Used to adjust the original mapping scheme to obtain a new scheme. Addition Is also a binary operator. The addition operation of vectors with the same dimension means that the corresponding components are added respectively. Among them, The result of ij Is the boundary node number pointed to by the index obtained by summing the index of the boundary node in the matching set of the j-th virtual node and the index increment ν ij Of the boundary node.

[0105] The above velocity and position update formulas make the grey wolf individuals randomly move between the original position and the local optimal position, and also make the algorithm more suitable for solving discretization problems.

[0106] In addition, in order to be applicable to the mapping of virtual nodes, the basic addition and subtraction operations of the mutation formula can be improved to the above addition operation And subtraction operation That is, randomly select three grey wolf individuals from the grey wolf population, and their positions are X a 、X b And X c The position of the mutated grey wolf individual is X d Satisfying the following formula Among them, F represents the scaling factor. The crossover operation is improved to swap some randomly selected components in the current grey wolf position vector with the corresponding components in the mutated grey wolf position vector to increase the diversity of the population. The selection operation is to select whether to accept the position of the new grey wolf individual after the crossover operation according to the fitness value.

[0107] Step 2: Calculate the second fitness of each grey wolf individual according to the objective function and the second position of each grey wolf individual.

[0108] Step 3: For each grey wolf individual, if the second fitness of the grey wolf individual is less than its first fitness, then use the second position of the grey wolf individual as the new first position of the grey wolf individual.

[0109] In this embodiment, if the second fitness of a grey wolf individual is less than its first fitness, it indicates that the second position of the grey wolf individual is superior to the previously determined first position. Therefore, the position of the grey wolf individual can be updated, and the second position is used as the new first position.

[0110] If the second fitness of a grey wolf individual is greater than or equal to its first fitness, it indicates that the second position of the grey wolf individual is not superior to the previously determined first position. Therefore, the second position of the grey wolf individual can be not accepted, no position update is performed, and the original first position is still retained.

[0111] Step 4: Obtain the first population based on the current first positions of all grey wolf individuals.

[0112] The mapping process of virtual nodes is introduced above. Next, the mapping process of virtual links will be continued.

[0113] In some embodiments, based on the virtual network mapping model and the optimal node mapping scheme, mapping multiple virtual links to obtain a virtual network mapping scheme may be as follows:

[0114] Step 1: According to the physical nodes corresponding to the start node and the target node in the virtual network mapping model and the optimal node mapping scheme, determine the set of candidate physical paths corresponding to each virtual link.

[0115] In this embodiment, after the start node and the target node are mapped, the start physical node and the final physical node of the physical path for virtual link mapping are also determined. Therefore, the optional physical paths corresponding to the virtual link can be found in the underlying physical network.

[0116] Since in the process of link mapping, the shorter the hop distance of the underlying physical path carrying the virtual link, the less network resources of the underlying layer the virtual link occupies, and the lower the mapping cost. Therefore, the path with the shortest hop distance can be found in the underlying physical network.

[0117] Considering the issues of the bandwidth and delay of the physical path, multiple candidate physical paths can be found for each virtual link, and then a suitable physical path is selected to map the virtual link.

[0118] Here, the Dijkstra algorithm can be used to find multiple physical paths with the shortest hop distance in the underlying physical network as candidate physical paths. The number of candidate physical paths specifically selected can be determined by the scale of the underlying physical network. For example, 3, 4, 5, 6, 7, etc. can be selected.

[0119] Step 2. Refer to Figure 5 the implementation flowchart of virtual link mapping shown in Figure 5. For each virtual link, retain the candidate physical paths that meet the bandwidth constraint in the set of candidate physical paths corresponding to the virtual link; calculate the path priority of each retained candidate physical path according to the bandwidth and delay of the candidate physical path; and use the candidate physical path with the highest path priority as the link mapping scheme for the virtual link.

[0120] In this embodiment, for each candidate physical path, it can be checked whether it meets the bandwidth constraint, that is if it meets the bandwidth constraint, it means that the candidate physical path can provide corresponding resources for the virtual link. If it does not meet the bandwidth constraint, it means that the candidate physical path cannot undertake the mapping of the virtual link, and the candidate physical path can be removed from the set of candidate physical paths.

[0121] For the retained candidate physical paths, priority division can be performed, and the candidate physical path with the highest priority is selected as the physical path for virtual link mapping. Here, in path selection, to achieve optimization in terms of delay, the priority can be calculated according to bandwidth and delay.

[0122] Optionally, according to the bandwidth and delay of the candidate physical path, calculating the path priority of each retained candidate physical path can be based on the expression: Calculate the path priority of each retained candidate physical path; in the formula, PathPri(p) represents the path priority of candidate physical path p, b(p) represents the bandwidth of candidate physical path p, delay(p) represents the delay of candidate physical path p, Paths(l s ) represents the set of candidate physical paths corresponding to virtual link l s γ represents the weight factor, p represents the candidate physical path, and l s represents the virtual link.

[0123] Optionally, after retaining the candidate physical paths that meet the bandwidth constraint in the set of candidate physical paths corresponding to the virtual link in this embodiment, it further includes: if the number of retained candidate physical paths in the set of candidate physical paths corresponding to the virtual link is 0, remove the node mappings of the start node and the target node corresponding to the virtual link from the optimal node mapping scheme.

[0124] Here, if the set of candidate physical paths of a virtual link is empty, it means that there is no corresponding physical path to realize the mapping of the virtual link. Then, the service request in the regional service request can be rejected, the node mapping of the start node and the target node corresponding to the virtual link can be removed from the optimal node mapping scheme, and the previously executed node mapping can be revoked.

[0125] Step 3: Based on all link mapping schemes and the optimal node mapping scheme, obtain the virtual network mapping scheme.

[0126] In this embodiment, after the virtual link mapping, the virtual link mapping scheme and the optimal node mapping scheme of the reserved virtual nodes can be combined to form the virtual network mapping scheme, so as to realize the virtual network mapping of the regional service request.

[0127] In addition, before determining the set of candidate physical paths of the virtual link, the virtual links can be sorted, for example, sorted according to the bandwidth required by the virtual link. The link mapping is preferentially performed for the virtual links with large required bandwidth, that is, the link mapping of the virtual links is sequentially performed in the order from large to small of the required bandwidth, so as to ensure that the virtual links with large required bandwidth can be successfully mapped.

[0128] The above text specifically introduces the mapping process of virtual nodes and virtual links. Before the mapping, the regional service requests can also be prioritized, and the regions with high priorities can be selected as the target regions.

[0129] As Figure 6 shown, the hierarchical multi-region resource management architecture includes the following parts:

[0130] Infrastructure provider (InP): Provides the network hardware facilities and physical network resources for supporting network virtualization technology.

[0131] Local Resource Management Center (LRMC): Responsible for managing and allocating the resources of each region.

[0132] Global Resource Management Center (GBMC): Responsible for managing the resources between regions and some resources within regions.

[0133] Service Provider (SP): Abstracts the logical network according to the user's resource request and generates the VNR. Generally, the request is sequentially sent to the normally operating GRMC, and the message of whether the mapping is successful is received from the GRMC and fed back to the user.

[0134] The hierarchical multi-region resource management architecture can not only avoid the huge additional overhead caused by the SP continuously negotiating directly with multiple InPs due to the inability to obtain global information during the mapping process, but also show certain advantages in avoiding single-center failure and scalability.

[0135] Based on the above architecture, virtual network mapping is performed in a regionalized manner.

[0136] In some embodiments, before obtaining the regional service request of the target region, the regional service requests of each region and the waiting time of each regional service request can also be obtained; then, according to the regional service requests of each region and the corresponding waiting times, the request priorities of each region are determined; the region with the highest request priority is taken as the target region.

[0137] In this embodiment, by the regional service requests and waiting times of each region, the request priorities of each region are determined, and the region with the highest request priority is preferentially taken as the target region for virtual network mapping, so as to shorten the waiting time of regional service requests and improve communication efficiency.

[0138] Optionally, determining the request priorities of each region according to the regional service requests of each region and the corresponding waiting times can be: according to the expression: Calculate the request priorities of each region; in the formula, A j (G V ) represents the request priority of the jth region, R j (G V ) represents the resources required for the regional service request of the jth region, T cj represents the waiting time of the regional service request of the jth region, η j is a binary value, q m represents the number of service requests in the regional service request, Q M represents the threshold of the number of service requests.

[0139] Among them, the resources required for the regional service request of the jth region can be calculated according to the following formula: In the formula, represents the computing resources required for the regional service request of the jth region, represents the link bandwidth resources that need to be maintained for the regional service request of the jth region.

[0140] In this embodiment, the number of service requests q m in the regional service request is less than the threshold Q M of the number of service requests, indicating that the number of regional service requests is small and the resource demand is small, and it can be preferentially processed, shortening the waiting time of virtual network requests and increasing the mapping rate of subsequent requests.

[0141] The quantity q of service requests in the regional service request m is greater than or equal to the service request quantity threshold Q M , indicating that the quantity of regional service requests is large. For this part of regional service requests, the regional service requests with less resources required for the service requests can be preferentially processed.

[0142] Here, the smaller the value of the request priority, the higher the priority level of the corresponding request priority, that is, the more preferentially processed. Among them, the request priority of the regional service request with a value of 0 can be determined as the highest level, and the rest are arranged and processed according to the size of the request priority value. The regional service requests with a request priority of 0 are always preferentially executed to ensure the fairness of service requests.

[0143] In addition, after determining the virtual network mapping scheme, the virtual network mapping scheme can also be evaluated using evaluation metrics.

[0144] The evaluation metrics can include mapping overhead, mapping revenue, revenue-overhead ratio, request acceptance rate, etc.

[0145] The mapping overhead is related to the size and resource cost of the physical resources provided by the underlying physical network. The calculation formula for the mapping overhead is as follows: In the formula, τ1(n s ) represents the unit cost of the computing resources provided by the physical node n v when the virtual node n s is embedded into the physical node n s , τ2(l s ) represents the unit cost of the bandwidth resources provided by the physical path l v when the virtual link l s is embedded into the physical path l s . P s represents the physical path selected for embedding the virtual link l v , and this physical path is a set composed of one or more physically connected physical paths. When the unit costs of both types of resources are 1, the mapping overhead represents the total size of the physical network resources when the virtual network is successfully mapped.

[0146] The mapping revenue consists of two parts: node mapping revenue and link mapping revenue. The calculation formula for the mapping revenue is as follows: In the formula, cpu(n v ) represents the computing resources required by the virtual node n v , b(l v ) represents the bandwidth resources required by the virtual link l v , μ1 represents the mapping unit price of the computing resources required by the virtual node n v , and μ2 represents the virtual link lv The mapped unit price of the required bandwidth resources. When the unit prices are all 1, the mapping revenue is an indicator measuring the total resource demand of the virtual network.

[0147] The revenue - cost ratio represents the ratio of the mapping revenue obtained within a period of time to the mapping cost, and its calculation formula can be:

[0148] The calculation formula of the request acceptance rate can be: In the formula, represents the number of virtual network requests (VNRs) accepted within the time period T, represents the total number of VNRs arriving within the time period T. The larger the value of the request acceptance rate, the better the effect of the virtual network mapping. Here, the VNR refers to the service requests in all regions.

[0149] In the embodiment of the present invention, through the constructed virtual network mapping model of source - load - storage communication and the regional service requests of the target area, with the goal of minimizing the sum of the propagation delays of physical nodes, a target function for mapping virtual nodes to physical nodes is established. The propagation delays of physical nodes can be considered during the virtual node mapping process, reducing the average delay, ensuring the delay requirements of source - load - storage services, and improving communication efficiency; by solving the target function, the optimal physical nodes for virtual node mapping can be determined, and the optimal node mapping scheme can be obtained quickly and accurately; then, through the above - mentioned virtual network mapping model and the above - mentioned node mapping scheme, multiple virtual links can be mapped, and the mapping of virtual links can be realized, so that the virtual network mapping scheme can be accurately obtained and the mapping of service requests can be realized. This application can effectively and accurately realize the mapping of service requests, while taking into account communication efficiency and ensuring the safe and economic operation of the power system.

[0150] It should be understood that the magnitudes of the sequence numbers of the steps in the above - mentioned embodiments do not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0151] The following is the device embodiment of the present invention. For the details not described in detail, reference can be made to the corresponding method embodiments above.

[0152] Figure 7 The structural schematic diagram of the source - load - storage communication virtual network mapping device provided by the embodiment of the present invention is shown. For the sake of convenience of description, only the parts related to the embodiment of the present invention are shown and are described in detail as follows:

[0153] As Figure 7 shown, the source - load - storage communication virtual network mapping device 70 includes:

[0154] A building module 71 for building a virtual network mapping model for source-load-storage communication;

[0155] An acquisition module 72 for acquiring regional service requests of a target area; wherein, the regional service requests include a plurality of virtual nodes and a plurality of virtual links, the virtual nodes include a starting node and a target node, and the virtual link is a link between the corresponding starting node and the target node;

[0156] A node mapping module 73 for establishing an objective function of mapping virtual nodes to physical nodes with the minimum sum of propagation delays of all physical nodes as the objective according to the virtual network mapping model and the regional service requests;

[0157] A solving module 74 for solving the objective function to obtain an optimal node mapping scheme;

[0158] A link mapping module 75 for mapping a plurality of virtual links based on the virtual network mapping model and the optimal node mapping scheme to obtain a virtual network mapping scheme.

[0159] In a possible implementation manner, the algorithm for solving the objective function is an improved grey wolf algorithm;

[0160] The solving module 74 is specifically configured to:

[0161] Set relevant parameters of the improved grey wolf algorithm; wherein, the relevant parameters include the population size and the maximum number of iterations;

[0162] Randomly generate grey wolf individuals with the population size; wherein, the first position of each grey wolf individual is used as a node mapping scheme;

[0163] Calculate the first fitness of each grey wolf individual according to the objective function and the first position of each grey wolf individual; determine the reference grey wolf individuals with the best first fitness, medium first fitness, and worst first fitness in the current population; update the position of each grey wolf individual according to the first position, the first speed of each grey wolf individual, and the first position of the reference grey wolf individuals to obtain the first population; perform mutation and crossover operations on the grey wolf individuals in the first population to obtain the second population; select grey wolf individuals with the population size from the first population and the second population to obtain a new population, and complete one iteration;

[0164] Continuously iterate until the maximum number of iterations is reached, determine the grey wolf individual with the best fitness in the current population, and use the first position of the grey wolf individual with the best fitness as the optimal node mapping scheme of the virtual nodes.

[0165] In a possible implementation manner, the solving module 74 is specifically configured to:

[0166] Calculate the second position and second velocity of each grey wolf individual according to the first position and first velocity of each grey wolf individual and the position of the reference grey wolf individual;

[0167] Calculate the second fitness of each grey wolf individual according to the objective function and the second position of each grey wolf individual;

[0168] For each grey wolf individual, if the second fitness of the grey wolf individual is less than its first fitness, then use the second position of the grey wolf individual as the new first position of the grey wolf individual;

[0169] Obtain the first population based on the current first positions of all grey wolf individuals.

[0170] In a possible implementation, the solving module 74 is specifically configured to:

[0171] According to Calculate the second position and second velocity of each grey wolf individual;

[0172] Where, X i ′ represents the second position of the i-th grey wolf individual, X i represents the first position of the i-th grey wolf individual, V′ represents the second velocity of the i-th grey wolf individual, V i represents the first velocity of the i-th grey wolf individual, represents the addition operation of the binary operator, represents the subtraction operation of the binary operator, ω represents the weight coefficient, X α represents the first position of the reference grey wolf individual with the best first fitness, X β represents the first position of the reference grey wolf individual with medium first fitness, X γ represents the first position of the reference grey wolf individual with the worst first fitness, C1, C2, and C3 represent acceleration coefficients, r1 represents the distance between the i-th grey wolf individual and the reference grey wolf individual with the best first fitness, r2 represents the distance between the i-th grey wolf individual and the reference grey wolf individual with medium first fitness, and r3 represents the distance between the i-th grey wolf individual and the reference grey wolf individual with the worst first fitness.

[0173] In a possible implementation, the obtaining module 72 is further configured to:

[0174] Obtain the regional service requests of each region and the waiting time of each regional service request;

[0175] Determine the request priorities of each region according to the regional service requests of each region and the corresponding waiting times;

[0176] Use the region with the highest request priority as the target region.

[0177] In a possible implementation, the obtaining module 72 is specifically configured to:

[0178] According to the expression: Calculate the request priority of each region;

[0179] In the formula, A j (G V ) represents the request priority of the jth region, R j (G V ) represents the resources required for the regional service request of the jth region, T cj represents the waiting time of the regional service request of the jth region, η j is a binary value, q m represents the number of service requests in the regional service request, Q M represents the threshold of the number of service requests.

[0180] In a possible implementation, the link mapping module 75 is specifically configured to:

[0181] According to the physical nodes corresponding to the start node and the target node in the virtual network mapping model and the optimal node mapping scheme, determine the set of candidate physical paths corresponding to each virtual link;

[0182] For each virtual link, retain the candidate physical paths that meet the bandwidth constraint in the set of candidate physical paths corresponding to the virtual link; calculate the path priority of each retained candidate physical path according to the bandwidth and delay of the candidate physical path; use the candidate physical path with the highest path priority as the link mapping scheme of the virtual link;

[0183] Based on all the link mapping schemes and the optimal node mapping scheme, obtain the virtual network mapping scheme.

[0184] In a possible implementation, the link mapping module 75 is further configured to:

[0185] If the number of retained candidate physical paths in the set of candidate physical paths corresponding to the virtual link is 0, remove the node mappings of the start node and the target node corresponding to the virtual link from the optimal node mapping scheme.

[0186] In a possible implementation, the link mapping module 75 is specifically configured to:

[0187] According to the expression: Calculate the path priority of each retained candidate physical path;

[0188] Wherein, PathPri(p) represents the path priority of the candidate physical path p, b(p) represents the bandwidth of the candidate physical path p, delay(p) represents the delay of the candidate physical path p, and Paths(l s ) represents the set of candidate physical paths corresponding to the virtual link l s , γ represents the weight factor, p represents the candidate physical path, and l s represents the virtual link.

[0189] In the above embodiments, the descriptions of the various embodiments have their own emphases. For parts not detailed or recorded in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0190] Those of ordinary skill in the art can realize that the templates, units, and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professionals can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0191] If the module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes of the above method embodiments of the present invention, it can also be completed by a computer program instructing the relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0192] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A source-load-storage communication virtual network mapping method, characterized in that: include: Construct a virtual network mapping model for source-load-storage communication; Obtaining a regional service request of a target area; wherein the regional service request includes a plurality of virtual nodes and a plurality of virtual links, the virtual nodes include a start node and a target node, and the virtual link is a link between the corresponding start node and the target node; According to the virtual network mapping model and the regional service request, an objective function for mapping virtual nodes to physical nodes is established with the goal of minimizing the sum of propagation delays of all physical nodes; Solving the objective function to obtain an optimal node mapping solution; Based on the virtual network mapping model and the optimal node mapping solution, the multiple virtual links are mapped to obtain a virtual network mapping solution.

2. The source-load-storage communication virtual network mapping method according to claim 1, characterized in that: The algorithm for solving the objective function is an improved grey wolf algorithm; The solving the objective function to obtain an optimal node mapping solution includes: Setting relevant parameters of the improved grey wolf algorithm; wherein the relevant parameters include population size and maximum number of iterations; Randomly generate gray wolf individuals of the population size; wherein the first position of each gray wolf individual is used as a node mapping scheme; According to the objective function and the first position of each gray wolf individual, the first fitness of each gray wolf individual is calculated; the reference gray wolf individuals with the best first fitness, the medium first fitness and the worst first fitness in the current population are determined; according to the first position and the first speed of each gray wolf individual and the first position of the reference gray wolf individual, the position of each gray wolf individual is updated to obtain a first population; the gray wolf individuals in the first population are mutated and crossovered to obtain a second population; gray wolf individuals of the population size are selected from the first population and the second population to obtain a new population, and one iteration is completed; Iterate continuously until the maximum number of iterations is reached, determine the gray wolf individual with the best fitness in the current population, and use the first position of the gray wolf individual with the best fitness as the optimal node mapping solution for the virtual node.

3. The source-load-storage communication virtual network mapping method according to claim 2, characterized in that: The method of updating the position of each individual gray wolf according to the first position and the first speed of each individual gray wolf and the position of the reference individual gray wolf to obtain a first population includes: Calculate the second position and the second speed of each individual gray wolf according to the first position and the first speed of each individual gray wolf and the position of the reference individual gray wolf; Calculating the second fitness of each gray wolf individual according to the objective function and the second position of each gray wolf individual; For each gray wolf individual, if the second fitness of the gray wolf individual is less than its first fitness, the second position of the gray wolf individual is used as the new first position of the gray wolf individual; Based on the current first positions of all gray wolf individuals, the first population is obtained.

4. The source-load-storage communication virtual network mapping method according to claim 3, characterized in that: The step of calculating the second position and the second speed of each individual gray wolf according to the first position and the first speed of each individual gray wolf and the position of the reference individual gray wolf comprises: according to Calculate the second position and second speed of each individual gray wolf; Where, X i ′ represents the second position of the i-th gray wolf individual, X i represents the first position of the i-th gray wolf individual, V′ represents the second speed of the i-th gray wolf individual, V i represents the first speed of the i-th gray wolf individual, Represents the addition operation of a binary operator, represents the subtraction operation of the binary operator, ω represents the weight coefficient, X α represents the first position of the reference gray wolf individual with the best first fitness, X β represents the first position of the reference gray wolf individual with medium fitness, X γ represents the first position of the reference gray wolf individual with the worst first fitness, C1, C2 and C3 represent acceleration coefficients, r1 represents the distance between the ith gray wolf individual and the reference gray wolf individual with the best first fitness, r2 represents the distance between the ith gray wolf individual and the reference gray wolf individual with medium first fitness, and r3 represents the distance between the ith gray wolf individual and the reference gray wolf individual with the worst first fitness.

5. The source-load-storage communication virtual network mapping method according to any one of claims 1 to 4, characterized in that: Before obtaining the area service request of the target area, the method further includes: Get the regional business requests of each region and the waiting time of each regional business request; Determine the request priority of each region based on the regional business requests and corresponding waiting time of each region; The region with the highest request priority is used as the target region.

6. The source-load-storage communication virtual network mapping method according to claim 5, characterized in that: Determining the request priority of each region according to the regional service requests of each region and the corresponding waiting time includes: According to the expression: Calculate the request priority for each region; In the formula, A j (G V ) represents the request priority of the jth region, R j (G V ) represents the resources required by the regional service request of the jth region, T cj represents the waiting time of the regional service request of the jth region, η j is a binary value, q m Indicates the number of service requests in the regional service request, Q M Indicates the threshold of the number of service requests.

7. The source-load-storage communication virtual network mapping method according to any one of claims 1 to 4, characterized in that: The mapping of the multiple virtual links based on the virtual network mapping model and the optimal node mapping scheme to obtain a virtual network mapping scheme includes: Determine a set of candidate physical paths corresponding to each virtual link according to the virtual network mapping model and the physical nodes corresponding to the start node and the target node in the optimal node mapping solution; For each virtual link, retain the candidate physical paths that meet the bandwidth constraint in the candidate physical path set corresponding to the virtual link; calculate the path priority of each retained candidate physical path according to the bandwidth and delay of the candidate physical path; and use the candidate physical path with the highest path priority as the link mapping scheme of the virtual link; Based on all link mapping schemes and the optimal node mapping scheme, a virtual network mapping scheme is obtained.

8. The source-load-storage communication virtual network mapping method according to claim 7, characterized in that: After retaining the candidate physical paths that meet the bandwidth constraint in the candidate physical path set corresponding to the virtual link, the method further includes: If the number of candidate physical paths retained in the candidate physical path set corresponding to the virtual link is 0, the node mappings of the start node and the target node corresponding to the virtual link are removed from the optimal node mapping solution.

9. The source-load-storage communication virtual network mapping method according to claim 7, characterized in that: The step of calculating the path priority of each retained candidate physical path according to the bandwidth and delay of the candidate physical path includes: According to the expression: Calculate the path priority of each candidate physical path retained; Where PathPri(p) represents the path priority of candidate physical path p, b(p) represents the bandwidth of candidate physical path p, delay(p) represents the delay of candidate physical path p, and Paths(l s ) indicates virtual link l s The corresponding candidate physical path set, γ represents the weight factor, p represents the candidate physical path, l s Indicates a virtual link.

10. A source-load-storage communication virtual network mapping device, characterized in that: include: A construction module for constructing a virtual network mapping model for source-load-storage communication; An acquisition module, used to acquire a regional service request of a target area; wherein the regional service request includes a plurality of virtual nodes and a plurality of virtual links, the virtual nodes include a start node and a target node, and the virtual link is a link between the corresponding start node and the target node; A node mapping module, used to establish an objective function for mapping virtual nodes to physical nodes based on the virtual network mapping model and the regional service request, with the goal of minimizing the sum of propagation delays of all physical nodes; A solution module, used for solving the objective function to obtain an optimal node mapping solution; The link mapping module is used to map the multiple virtual links based on the virtual network mapping model and the optimal node mapping solution to obtain a virtual network mapping solution.

Images