A virtual network mapping method based on node delay perception

Abstract

The present invention relates to a virtual network mapping method based on node delay perception, including: obtaining virtual node sorting values according to a node sorting algorithm, and sorting virtual nodes based on the virtual node sorting values; introducing the physical node delay importance into the node sorting algorithm, weighting the deployment delay of physical nodes, node topology attributes, and global network resources, and then sorting physical nodes through a physical node sorting algorithm based on delay perception to obtain physical node sorting values; performing virtual node mapping according to the virtual node sorting values and physical node sorting values; and using the K - shortest path algorithm to find K shortest physical paths between the two physical nodes according to the physical nodes mapped by the end nodes of the virtual link. The present invention reduces the deployment delay of the virtual network and improves the virtual network mapping revenue - expenditure ratio on the premise of ensuring the mapping success rate of the virtual network, achieving the purpose of balancing the performance and delay of the virtual network mapping algorithm.

Description

Technical Field

[0001] The present invention relates to the technical field of network function virtualization, and in particular to a virtual network mapping method based on node delay perception. Background Art

[0002] With the continuous expansion of the Internet scale, the rapid growth of the number of users, and the emergence of various emerging applications, the traditional Internet has gradually been unable to meet and adapt to the changes in current network service requirements, and has become increasingly bloated and rigid. In recent years, Network Function Virtualization (NFV) has attracted the attention and research of many scholars in the academic community as an effective method to solve the problem of network rigidity.

[0003] The core idea of network function virtualization is to abstract and encapsulate existing network resources, virtualize and flatten the physical resources of the underlying physical network, so as to realize the decoupling of the basic capabilities of the network itself and the specific functions of network services. End users can customize various virtual network services according to their actual needs to share the physical resources of the underlying physical network, and they do not interfere with each other and are independent of each other. Due to the high inclusiveness and compatibility of network virtualization, it is conducive to the evolution of the current communication network towards the future network. Therefore, network virtualization technology is regarded as a key technology for building the future network.

[0004] Existing virtual network mapping methods optimize the consumption of mapping to maximize the mapping success rate of virtual network service providers, but they do not consider the deployment delay of mapping and the deployment delay of underlying physical nodes. The current 5G mobile communication system has high requirements for the delay of communication services, and the end-to-end delay in low-delay scenarios even needs to reach the millisecond level. Optimizing the deployment delay of virtual networks has become increasingly important. In addition, most of the node sorting methods in existing virtual network mapping methods only consider multi-topological attributes and some global resources, without considering the functional constraints of node deployment delay and link deployment delay, and do not consider node deployment delay in the sorting algorithm. Although such a node sorting strategy can more accurately reflect the resource status of any node in the network, the disadvantage is that it will greatly reduce the sensitivity of the virtual network mapping algorithm to deployment delay, resulting in poor performance of the virtual network mapping algorithm in terms of delay. Summary of the Invention

[0005] The purpose of the present invention is to provide a virtual network mapping method based on node delay perception that introduces node deployment delay and link deployment delay into physical node sorting, virtual node mapping, and virtual link mapping, reduces the deployment delay of virtual networks, improves the revenue-expenditure ratio of virtual network mapping, and achieves a balance between the performance and delay of the virtual network mapping algorithm on the premise of ensuring the mapping success rate of virtual networks.

[0006] To achieve the above object, the present invention adopts the following technical solutions: A virtual network mapping method based on node delay perception, the method comprising the following steps in sequence:

[0007] (1) Virtual node sorting: According to the topological attributes of virtual nodes and the global network resources, obtain the virtual node sorting value according to the node sorting algorithm, and sort the virtual nodes according to the virtual node sorting value; the topological attributes of the virtual nodes include node degree, node link strength, node intimacy, and link interference; the global network resources of the virtual nodes include the node capacity of the entire virtual network and the link bandwidth of the entire virtual network;

[0008] (2) Physical node sorting: According to the topological attributes of virtual nodes and the global network resources, through the physical node deployment delay measurement method, introduce the physical node delay importance into the node sorting algorithm, and after weighting the deployment delay of physical nodes, the topological attributes of nodes, and the global network resources, sort the physical nodes through the physical node sorting algorithm based on delay perception to obtain the physical node sorting value;

[0009] (3) Perform mapping of virtual nodes to physical nodes: Perform virtual node mapping according to the virtual node sorting value and the physical node sorting value;

[0010] (4) Perform mapping of virtual links to physical paths: According to the physical nodes mapped by the end nodes of the virtual link, use the K - shortest path algorithm to find K shortest physical paths between these two physical nodes.

[0011] In step (1), the node degree refers to the sum of the number of all directly connected links to this virtual node, and its formula is as follows:

[0012]

[0013] Wherein, represents the sum of the number of all directly connected links; to the virtual node

[0014] The node link strength refers to the sum of the bandwidths of all directly connected links to this virtual node, and its formula is:

[0015]

[0016] Wherein, represents the virtual link between virtual node and virtual node ; represents the bandwidth of virtual link ;

[0017] The node intimacy refers to the reciprocal of the sum of the shortest distances from a virtual node to all other virtual nodes in the virtual network, and is used to describe the centrality of the virtual node in the virtual network. The formula is as follows:

[0018]

[0019] Among them, respectively represent the node positions of the virtual nodes ; represents the Euclidean distance between the virtual nodes ;

[0020] The link interference refers to the interference attribute of the link, aiming to find the link with the least interference and average number of branches. The formula is as follows:

[0021]

[0022] In the formula, represents the node degree of the virtual node ; represents the node intimacy of the virtual node ;

[0023] In step (1), the node sorting algorithm is as follows:

[0024] First, define the resource amount V of any virtual node in the virtual network G

[0025]

[0026] Among them, represents the capacity of the virtual node ; is the link strength of the virtual node ; is the link interference of the virtual link ;

[0027] Then, further quantify the interaction between any two virtual nodes in the network according to the resource amount of the virtual node :

[0028]

[0029] Among them, represents the transmission delay of the virtual link , and its value is equal to the sum of the transmission delays of each link in the path ;

[0030] According to the above formula, obtain any virtual node Resource relationship with the remaining virtual nodes in the entire virtual network:

[0031]

[0032] Further quantify the virtual node In the global network resource volume The proportion of the virtual node The interaction between the virtual node and the remaining nodes In the global network:

[0033]

[0034]

[0035] According to the topological attributes of the virtual node And the global network resources, its sorting value is defined as:

[0036]

[0037] Among them, Is the sorting value of the virtual node α and β are respectively And The weight factors of, and α + β = 1.

[0038] The specific content of step (2) is:

[0039] First, define the transmission delay of the physical link As:

[0040]

[0041] Among them, from the physical node To the physical node The physical link is denoted as Is the position of the physical node , Represents the Euclidean distance between the physical nodes , Bl is the side length of the physical network range;

[0042] The physical node deployment delay measurement method is:

[0043]

[0044] In the formula, Is the deployment delay of the physical node ; Is the delay importance of the physical node ;

[0045] The physical node sorting algorithm based on delay perception is as follows:

[0046]

[0047] Among them, is the sorting value of the physical node α, β, and γ are the weight factors of the physical node topology attribute, global network resource, and functional attribute respectively, used to balance, α + β + γ = 1, is the physical node The proportion of the interaction between and the rest of the nodes in the global network; is the physical node In the global network resource volume The proportion in;

[0048] The specific steps of step (3) are as follows:

[0049] (3a) When the virtual network G V Arrives at the underlying physical network G S At this time, according to the virtual node sorting value and the physical node sorting value, the virtual nodes and physical nodes are stored in their respective node sets respectively, and then these two node sets are sorted in descending order according to the virtual node sorting value and the physical node sorting value. The virtual node with the highest node sorting value has the priority mapping right;

[0050] (3b) Assume that the virtual node Has the highest node sorting value, then the virtual node Is mapped preferentially: First, select the physical node with the highest physical node sorting value from the physical node set If the physical node Can meet the resource constraints of the virtual node , that is, the node CPU resource and the node capacity, and the function constraints, that is, the node location and the node deployment time, then the virtual node Is successfully mapped to the physical node On, and allocate resources for its use according to the requirements of the virtual node . Its constraint conditions are specifically expressed as follows:

[0051]

[0052]

[0053]

[0054]

[0055] Among them, is the CPU resource of the physical node ; is the CPU resource of the virtual node ; is the capacity of the physical node ; is the capacity of the virtual node ; is the deployment delay of the physical node ; is the deployment delay of the virtual node ; represents the Euclidean distance between the physical node and the virtual node ; is the maximum position deviation of the virtual node ;

[0056] If the physical nodes do not all meet the constraint conditions of the virtual node , then select the physical node with the second highest sorting value from the physical node set for constraint condition comparison. If all the constraint conditions are met, map the virtual node to the physical node and allocate resources for it according to the requirements of the virtual node for its use; if the physical node still cannot meet the constraint conditions of the virtual node , continue to search in the physical node set until a physical node that meets the constraint conditions is found; if no physical node that meets the constraint conditions of the virtual node is found after traversing the entire physical node set, then the mapping of the virtual node fails, that is, the virtual network G V will be directly rejected;

[0057] (3c) For the remaining nodes, repeat step (3b) until all virtual nodes are successfully mapped and the virtual node mapping phase ends.

[0058] The said step (4) specifically includes the following steps:

[0059] (4a) First, sort all virtual links in descending order according to their bandwidth requirements, and preferentially select the virtual link with the largest bandwidth requirement for mapping;

[0060] (4b) Assume that there is a virtual link between the virtual nodes and and the virtual link The bandwidth demand is the largest, and it is known that in the node mapping stage, the virtual node and Already mapped to a physical node and Then the virtual link Mapping to physical nodes On the physical path between

[0061] (4c) In the physical path search phase, the K-shortest path algorithm is first used to select k physical nodes. and The shortest path between them, and judge in order whether these k physical paths satisfy the virtual link at the same time The bandwidth requirements and functional requirements of the physical node include transmission delay and link deployment delay; and The shortest path between Virtual Link The virtual link Successfully mapped to the physical path On the physical path The resource allocation requirements are as follows:

[0062]

[0063]

[0064]

[0065] in, For physical path bandwidth resources, Virtual Link bandwidth resources; For physical path The transmission delay of Virtual Link transmission delay; For physical path Link deployment delay, Virtual Link Link deployment delay;

[0066] (4d) The remaining virtual links are mapped in the same way as in (4b) and (4c) until all virtual links are successfully mapped. At this point, the virtual network G V The mapping is completed.

[0067] As can be seen from the above technical solutions, the beneficial effects of the present invention are as follows: First, compared with the prior art, the present invention incorporates node deployment delay and link deployment delay, calculates the node sorting value by using more topological attributes and global network resources, and takes into account the deployment delay of physical nodes and the transmission delay of physical links, which can more accurately represent the importance of any node in the network in the entire network; Second, the present invention is first practically utilized in the calculation of the sorting value of physical nodes, no longer simply regarding functional attributes as constraints, thereby improving the sensitivity of the virtual network mapping algorithm to deployment delay; Third, compared with the prior art, the present invention reduces the virtual network deployment delay and improves the virtual network mapping revenue-expenditure ratio on the premise of ensuring the mapping success rate of the virtual network, achieving the purpose of balancing the performance and delay of the virtual network mapping algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 is the flowchart of the method of the present invention;

[0069] Figure 2 is the virtual network mapping deployment delay;

[0070] Figure 3 is the virtual network mapping revenue-expenditure ratio. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0071] As Figure 1 shown, a virtual network mapping method based on node delay awareness, the method includes the following steps in sequence:

[0072] (1) Virtual node sorting: According to the topological attributes of virtual nodes and global network resources, obtain the virtual node sorting value according to the node sorting algorithm, and sort the virtual nodes according to the virtual node sorting value; the topological attributes of the virtual nodes include node degree, node link strength, node intimacy and link interference; the global network resources of the virtual nodes include the node capacity of the entire virtual network and the link bandwidth of the entire virtual network;

[0073] (2) Physical node sorting: According to the topological attributes of virtual nodes and global network resources, through the physical node deployment delay measurement method, introduce the physical node delay importance into the node sorting algorithm, and after weighting the deployment delay of physical nodes, node topological attributes and global network resources, sort the physical nodes through the physical node sorting algorithm based on delay awareness to obtain the physical node sorting value;

[0074] (3) Perform mapping from virtual nodes to physical nodes: Perform virtual node mapping according to the virtual node sorting value and the physical node sorting value;

[0075] (4) Map the virtual link to the physical path: According to the physical nodes mapped by the end nodes of the virtual link, use the K - shortest path algorithm to find the K shortest physical paths between these two physical nodes.

[0076] In step (1), the node degree refers to the sum of the number of all directly connected links to this virtual node, and its formula is as follows:

[0077]

[0078] Where, represents the sum of the number of all directly connected links to the virtual node ;

[0079] Generally speaking, the higher the node degree, the higher the connection degree of this node in the whole network.

[0080] The node link strength refers to the sum of the bandwidths of all directly connected links to this virtual node, and its formula is:

[0081]

[0082] Where, represents the virtual link between the virtual node and the virtual node ; represents the virtual link ;

[0083] Generally speaking, the higher the link strength of a node, the more likely this node is to be considered first during virtual link mapping and it is easier to complete link mapping with high - bandwidth requirements.

[0084] The node intimacy refers to the reciprocal of the sum of the shortest distances from a virtual node to all other virtual nodes in the virtual network, which is used to describe the centrality of the virtual node in the virtual network. Its formula is as follows:

[0085]

[0086] Where, respectively represent the node positions of the virtual nodes ; represents the Euclidean distance between the virtual nodes ; The higher the node intimacy, the higher the status of this node in the whole network and the easier it is to have connection links or paths with other nodes.

[0087] The link interference refers to the interference attribute of the link, aiming to find the link with the least interference and average number of branches, which is beneficial to reducing the probability of traffic blockage. Its formula is as follows:

[0088]

[0089] In the formula, represents the node degree of the virtual node , represents the node intimacy of the virtual node .

[0090] Among them, the above formula calculates the interference attribute of the link . For the physical network, the interference attribute of the physical path is determined by the interference attribute of the link with the least number of components in the path.

[0091] In step (1), the node sorting algorithm is as follows:

[0092] First, define the virtual network G V for any virtual node in it

[0093]

[0094] where represents the capacity of the virtual node ; is the link strength of the virtual node ; is the link interference of the virtual link ;

[0095] Then, further quantify the interaction between any two virtual nodes in the network according to the resource amount of the virtual node :

[0096]

[0097] where represents the transmission delay of the virtual link , and its value is equal to the sum of the transmission delays of each link in the path ;

[0098] According to the above formula, obtain the resource relationship between any virtual node and the remaining virtual nodes in the entire virtual network:

[0099]

[0100] Further quantify the proportion of the virtual node in the global network resource amount , and the proportion of the interaction between the virtual node and the remaining nodes in the global network:

[0101]

[0102]

[0103] According to the topological attributes of virtual nodes and the global network resources, its sorting value is defined as:

[0104]

[0105] where is the sorting value of the virtual node α and β are respectively and weight factors, and α + β = 1.

[0106] The specific step (2) refers to:

[0107] First, define the transmission delay of the physical link as:

[0108]

[0109] where the physical link from the physical node to the physical node is denoted as is the position of the physical node , represents the Euclidean distance between the physical nodes , and Bl is the side length of the range where the physical network is located;

[0110] The method for measuring the deployment delay of the physical node is:

[0111]

[0112] In the formula, is the deployment delay of the physical node ; is the delay importance of the physical node ;

[0113] The physical node sorting algorithm based on delay perception is:

[0114]

[0115] where is the sorting value of the physical node ; α, β, and γ are respectively the weight factors of the physical node topological attributes, global network resources, and functional attributes, used to balance, α + β + γ = 1, is the interaction between the physical node and the remaining nodes The proportion in the global network; is a physical node in the global network resource volume The proportion.

[0116] The sorting value of the physical node has comprehensively considered the topological attributes of the physical node itself, the interaction with its adjacent physical nodes, and its deployment delay, and comprehensively evaluated and quantified the sorting value of the physical node.

[0117] The specific steps of step (3) include the following steps:

[0118] (3a) When the virtual network G V reaches the underlying physical network G S At this time, according to the virtual node sorting value and the physical node sorting value, the virtual nodes and physical nodes are respectively stored in their respective node sets, and then these two node sets are sorted in descending order according to the virtual node sorting value and the physical node sorting value. The virtual node with the highest node sorting value has the priority mapping right;

[0119] (3b) Assume that the virtual node has the highest node sorting value. Then, the virtual node is preferentially mapped: First, select the physical node with the highest physical node sorting value from the physical node set If the physical node can meet the resource constraints of the virtual node , that is, the node CPU resource and the node capacity, as well as the function constraints, that is, the node location and the node deployment time, then the virtual node is successfully mapped to the physical node and resources are allocated for its use according to the requirements of the virtual node . Its constraint conditions are specifically expressed as follows:

[0120]

[0121]

[0122]

[0123]

[0124] Among them, is the CPU resource of the physical node , is the CPU resource of the virtual node ; is the capacity of the physical node , is the virtual node The capacity; is the physical node The deployment delay of, is the virtual node The deployment delay of; represents the physical node Virtual node The Euclidean distance between, is the virtual node The maximum position deviation of;

[0125] If the physical node does not fully meet the constraints of the virtual node then select the physical node with the second-highest sorting value from the set of physical nodes for constraint comparison. If all constraints are met, then map the virtual node to the physical node and allocate resources for it to use according to the requirements of the virtual node ; If the physical node still cannot meet the constraints of the virtual node then continue to search in the set of physical nodes until a physical node that meets the constraints is found; If no physical node that meets the constraints of the virtual node is found after traversing the entire set of physical nodes, then the virtual node mapping fails, that is, the virtual network G V will be directly rejected;

[0126] (3c) For the remaining nodes, repeat step (3b) until all virtual nodes are successfully mapped, and the virtual node mapping phase ends.

[0127] The specific steps of step (4) include the following steps:

[0128] (4a) First, sort all virtual links in descending order according to their bandwidth requirements, and preferentially select the virtual link with the largest bandwidth requirement for mapping;

[0129] (4b) Assume that there is a virtual link and between the virtual nodes and the virtual link has the largest bandwidth requirement, and it is known that in the node mapping phase, the virtual nodes and have been mapped to the physical nodes and respectively. Then the virtual link has to be mapped to the physical path between the physical nodes ;

[0130] (4c) In the physical path search phase, the K-shortest path algorithm is first used to select k physical nodes. and The shortest path between them, and judge in order whether these k physical paths satisfy the virtual link at the same time The bandwidth requirements and functional requirements of the physical node include transmission delay and link deployment delay; and The shortest path between Virtual Link The virtual link Successfully mapped to the physical path On the physical path The resource allocation requirements are as follows:

[0131]

[0132]

[0133]

[0134] in, For physical path bandwidth resources, Virtual Link bandwidth resources; For physical path The transmission delay of Virtual Link transmission delay; For physical path Link deployment delay, Virtual Link Link deployment delay;

[0135] (4d) The remaining virtual links are mapped in the same way as in (4b) and (4c) until all virtual links are successfully mapped. At this point, the virtual network G V The mapping is completed.

[0136] The simulation experiment uses the improved Salam network topology random generation algorithm to randomly generate the underlying physical network and virtual network. The underlying physical network generates 100 nodes uniformly distributed in the range of 1000×1000, and the arrival of the virtual network follows the Poisson distribution, the arrival time is 1000 units of time, and the expected number of arrivals is 5. The survival time of the virtual network follows the exponential distribution with an expected time unit of 1000.

[0137] like Figure 2As shown in the figure, the virtual network deployment delay of the three algorithms is continuously increasing over time, but the virtual network deployment delay of the algorithm of the present invention is about 10% lower than that of the other two algorithms.

[0138] As Figure 3 shown in the figure, the NRM-VNE algorithm and the VNE-DCC algorithm take into account the local node resources, but only quantify the distance relationship between neighboring nodes and can only save part of the bandwidth resources. Therefore, the mapping revenue-expenditure ratio of the algorithms is relatively low. When quantifying the node sorting value in the present invention, a variety of topological attributes, all network resource factors, and node function attributes are taken into account. Therefore, the mapping revenue-expenditure ratio of the VNE-NDP algorithm of the present invention performs better.

[0139] In summary, the present invention adds the node deployment delay and the link deployment delay, uses more topological attributes and global network resources to calculate the node sorting value, and takes into account the deployment delay of physical nodes and the transmission delay of physical links, which can more accurately show the importance of any node in the entire network; the present invention is the first to practically use it in the calculation of the sorting value of physical nodes, no longer only regarding the function attribute as a constraint condition, thereby improving the sensitivity of the virtual network mapping algorithm to the deployment delay; on the premise of ensuring the mapping success rate of the virtual network, the present invention reduces the virtual network deployment delay, improves the mapping revenue-expenditure ratio of the virtual network, and achieves the purpose of balancing the performance and delay of the virtual network mapping algorithm.

Claims

1. A virtual network mapping method based on node delay awareness, characterized in that: The method includes the following steps in sequence: (1) Virtual node sorting: According to the topological attributes and global network resources of virtual nodes, obtain the virtual node sorting value according to the node sorting algorithm, and sort the virtual nodes according to the virtual node sorting value; the topological attributes of the virtual nodes include node degree, node link strength, node intimacy, and link interference; the global network resources of the virtual nodes include the node capacity of the entire virtual network and the link bandwidth of the entire virtual network; (2) Physical node sorting: According to the topological attributes and global network resources of virtual nodes, through the physical node deployment delay measurement method, introduce the physical node delay importance into the node sorting algorithm, and after weighting the deployment delay of physical nodes, the topological attributes of nodes, and global network resources, sort the physical nodes through the delay-aware physical node sorting algorithm to obtain the physical node sorting value; (3) Perform mapping from virtual nodes to physical nodes: Perform virtual node mapping according to the virtual node sorting value and the physical node sorting value; (4) Perform mapping from virtual links to physical paths: According to the physical nodes mapped by the end nodes of the virtual links, use the K-shortest path algorithm to find K shortest physical paths between these two physical nodes; In step (1), the node degree refers to the sum of the numbers of all directly connected links of the virtual node, and its formula is as follows: Among them, represents the sum of the number of all directly connected links to the virtual node The node link strength refers to the sum of the bandwidths of all directly connected links of the virtual node, and its formula is: Among them, represents a virtual node and a virtual node the virtual link between them, represents the virtual link bandwidth; The node intimacy refers to the reciprocal of the sum of the shortest distances from a virtual node to all other virtual nodes in the virtual network, which is used to describe the centrality of the virtual node in the virtual network, and its formula is as follows: Among them, respectively represent the node positions of virtual nodes . represents the Euclidean distance between virtual nodes . The link interference refers to the interference attribute of the link, aiming to find the link with the least interference and average number of branches, and its formula is as follows: In the formula, represents the node degree of the virtual node , represents the node intimacy of the virtual node ; In step (1), the node sorting algorithm is as follows: First, define the virtual network G V for any virtual node in the resource amount Among them, represents the capacity of the virtual node ; is the link strength of the virtual node ; is the link interference of the virtual link ; According to the resource amount of the virtual node to further quantify the interaction between any two virtual nodes in the network : Among them, represents the transmission delay of the virtual link , and its value is equal to the sum of the transmission delays of each link in this path; Obtain any virtual node according to the above formula Resource relationship with the remaining virtual nodes in the entire virtual network: Further quantify virtual nodes in the global network resource volume proportion, the interaction between virtual nodes and the remaining nodes proportion in the global network: According to the topological attributes of the virtual node and the global network resources, its sorting value is defined as: Among them, is a virtual node sorting value, where α and β are respectively and weight factors, and α + β = 1; The specific content of step (2) is: First, define the transmission delay of the physical link as follows: Among them, from physical node to physical node the physical link is denoted as is the position of physical node represents the Euclidean distance between physical nodes ; Bl is the side length of the range where the physical network is located.​ The physical node deployment delay measurement method is: In the formula, is the deployment delay of the physical node ; is the delay importance of the physical node ; The delay-aware physical node sorting algorithm is: Among them, is a physical node Sorting value; α, β, and γ are the weight factors of the physical node topology attribute, global network resource, and functional attribute respectively, used to balance, α + β + γ = 1, is a physical node The interaction between and the rest of the nodes The proportion in the global network; is a physical node In the global network resource volume The proportion in; The specific content of step (3) includes the following steps: (3a) When the virtual network G V reaches the underlying physical network G S according to the virtual node sorting value and the physical node sorting value, the virtual nodes and physical nodes are stored in their respective node sets respectively, and then these two node sets are sorted in descending order according to the virtual node sorting value and the physical node sorting value. The virtual node with the highest node sorting value has the priority mapping right; (3b) Assume a virtual node has the highest node sorting value, then the virtual node is preferentially mapped as follows: First, select the physical node with the highest physical node sorting value from the set of physical nodes If the physical node can meet the resource constraints of the virtual node , i.e., the node CPU resources and node capacity, and the function constraints, i.e., the node location and node deployment time, then the virtual node is successfully mapped to the physical node , and resources are allocated for its use according to the requirements of the virtual node . Its constraint conditions are specifically expressed as follows: Among them, is the CPU resource of the physical node . is the CPU resource of the virtual node . is the capacity of the physical node . is the capacity of the virtual node . is the deployment delay of the physical node . is the deployment delay of the virtual node . represents the Euclidean distance between the physical node and the virtual node . is the maximum position deviation of the virtual node . If a physical node does not fully meet the constraints of a virtual node , then select the physical node with the second highest sorting value from the physical node set for constraint condition comparison. If all the constraint conditions are met, map the virtual node to the physical node , and allocate resources according to the requirements of the virtual node for its use; if the physical node still cannot meet the constraints of the virtual node , continue to search in the physical node set until a physical node that meets the constraints is found; if no physical node that meets the constraints of the virtual node is found after traversing the entire physical node set, the virtual node mapping fails, that is, the virtual network G V will be directly rejected; (3c) For the remaining nodes, repeat step (3b) until all virtual nodes are successfully mapped and the virtual node mapping stage ends; The specific content of step (4) includes the following steps: (4a) First, sort all virtual links in descending order according to their bandwidth requirements, and preferentially select the virtual link with the largest bandwidth requirement for mapping; (4b) Assume a virtual node and There is a virtual link between them and the virtual link has the largest bandwidth requirement. And it is known that during the node mapping phase, the virtual nodes and have been mapped to the physical nodes and respectively. Then the virtual link should be mapped to the physical path between the physical nodes ; (4c) In the physical path search phase, the K-shortest path algorithm is first used to select k physical nodes. and The shortest path between them, and judge in order whether these k physical paths satisfy the virtual link at the same time The bandwidth requirements and functional requirements of the physical node include transmission delay and link deployment delay; and The shortest path between Virtual Link The virtual link Successfully mapped to the physical path On the physical path The resource allocation requirements are as follows: Among them, is the bandwidth resource of the physical path ; is the bandwidth resource of the virtual link ; is the transmission delay of the physical path ; is the transmission delay of the virtual link ; is the link deployment delay of the physical path ; is the link deployment delay of the virtual link ; (4d) The remaining virtual links are mapped in the same way as in (4b) and (4c) until all virtual links are successfully mapped. At this point, the virtual network G V The mapping is completed.

Images