WO2017117951A1 - 一种虚拟映射方法及装置 - Google Patents
一种虚拟映射方法及装置 Download PDFInfo
- Publication number
- WO2017117951A1 WO2017117951A1 PCT/CN2016/088908 CN2016088908W WO2017117951A1 WO 2017117951 A1 WO2017117951 A1 WO 2017117951A1 CN 2016088908 W CN2016088908 W CN 2016088908W WO 2017117951 A1 WO2017117951 A1 WO 2017117951A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- node
- virtual
- mapping
- nodes
- topology network
- Prior art date
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/12—Discovery or management of network topologies
Definitions
- This document relates to but not limited to the field of communication, and relates to a virtual mapping method and device.
- Network virtualization is designed to create multiple virtual networks on top of a shared physical network resource while being deployed and managed on its own for each virtual network.
- the essence of network virtualization is the sharing of resources, pooling physical network resources, and achieving the purpose of arbitrarily dividing or merging resources, and building virtual networks that meet the needs of upper-layer services.
- SDN Software Defined Network
- SDN is a new type of network innovation architecture. Its core technology is to separate the control plane and forwarding plane of the network to achieve centralized control, thus achieving flexible control of network traffic.
- SDN requires centralized control, and network virtualization requires centralized control. Therefore, applying SDN to network virtualization has become a hot research direction.
- control plane maps multiple virtual networks with different topologies to a common data plane at the same time, and ensures efficient utilization of the underlying resources, which is called virtual network mapping problem.
- Virtual network mapping is divided into first-order mapping and second-order mapping.
- the first-order mapping algorithm considers the node mapping and the link mapping as a whole.
- the algorithm is a traceable mapping algorithm, that is, traversing all the nodes that satisfy the requirements, looking for a feasible link mapping, if no suitable link mapping can be found. Then, back to the last feasible node mapping scheme for recalculation; if a suitable link mapping is found, the corresponding node is added to the mappable set, and the mapping scheme of the next node is continuously calculated.
- the second-order mapping is divided into two stages: node mapping and link mapping. First, all nodes of the virtual network are mapped, and all links of the virtual network are mapped after the node mapping is completed.
- the virtual network mapping method in the related art needs to traverse all the nodes that meet the requirements, and the data processing amount is large, and the factors such as bandwidth and distance are not taken into consideration, and the obtained mapping relationship is inaccurate and the system performance is poor.
- the embodiment of the invention provides a virtual mapping method and device, which solves the problem that the virtual network mapping method in the related art needs to traverse all the nodes that meet the requirements, the data processing amount is large, and the factors such as bandwidth and distance are not taken into consideration. The mapping relationship is inaccurate and the system performance is poor.
- An embodiment of the present invention provides a virtual mapping method, including: calculating resource values of each virtual node in a physical topology network and each virtual node in a virtual topology network;
- a virtual node having the largest resource value in the virtual topology network is used as a root node, and each layer to be mapped node of the breadth-first search tree is sequentially constructed according to the descending order of the virtual topology network resource values, so as to be a node in the virtual topology network.
- a virtual node having the largest resource value in the virtual topology network is used as a root node, and each layer to be mapped node of the breadth-first search tree is sequentially constructed according to the descending order of the virtual topology network resource values, as a virtual topology.
- the mapping order of nodes in the network including:
- each layer to be mapped node of the breadth-first search tree is constructed in order from the root node to serve as a mapping order of nodes in the virtual topology network.
- calculating the attractiveness of all candidate mapping nodes includes:
- each candidate mapping node and each of the adjacent mapped physical nodes is separately weighted as the attraction of each candidate mapping node.
- calculating the attractiveness between each candidate mapping node and each adjacent mapped physical node according to the two-way breadth search algorithm including:
- calculate the attractiveness of all candidate mapping nodes including:
- An embodiment of the present invention provides a virtual mapping apparatus, including:
- a calculation module configured to calculate a resource value of each physical node in the physical topology network and each virtual node in the virtual topology network;
- a building module is configured to use a virtual node with the largest resource value in the virtual topology network as a root node, and sequentially construct each layer to be mapped node of the breadth-first search tree according to the descending order of the virtual topology network resource values, as a virtual The mapping order of nodes in the topology network;
- the mapping module is configured to calculate the attractiveness of all candidate mapping nodes, and select the most attractive nodes of all candidate mapping nodes to map to establish a mapping between the virtual topology network and the physical topology network.
- the building module includes:
- a building unit configured to sort the descending order of the virtual topological network resource values, and construct each layer to be mapped node of the breadth-first search tree in order from the root node to serve as a mapping order of nodes in the virtual topology network.
- mapping module includes:
- a calculating unit configured to calculate an attractive force between each candidate mapping node and each of the mapped physical nodes according to the two-way breadth search algorithm
- a determining unit is arranged to weight the attractiveness between each candidate mapping node and each of the mapped mapped physical nodes as the attraction of each candidate mapping node.
- the computing unit is configured to determine a shortest path between each candidate mapping node and each of the mapped mapped physical nodes according to the two-way breadth search algorithm; calculate an attractiveness between the nodes on each of the shortest paths.
- mapping module implements calculating the attractiveness of all candidate mapping nodes by:
- the embodiment of the invention further provides a computer readable storage medium, wherein the computer readable storage medium stores computer executable instructions, and the computer executable instructions are implemented to implement a virtual mapping method.
- resource values are calculated for each node in the physical network and the virtual network, and candidate nodes are selected according to the calculated resource values, and then the attractiveness of the candidate nodes is calculated, and the nodes with large attraction are selected for mapping.
- the mapping can be completed quickly, and the virtual network mapping method in the related art is solved. It is necessary to traverse all the nodes that meet the requirements, the data processing amount is large, and the bandwidth and distance are not considered. Other factors, the resulting mapping relationship is not accurate, and the system performance is poor.
- Embodiment 1 is a flowchart of a virtual mapping method in Embodiment 1 of the present invention.
- FIG. 2 is a schematic structural diagram of a virtual mapping apparatus in Embodiment 1 of the present invention.
- FIG. 3 is a schematic structural diagram of a virtual mapping device construction module in Embodiment 1 of the present invention.
- FIG. 4 is a schematic structural diagram of a mapping module of a virtual mapping device according to Embodiment 1 of the present invention.
- FIG. 5 is a flowchart of virtual network mapping in Embodiment 2 of the present invention.
- FIG. 6 is a schematic diagram of a physical network topology in Embodiment 2 of the present invention.
- FIG. 7 is a schematic diagram of a virtual network topology in Embodiment 2 of the present invention.
- the embodiment of the invention provides a virtual mapping method.
- the process of the method is as shown in FIG. 1 and includes steps S102 to S108:
- S104 The virtual node with the largest resource value in the virtual topology network is used as the root node, and each layer to be mapped node of the breadth-first search tree is sequentially constructed according to the descending order of the virtual topology network resource values, so as to be the mapping order of the nodes in the virtual topology network. ;
- the embodiment of the present invention calculates resource values of each node in the physical topology network and the virtual topology network, selects candidate nodes according to the calculated resource values, calculates the attractiveness of the candidate nodes, and selects nodes with large attraction points for mapping.
- the network topology is considered in the mapping, the acceptance rate of the virtual network is increased, and the mapping can be completed quickly.
- the virtual network mapping method in the related art needs to traverse all the nodes that meet the requirements, the data processing amount is large, and the bandwidth is not considered. Distance and other factors, the resulting mapping relationship is not accurate, the system performance is poor.
- the virtual node with the largest resource value in the virtual topology network is used as the root node, and each layer to be mapped node of the breadth-first search tree is constructed in descending order of the virtual topology network resource value to serve as a node in the virtual topology network.
- the order of the mapping may include: selecting a virtual node with the largest resource value as the root node from the virtual topology network; sorting the virtual topology network resource values in descending order, and constructing each of the breadth-first search trees in order from the root node. Layers are mapped to nodes as the mapping order of nodes in the virtual topology network.
- the process of calculating the attractiveness of all candidate mapping nodes may include: calculating, respectively, each candidate mapping node and each of the mapped mapped physical nodes according to the two-way breadth search algorithm.
- the attractiveness of each; the attractiveness of the attraction between each candidate mapping node and each of the adjacent mapped physical nodes is respectively weighted as the attraction of each candidate mapping node.
- calculating the attractiveness between each candidate mapping node and each of the mapped mapped physical nodes according to the two-way breadth search algorithm may include: determining each candidate mapping node and phase separately according to the two-way breadth search algorithm. The shortest path between each physical node that has been mapped; the attraction between nodes on each shortest path is calculated.
- the embodiment of the present invention further provides a virtual mapping device, which is shown in FIG. 2, and includes a computing module 10 configured to calculate resource values of each physical node in the physical topology network and each virtual node in the virtual topology network.
- the construction module 20 is coupled with the calculation module 10, and is configured to use the virtual node with the largest resource value in the virtual topology network as the root node, and sequentially construct each layer to be mapped node of the breadth-first search tree according to the descending order of the virtual topology network resource values.
- the mapping module is used as a mapping order of the nodes in the virtual topology network;
- the selection module 30 is coupled with the building module 20, and is configured to select a pre-predetermined unmapped node with the largest resource value in the physical topology network as the candidate mapping node of the current virtual node;
- the module 40 is coupled to the selection module 30, configured to calculate the attractiveness of all candidate mapping nodes, and selects the most attractive nodes among all the candidate mapping nodes for mapping to establish a mapping between the virtual topology network and the physical topology network.
- the building module 20 includes: a selecting unit 202, configured to select a virtual node with the largest resource value as the root node from the virtual topology network; and the building unit 204 is coupled with the selecting unit 202, and is configured to be configured according to The descending order of the virtual topology network resource values, starting from the root node, constructs each layer of the breadth-first search tree to be mapped in order to be the mapping order of the nodes in the virtual topology network.
- the mapping module 40 includes: a calculating unit 402 configured to calculate an attractive force between each candidate mapping node and each of the mapped physical nodes according to the bidirectional breadth search algorithm; 404, coupled with the computing unit 402, is configured to weight the attractiveness between each candidate mapping node and each of the mapped mapped physical nodes as the attraction of each candidate mapping node.
- the calculating unit 402 is configured to respectively determine a shortest path between each candidate mapping node and each adjacent mapped physical node according to the bidirectional breadth search algorithm; calculate an attraction between the nodes on each shortest path .
- the mapping module 40 may implement the calculation of the attractiveness of all candidate mapping nodes by calculating the attractiveness of the candidate mapping nodes according to the following formula: candidate mapping nodes
- the attraction is: among them, Indicates a virtual node to be mapped in the virtual topology network; Express Mapping to candidate mapping nodes in the physical topology network; Express Adjacent virtual nodes that have been mapped, Representation node The i-th adjacent virtual node that has been mapped; Express Map to nodes in the physical topology network, Express Map to nodes in the physical topology network; Representing a node pair with The attraction between the size is p represents the physical topology network with Satisfy bandwidth The shortest path of demand; hop represents the number of hops of the node on path p; Ave(p) represents the average bandwidth of path p, the size is l is a link on path p; Is a neighbor in the virtual topology network with Bandwidth.
- the virtual network mapping method in the related art needs to traverse all the nodes that meet the requirements, the data processing amount is large, and the factors such as bandwidth and distance are not considered, and the obtained mapping relationship is inaccurate.
- the system performance is poor.
- the problem to be solved by the embodiments of the present invention is that the mapping distance between adjacent virtual network (virtual topology network) nodes is too far, and the detection threshold of the first-order mapping algorithm is limited.
- the physical network (physical topology network, or physical network) is provided by a single network facility provider; the computing resource constraints of the node Not considered; link mapping does not support path segmentation.
- the technical solutions and parameter settings of the embodiments of the present invention are as follows:
- the nodes of the physical network and the virtual network are calculated and sorted according to the resource values.
- the node with the largest resource value in the virtual network is used as the root node to construct the breadth-first search tree, and each node of the breadth-first search tree is sorted according to the resource value. , sorted as the mapping order of the nodes. For each node mapping, the first k unmapped nodes with the largest resource value are selected as candidate nodes, the bandwidth-first bidirectional breadth search algorithm is used for link mapping, and the attractiveness of the candidate nodes is calculated, and the most attractive candidate node mapping is selected. And then map the next node until all nodes and links are mapped.
- G s (N s , L s ) with weights.
- N s is a collection of nodes in the physical network
- L s is a collection of links in the physical network.
- G v (N v , L v ) with weights.
- the relevant parameters are as follows:
- N v is a collection of nodes in the virtual network.
- n
- L v is a collection of links in the virtual network.
- the relative resource values of the nodes can be represented by the bandwidth sum of the adjacent links, and the node resource values are defined for the ranking of the nodes.
- the relevant parameters are as follows:
- RV is the resource value of the node, and its size can be expressed as ⁇ l ⁇ L bw(l), where L is a set of adjacent links of the node.
- Adjacent nodes have not been mapped to the physical network
- Virtual network node Map Map to nodes in the physical network
- Virtual network node Adjacent nodes subscripted i in the set of neighboring nodes that have been mapped;
- p indicates the node in the network with Satisfy bandwidth The shortest path of demand
- Hop indicates the hop count of the node on the path p
- Ave(p) indicates the average bandwidth of the path p, and its size can be expressed as
- FIG. 5 The flowchart of completing the virtual network mapping in the embodiment of the present invention is shown in FIG. 5, and the steps are as follows:
- Step S501 the resource value G s and G v of all nodes are calculated virtual network physical network
- Step S502 the physical network and the virtual network are arranged in descending order of resource values
- Step S503 selecting a point with the largest resource value in the virtual network G v as a root node, and constructing a breadth-first search tree;
- Step S504 the nodes of each layer of the breadth-first search tree are arranged in descending order of resource values, and the virtual network vertex mapping order is constructed.
- Step S505 assigning an initial value, and assigning an initial value to the backtracking upper bound and the number of candidate nodes;
- Step S506 mapping the i-th node of the virtual network and the corresponding link, and determining whether the mapping is successful;
- step S507 If the mapping is successful, the process goes to step S507; if the mapping fails, the process goes to step S508.
- Step S507 to see whether i is less than the number of virtual network nodes, if i is less than the number of virtual network nodes, then i is added to 1 and then proceeds to step S506; if i is not less than the number of virtual network nodes, then step S511;
- step S508 it is checked whether i is the root node. If i is the root node, go to step 5S10, if i is not the root node, go to step S509;
- Step S509 checking whether the backtracking value is smaller than the backtracking upper bound. If the backtracking value is less than the backtracking upper bound, searching for the mapped neighboring node that causes the mapping failure, assigning the value to i, and proceeding to step S506; if the backtracking value is not less than the backtracking upper bound, Clear the mapping information, go to step S510;
- Step S510 the virtual network mapping fails, and the algorithm ends.
- step S511 the virtual network mapping is successful, and the algorithm ends.
- Step S10 establishing an initial network physical topology and a network virtual topology according to each node
- Step S20 Calculate resource values of each node of the physical topology network and the virtual topology network, and arrange the nodes in descending order of resource values.
- the sorted network physical topology is as shown in FIG. 6, and the network virtual topology is as shown in FIG. 7.
- step S30 the breadth-first search tree is constructed by using the node c with the highest resource value as the root node in the virtual topology network, and each node of the tree is arranged in descending order of resource values, and the node mapping order of the virtual topology is c, b, a. .
- step S40 nodes and links are sequentially mapped according to the node mapping order of the virtual topology network.
- Step S4011 mapping the virtual topology node c, calculating the resource value of c is 7, and obtaining the node that meets the node resource value requirement in the physical topology and is not mapped has 5, 4, 2 1, 6, 3, the first two nodes 5 and 4 with the largest resource value are selected as the candidate nodes; in step S4012, the node 5 with the largest resource value is selected as the mapping node of the virtual topology root node c.
- step S50 the virtual topology node c is mapped, and the mapping node is the node 5 in the physical topology, and the node and link mapping information is saved.
- step S4021 the virtual topology node b is mapped, and the resource value of b is calculated to be 6, and the nodes that satisfy the node resource value requirement in the physical topology and are not mapped have 4, 2, 1, 6, and 3, and the first two resources with the largest resource value are selected. Nodes 4 and 2 serve as candidate nodes;
- step S4022 when the candidate node is 4, the neighboring node that has been mapped to the node b has only c, and the mapping node of the node c is 5, and the minimum priority of the node 4 to the node 5 that satisfies the bandwidth requirement is obtained by using the bandwidth-first two-way breadth search algorithm.
- the path is 4-5, the attraction between the two points is calculated as 7, the node c has only one adjacent mapped node, and the attractiveness of the candidate node 4 is also 7;
- step S4023 when the candidate node is 2, the neighboring node that has been mapped to the node b has only c, the mapping node of the node c is 5, and the shortest of the node 2 to the node 5 satisfying the bandwidth requirement by using the bandwidth-first two-way breadth search algorithm.
- the path is 2-5, the attraction between the two points is calculated as 6, the node c has only one adjacent mapped node, and the attractiveness of the calculated candidate node 2 is also 6;
- step S4024 the candidate nodes are arranged into 4 and 2 in descending order of attraction, and the node 4 with the strongest attraction is selected as the mapping node of the node b.
- step S50 is performed.
- the virtual topology node b is mapped, the mapping node is the node 4 in the physical topology, the node and link mapping information is saved, and the physical topology link information is updated.
- step S4031 the virtual topology node a is mapped, and the resource value of the calculation a is 5, and the nodes in the physical topology that satisfy the node resource value requirement and are not mapped have 2, 1, 6, and 3, and the first two nodes with the largest resource value are selected. 2 and 1 as candidate nodes;
- step S4032 when the candidate node is 2, the neighbor nodes that have been mapped to the node a have c and b, the mapping node of the node c is 5, and the mapping node of the node b is 4.
- the bandwidth-first two-way breadth search algorithm to obtain the shortest path that satisfies the bandwidth requirement, the shortest path from node 5 to node 2 is 5-2, the attraction between two points is calculated to be 6, and the shortest path from node 4 to node 2 is 4. -1-2, count The attraction between the two points is 2.5.
- Node b has two adjacent nodes that have been mapped, and the attractiveness of 2 for calculating candidate nodes is also 4.6;
- step S4033 when the candidate node is 1, the neighboring nodes that have been mapped to the node a have c and b, the mapping node of the node c is 5, and the mapping node of the node b is 4.
- the bandwidth-first two-way breadth search algorithm to obtain the shortest path that satisfies the bandwidth requirement, the shortest path from node 5 to node 1 is 5-4-1, and the attraction between nodes is 2.75, and the shortest path from node 4 to node 1 is calculated.
- 4-1 calculate the attraction between the two points to 5.
- Node b has two adjacent nodes that have been mapped, and the attractiveness of 2 for calculating candidate nodes is also 3.65;
- step S4034 the candidate nodes are arranged in the descending order of attraction to 2 and 1, and the node 2 with the strongest attraction is selected as the mapping node of the node b.
- step S50 is performed.
- the virtual topology node a is mapped, the mapping node is node 2 in the physical topology, the node and link mapping information is saved, and the physical topology bandwidth information is updated.
- step S60 is performed, that is, the virtual topology mapping is completed, the node c is mapped to the node 5, the node b is mapped to the node 4, and the mapping of the link cb is 4-5, and the node a is mapped.
- the mapping of link ac is 5-2, the mapping of link ab is 4-1-2, the physical topology information is updated, and the virtual network mapping is completed.
- the virtual mapping method calculates resource values for each node in the physical topology network and the virtual topology network, and selects a predetermined unmapped node with the largest resource value as a candidate node according to the calculated resource value, and comprehensively considers The factors of bandwidth and distance, calculate the attractiveness of the candidate nodes, select the nodes with strong attraction to map, and select the shortest path with bandwidth priority that meets the bandwidth requirement as the link mapping.
- the network topology is considered, the acceptance rate of the virtual network is increased, the bandwidth and the distance are comprehensively considered in the link mapping, and the link cost of the virtual network is reduced.
- the embodiment of the invention further provides a computer readable storage medium, wherein the computer readable storage medium stores computer executable instructions, and the computer executable instructions are implemented to implement a virtual mapping method.
- each module/unit in the above embodiment may be implemented in the form of hardware, for example, by implementing an integrated circuit to implement its corresponding function, or may be implemented in the form of a software function module, for example, executing a program stored in the memory by a processor. / instruction to achieve its corresponding function.
- This application is not limited to any specific combination of hardware and software.
- the above technical method increases the acceptance rate of the virtual network and reduces the link cost of the virtual network.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
一种虚拟映射方法及装置，该虚拟映射方法包括：计算物理拓扑网中每个物理节点和虚拟拓扑网中每个虚拟节点的资源值；以虚拟拓扑网中资源值最大的虚拟节点作为根节点，按照虚拟拓扑网资源值的降序排序依次构建广度优先搜索树的每一层待映射节点，以作为虚拟拓扑网中节点的映射顺序；选取物理拓扑网中资源值最大的前预定个未映射的节点作为当前虚拟节点的候选映射节点；计算所有候选映射节点的吸引力，并选取所有候选映射节点中吸引力最大的节点进行映射，以建立虚拟拓扑网与物理拓扑网的映射。上述技术方案解决了相关技术中的虚拟网络映射方法需要遍历所有满足需求的节点集合，数据处理量较大的问题。
Description
本文涉及但不限于通讯领域，涉及一种虚拟映射方法及装置。
互联网经过几十年的发展，展示了其强大的生命力和广阔的发展空间。但随着技术的发展，传统互联网由于自身体系架构的缺陷，难以适应新兴业务的不断发展，网络虚拟化技术被认为是解决现有网络体系僵化问题，构建下一代互联网最好的方案。
网路虚拟化旨在一个共享的物理网络资源之上创建多个虚拟网络，同时为每个虚拟网络独自地部署以及管理。网络虚拟化的本质是资源的共享，将物理网络资源池化，达到资源任意分割或者合并的目的，用于构建满足上层服务需求的虚拟网络。SDN(软件定义网络，Software Defined Network)是一种新型网络创新架构，其核心技术是将网络的控制平面和转发平面面相分离，实现集中控制，从而实现网络流量的灵活控制。SDN要求集中控制，网络虚拟化要求集中控制，因此，将SDN运用于网络虚拟化是已然成为一个热门的研究方向。
而在这其中，网络虚拟化映射又是一个研究重点。在不破坏底层资源约束的前提下，控制平面将多个具有不同拓扑的虚拟网络同时映射到共用的数据平面中，并且保证底层资源的高效利用率，被称为虚拟网络映射问题。
虚拟网络映射被分为一阶映射和二阶映射。一阶映射算法是将节点映射和链路映射作为一个整体考虑，算法是可回溯的映射算法，即遍历所有满足需求的节点集合，寻找可行的链路映射，如果找不到合适的链路映射，则回溯至上一次可行的节点映射方案进行重新计算；如果找到合适的链路映射，则将对应的节点加入至可映射集合，继续计算下一个节点的映射方案。二阶映射分为节点映射和链路映射两个阶段，先映射虚拟网所有的节点，在节点映射完毕后再映射虚拟网所有的链路。
相关技术中的虚拟网络映射方法，需要遍历所有满足需求的节点集合，数据处理量较大，且没有考虑到带宽和距离等因素，得到的映射关系不准确，系统性能较差。
发明内容
以下是对本文详细描述的主题的概述。本概述并非是为了限制权利要求的保护范围。
本发明实施例提供一种虚拟映射方法及装置，解决了相关技术中的虚拟网络映射方法需要遍历所有满足需求的节点集合，数据处理量较大，且没有考虑到带宽和距离等因素，得到的映射关系不准确，系统性能较差的问题。
本发明实施例提供一种虚拟映射方法，包括：计算物理拓扑网中每个物理节点和虚拟拓扑网中每个虚拟节点的资源值；
以所述虚拟拓扑网中资源值最大的虚拟节点作为根节点，按照所述虚拟拓扑网资源值的降序排序依次构建广度优先搜索树的每一层待映射节点，以作为虚拟拓扑网中节点的映射顺序；
选取所述物理拓扑网中资源值最大的前预定个未映射的节点作为当前虚拟节点的候选映射节点；
计算所有候选映射节点的吸引力，并选取所有候选映射节点中吸引力最大的节点进行映射，以建立虚拟拓扑网与物理拓扑网的映射。
可选地，以所述虚拟拓扑网中资源值最大的虚拟节点作为根节点，按照所述虚拟拓扑网资源值的降序排序依次构建广度优先搜索树的每一层待映射节点，以作为虚拟拓扑网中节点的映射顺序，包括：
从所述虚拟拓扑网中选取资源值最大的虚拟节点作为根节点；
按照所述虚拟拓扑网资源值的降序排序，从所述根节点开始依次构建广度优先搜索树的每一层待映射节点，以作为虚拟拓扑网中节点的映射顺序。
可选地，计算所有候选映射节点的吸引力包括：
根据双向广度搜索算法分别计算每个候选映射节点与相邻已映射的每个
物理节点之间的吸引力；
分别将每个候选映射节点与相邻已映射的每个物理结点之间的吸引力加权求和，作为每个候选映射节点的吸引力。
可选地，根据双向广度搜索算法分别计算每个候选映射节点与相邻已映射的每个物理节点之间的吸引力，包括：
根据双向广度搜索算法分别确定每个候选映射节点与相邻已映射的每个物理节点之间的最短路径；
计算每条最短路径上节点之间的吸引力。
可选地，计算所有候选映射节点的吸引力，包括：
候选映射节点
吸引力按照如下公式计算：
其中，表示虚拟拓扑网中待映射的虚拟节点；
表示
映射到物理拓扑网中的候选映射节点；
表示
已经映射过的相邻的虚拟节点，
表示节点
已经映射过的第i个相邻虚拟节点；
表示
映射到物理拓扑网中的节点，
表示
映射到物理拓扑网中的节点；
表示节点对
和
之间的吸引力，大小为
p表示物理拓扑网中
和
满足带宽
需求的最短路径；hop表示路径p上节点的跳数；Ave(p)表示路径p的平均带宽，大小为
l是路径p上的一条链路；
是虚拟拓扑网中相邻节点
和
的带宽。
本发明实施例一种虚拟映射装置，包括：
计算模块，设置为计算物理拓扑网中每个物理节点和虚拟拓扑网中每个虚拟节点的资源值；
构建模块，设置为以所述虚拟拓扑网中资源值最大的虚拟节点作为根节点，按照所述虚拟拓扑网资源值的降序排序依次构建广度优先搜索树的每一层待映射节点，以作为虚拟拓扑网中节点的映射顺序；
选取模块，设置为选取所述物理拓扑网中资源值最大的前预定个未映射
的节点作为当前虚拟节点的候选映射节点；
映射模块，设置为计算所有候选映射节点的吸引力，并选取所有候选映射节点中吸引力最大的节点进行映射，以建立虚拟拓扑网与物理拓扑网的映射。
可选地，所述构建模块包括：
选取单元，设置为从所述虚拟拓扑网中选取资源值最大的虚拟节点作为根节点；
构建单元，设置为按照所述虚拟拓扑网资源值的降序排序，从所述根节点开始依次构建广度优先搜索树的每一层待映射节点，以作为虚拟拓扑网中节点的映射顺序。
可选地，所述映射模块包括：
计算单元，设置为根据双向广度搜索算法计算每个候选映射节点与相邻已映射的各物理结点之间的吸引力；
确定单元，设置为将每个候选映射节点与相邻已映射的各物理节点之间吸引力加权求和，作为每个候选映射节点的吸引力。
可选地，所述计算单元是设置为根据双向广度搜索算法确定每个候选映射节点与相邻已映射的各物理节点之间的最短路径；计算每条最短路径上节点之间的吸引力。
可选地，所述映射模块通过如下方式实现计算所有候选映射节点的吸引力：
按照如下公式计算候选映射节点的吸引力：
候选映射节点
吸引力为：
其中，
表示虚拟拓扑网中待映射的虚拟节点；
表示
映射到物理拓扑网中的候选映射节点；
表示
已经映射过的相邻的虚拟节点，其中
表示节点
已经映射过的第i个相邻虚拟节点；
表示
映射到物理拓扑网中的节点，其中
表示
映射到物理拓扑网中的节点；
表示节点对
和
之间的吸引力，大小为
p表示物理拓扑网中
和
满足带宽
需求的最短路径；hop表示路径p上节点的跳数；Ave(p)表示路径p的平均带宽，大小为
其中l是路径p上的一条链路；
是虚拟拓扑网中相邻节点
和
的带宽。
本发明实施例还提供一种计算机可读存储介质，所述计算机可读存储介质中存储有计算机可执行指令，所述计算机可执行指令被执行时实现虚拟映射方法。
本发明实施例为物理网和虚拟网中的每个节点计算资源值，根据计算的资源值，选取候选节点，再计算候选节点的吸引力，选取吸引力大的节点进行映射，在节点映射时考虑网络拓扑，增加了虚拟网的接受率，可以快速完成映射，解决了相关技术中的虚拟网络映射方法，需要遍历所有满足需求的节点集合，数据处理量较大，且没有考虑到带宽和距离等因素，得到的映射关系不准确，系统性能较差的问题。
在阅读并理解了附图和详细描述后，可以明白其它方面。
图1是本发明实施例一中的虚拟映射方法的流程图；
图2是本发明实施例一中的虚拟映射装置的结构示意图；
图3是本发明实施例一中的虚拟映射装置构建模块的结构示意图；
图4是本发明实施例一中的虚拟映射装置的映射模块结构示意图；
图5是本发明实施例二中的虚拟网映射流程图；
图6是本发明实施例二中的物理网拓扑示意图；
图7是本发明实施例二中的虚拟网拓扑示意图。
以下结合附图以及实施例，对本申请进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不限定本申请。
实施例一
本发明实施例提供一种虚拟映射方法，该方法的流程如图1所示，包括步骤S102至S108：
S102，计算物理拓扑网中每个物理节点和虚拟拓扑网中每个虚拟节点的资源值；
S104，以虚拟拓扑网中资源值最大的虚拟节点作为根节点，按照虚拟拓扑网资源值的降序排序依次构建广度优先搜索树的每一层待映射节点，以作为虚拟拓扑网中节点的映射顺序；
S106，选取物理拓扑网中资源值最大的前预定个未映射的节点作为当前虚拟节点的候选映射节点；
S108，计算所有候选映射节点的吸引力，并选取所有候选映射节点中吸引力最大的节点进行映射，以建立虚拟拓扑网与物理拓扑网的映射。
本发明实施例计算物理拓扑网和虚拟拓扑网中的每个节点的资源值，根据计算的资源值，选取候选节点，再计算候选节点的吸引力，选取吸引力大的节点进行映射，在节点映射时考虑网络拓扑，增加了虚拟网的接受率，可以快速完成映射，解决了相关技术中的虚拟网络映射方法需要遍历所有满足需求的节点集合，数据处理量较大，且没有考虑到带宽和距离等因素，得到的映射关系不准确，系统性能较差的问题。
实现过程中，以虚拟拓扑网中资源值最大的虚拟节点作为根节点，按照虚拟拓扑网资源值的降序排序依次构建广度优先搜索树的每一层待映射节点，以作为虚拟拓扑网中节点的映射顺序。在本实施例中，上述过程可以包括：从虚拟拓扑网中选取资源值最大的虚拟节点作为根节点；按照虚拟拓扑网资源值的降序排序，从根节点开始依次构建广度优先搜索树的每一层待映射节点，以作为虚拟拓扑网中节点的映射顺序。
在本实施例中，计算所有候选映射节点的吸引力的过程可以包括：根据双向广度搜索算法分别计算每个候选映射节点与相邻已映射的每物理节点之
间的吸引力；分别将每个候选映射节点与相邻已映射的每个物理结点之间的吸引力加权求和，作为每个候选映射节点的吸引力。
在本实施例中，根据双向广度搜索算法分别计算每个候选映射节点与相邻已映射的每个物理节点之间的吸引力可以包括：根据双向广度搜索算法分别确定每个候选映射节点与相邻已映射的每个物理节点之间的最短路径；计算每条最短路径上节点之间的吸引力。
实现的过程中，候选映射节点
吸引力可以按照如下公式计算：
其中，
表示虚拟拓扑网中待映射的虚拟节点；
表示
映射到物理拓扑网中的候选映射节点；
表示
已经映射过的相邻的虚拟节点，
表示节点
已经映射过的第i个相邻虚拟节点；
表示
映射到物理拓扑网中的节点，
表示
映射到物理拓扑网中的节点；
表示节点对
和
之间的吸引力，大小为
p表示物理拓扑网中
和
满足带宽
需求的最短路径；hop表示路径p上节点的跳数；Ave(p)表示路径p的平均带宽，大小为
l是路径p上的一条链路；
是虚拟拓扑网中相邻节点
和
的带宽。
本发明实施例还提供一种虚拟映射装置，其结构示意如图2所示，包括：计算模块10，设置为计算物理拓扑网中每个物理节点和虚拟拓扑网中每个虚拟节点的资源值；构建模块20，与计算模块10耦合，设置为以虚拟拓扑网中资源值最大的虚拟节点作为根节点，按照虚拟拓扑网资源值的降序排序依次构建广度优先搜索树的每一层待映射节点，以作为虚拟拓扑网中节点的映射顺序；选取模块30，与构建模块20耦合，设置为选取物理拓扑网中资源值最大的前预定个未映射的节点作为当前虚拟节点的候选映射节点；映射模块40，与选取模块30耦合，设置为计算所有候选映射节点的吸引力，并选取所有候选映射节点中吸引力最大的节点进行映射，以建立虚拟拓扑网与物理拓扑网的映射。
可选地，如图3所示，构建模块20包括：选取单元202，设置为从虚拟拓扑网中选取资源值最大的虚拟节点作为根节点；构建单元204，与选取单元202耦合，设置为按照虚拟拓扑网资源值的降序排序，从根节点开始依次构建广度优先搜索树的每一层待映射节点，以作为虚拟拓扑网中节点的映射顺序。
可选地，如图4所示，映射模块40包括：计算单元402，设置为根据双向广度搜索算法计算每个候选映射节点与相邻已映射的各物理结点之间的吸引力；确定单元404，与计算单元402耦合，设置为将每个候选映射节点与相邻已映射的各物理节点之间吸引力加权求和，作为每个候选映射节点的吸引力。
可选地，计算单元402是设置为根据双向广度搜索算法分别确定每个候选映射节点与相邻已映射的每个物理节点之间的最短路径；计算每条最短路径上节点之间的吸引力。
可选地，映射模块40可以通过如下方式实现计算所有候选映射节点的吸引力：按照如下公式计算候选映射节点的吸引力：候选映射节点
吸引力为：
其中，
表示虚拟拓扑网中待映射的虚拟节点；
表示
映射到物理拓扑网中的候选映射节点；
表示
已经映射过的相邻的虚拟节点，
表示节点
已经映射过的第i个相邻虚拟节点；
表示
映射到物理拓扑网中的节点，
表示
映射到物理拓扑网中的节点；
表示节点对
和
之间的吸引力，大小为
p表示物理拓扑网中
和
满足带宽
需求的最短路径；hop表示路径p上节点的跳数；Ave(p)表示路径p的平均带宽，大小为
l是路径p上的一条链路；
是虚拟拓扑网中相邻节点
和
的带宽。
实施例二
相关技术中的虚拟网络映射方法，需要遍历所有满足需求的节点集合，数据处理量较大，且没有考虑到带宽和距离等因素，得到的映射关系不准确，
系统性能较差。本发明实施例要解决的问题是相邻的虚拟网(虚拟拓扑网)节点映射距离过远问题，以及一阶映射算法的探测阀值的限制。
本发明实施例在实现时，设置满足以下限制，以使得实施例实现的效果更好：物理网络(物理拓扑网，或称为物理网)是由单个网络设施提供商提供；节点的计算资源约束不予考虑；链路映射不支持路径分割。本发明实施例的技术方案及参数设置如下：
a)具体规则：
将物理网和虚拟网的节点计算资源值并按照资源值排序，虚拟网中资源值最大的节点作为根结点构建广度优先搜索树，将广度优先搜索树的每层节点按照资源值大小进行排序，排序后作为节点的映射顺序。对于每个节点映射，选择资源值最大的前k个未映射节点作为候选节点，采用带宽优先的双向广度搜索算法进行链路映射，并计算候选节点的吸引力，选取吸引力最大的候选节点映射，然后映射下一个节点，直到所有节点和链路映射完毕。
b)相关参数：
物理网络图可以通过一个带权值的无向图Gs＝(Ns,Ls)表示。相关参数如下：
Ns：为物理网中节点的集合；
Ls：为物理网中链路的集合。
虚拟网络图可以通过一个带权值的无向图Gv＝(Nv,Lv)表示。相关参数如下：
Nv：为虚拟网中节点的集合。n＝|Nv|为节点个数；
Lv：为虚拟网中链路的集合。
节点的相对资源值可以通过相邻链路的带宽和来表示，定义节点资源值以用于节点的排名。相关参数如下：
RV：为节点的资源值，其大小可以表示为∑l∈Lbw(l)，其中，L是节点的相邻链路的集合。
吸引力与路径的平均带宽成正比，与路径的节点跳数成反比。相关参数
如下：
p：表示理网中节点
和
满足带宽
需求的最短路径；
hop：表示路径p上节点的跳数；
Ave(p)：表示路径p的平均带宽，其大小可以表示为
本发明实施例完成虚拟网映射的流程图见图5，步骤如下：
步骤S501，分别计算物理网Gs和虚拟网Gv所有节点的资源值；
步骤S502，将物理网络和虚拟网络按照资源值的降序排列；
步骤S503，选取虚拟网Gv中资源值最大的点，作为根节点，构建广度优先搜索树；
步骤S504，将广度优先搜索树的每一层节点按照资源值的降序排列，构建虚拟网顶点映射顺序；
步骤S505，赋初值，将回溯上界和候选节点个数赋初值；
步骤S506，映射虚拟网第i个节点和相应链路，判断是否映射成功；
若映射成功，则转步骤S507；，若映射失败则转步骤S508。
步骤S507，查看i是否小于虚拟网节点个数，若i小于虚拟网节点个数，则i加1后转到步骤S506；若i不小于虚拟网节点个数，则转步骤S511；
步骤S508若映射失败，则查看i是否是根结点。若i是根结点，则转到步骤5S10，若i不是根节点，转到步骤步骤S509；
步骤S509查看回溯值是否小于回溯上界，若回溯值小于回溯上界，则找寻导致映射失败的已映射相邻节点，赋值给i，转到步骤S506；若回溯值不小于回溯上界，则清除映射信息，转到步骤S510；
步骤S510虚拟网映射失败，算法结束；
步骤S511虚拟网映射成功，算法结束。
下面结合附图和具体实施实例对本申请作进一步的详细描述。
本发明实施例提供的虚拟映射方法包括：
步骤S10，根据每个节点建立初始网络物理拓扑和网络虚拟拓扑；
步骤S20，计算物理拓扑网和虚拟拓扑网的每个节点的资源值，并将节点按照资源值的降序排列，排序后的网络物理拓扑如图6所示，网络虚拟拓扑如图7所示。
步骤S30，虚拟拓扑网中以资源值最高的节点c为根节点构建广度优先搜索树，将树的每一层节点按照资源值的降序排列，构建虚拟拓扑的节点映射顺序为c、b、a。
步骤S40，按照虚拟拓扑网的节点映射顺序，依次映射节点和链路。
在执行S40时，可以按照如下过程执行：步骤S4011，映射虚拟拓扑节点c，计算c的资源值为7，得到物理拓扑中满足节点资源值需求且未被映射的节点有5、4、2、1、6、3，选取资源值最大的前两个节点5和4作为候选节点；步骤S4012，在候选节点中选取资源值最大的节点5作为虚拟拓扑根节点c的映射节点。
步骤S50，虚拟拓扑节点c映射完毕，映射节点为物理拓扑中节点5，保存节点和链路映射信息。
随后，重复执行S10至S30，再次执行到S40时，按照如下过程执行：
步骤S4021，映射虚拟拓扑节点b，计算b的资源值为6，得到物理拓扑中满足节点资源值需求且未被映射的节点有4、2、1、6、3，选取资源值最大的前两个节点4和2作为候选节点；
步骤S4022，候选节点为4时，得到节点b的已经映射的相邻节点只有c，节点c的映射节点为5，利用带宽优先的双向广度搜索算法得到满足带宽需求的节点4到节点5的最短路径为4-5，计算两点之间吸引力为7,节点c只有一个相邻已映射节点，计算候选节点的4的吸引力也为7；
步骤S4023，候选节点为2时，得到节点b的已经映射的相邻节点只有c，节点c的映射节点为5，利用带宽优先的双向广度搜索算法得满足带宽需求的节点2到节点5的最短路径为2-5，计算两点之间吸引力为6,节点c只有一个相邻已映射节点，计算候选节点的2的吸引力也为6；
步骤S4024，将候选节点按照吸引力的降序排列为4、2，选择吸引力最大的节点4作为节点b的映射节点。
在按照上述过程执行完S40的过程后，执行步骤S50，此时对应的是虚拟拓扑节点b映射完毕，映射节点为物理拓扑中节点4，保存节点和链路映射信息，更新物理拓扑链路信息。
随后，再次重复执行S10至S30，当再次执行到S40时，再次按照如下的过程执行：
步骤S4031，映射虚拟拓扑节点a，计算a的资源值为5，得到物理拓扑中满足节点资源值需求且未被映射的节点有2、1、6、3，选取资源值最大的前两个节点2和1作为候选节点；
步骤S4032，候选节点为2时，得到节点a的已经映射的相邻节点有c和b，节点c的映射节点为5，节点b的映射节点为4。利用带宽优先的双向广度搜索算法得到满足带宽需求的最短路径，得到节点5到节点2的最短路径为5-2，计算两点之间吸引力为6,节点4到节点2的最短路径为4-1-2，计
算两点之间吸引力为2.5。节点b有两个已经映射过的相邻节点，计算候选节点的2的吸引力也为4.6；
步骤S4033，候选节点为1时，得到节点a的已经映射的相邻节点有c和b，节点c的映射节点为5，节点b的映射节点为4。利用带宽优先的双向广度搜索算法得到满足带宽需求的最短路径，得到节点5到节点1的最短路径为5-4-1，计算两点之间吸引力为2.75,节点4到节点1的最短路径为4-1，计算两点之间吸引力为5。节点b有两个已经映射过的相邻节点，计算候选节点的2的吸引力也为3.65；
步骤S4034，将候选节点按照吸引力的降序排列为2、1，选择吸引力最大的节点2作为节点b的映射节点。
在按照上述过程执行完S40的过程后，执行步骤S50，此时的虚拟拓扑节点a映射完毕，映射节点为物理拓扑中节点2，保存节点和链路映射信息，更新物理拓扑带宽信息。
在上述过程将三个虚拟节点都映射完毕后，执行步骤S60，即虚拟拓扑映射完毕，节点c映射到节点5，节点b映射到节点4，链路c-b的映射为4-5，节点a映射到节点2，链路a-c的映射为5-2，链路a-b的映射为4-1-2，更新物理拓扑信息，虚拟网映射完毕。
本发明实施例提供的虚拟映射方法为物理拓扑网和虚拟拓扑网中的每个节点计算资源值，根据计算的资源值，选取资源值最大的前预定个未映射的节点作为候选节点，综合考虑带宽和距离的因素，计算候选节点的吸引力，选取吸引力大的节点进行映射，并选择满足带宽需求的带宽优先的最短路径作为链路映射。在节点映射时考虑网络拓扑，增加了虚拟网的接受率，在链路映射时综合考虑带宽和距离，降低虚拟网的链路代价。
本发明实施例还提供一种计算机可读存储介质，所述计算机可读存储介质中存储有计算机可执行指令，所述计算机可执行指令被执行时实现虚拟映射方法。
本领域普通技术人员可以理解上述方法中的全部或部分步骤可通过程序来指令相关硬件(例如处理器)完成，所述程序可以存储于计算机可读存储介质中，如只读存储器、磁盘或光盘等。可选地，上述实施例的全部或部分
步骤也可以使用一个或多个集成电路来实现。相应地，上述实施例中的各模块/单元可以采用硬件的形式实现，例如通过集成电路来实现其相应功能，也可以采用软件功能模块的形式实现，例如通过处理器执行存储于存储器中的程序/指令来实现其相应功能。本申请不限制于任何特定形式的硬件和软件的结合。本领域的普通技术人员应当理解，可以对本申请的技术方案进行修改或者等同替换，而不脱离本申请技术方案的精神和范围，均应涵盖在本申请的权利要求范围当中。
上述技术方法增加了虚拟网的接受率，降低虚拟网的链路代价。
Claims (10)
- 一种虚拟映射方法，包括：计算物理拓扑网中每个物理节点和虚拟拓扑网中每个虚拟节点的资源值；以所述虚拟拓扑网中资源值最大的虚拟节点作为根节点，按照所述虚拟拓扑网资源值的降序排序依次构建广度优先搜索树的每一层待映射节点，以作为虚拟拓扑网中节点的映射顺序；选取所述物理拓扑网中资源值最大的前预定个未映射的节点作为当前虚拟节点的候选映射节点；计算所有候选映射节点的吸引力，并选取所有候选映射节点中吸引力最大的节点进行映射，以建立虚拟拓扑网与物理拓扑网的映射。
- 如权利要求1所述的虚拟映射方法，其中，以所述虚拟拓扑网中资源值最大的虚拟节点作为根节点，按照所述虚拟拓扑网资源值的降序排序依次构建广度优先搜索树的每一层待映射节点，以作为虚拟拓扑网中节点的映射顺序，包括：从所述虚拟拓扑网中选取资源值最大的虚拟节点作为根节点；按照所述虚拟拓扑网资源值的降序排序，从所述根节点开始依次构建广度优先搜索树的每一层待映射节点，以作为虚拟拓扑网中节点的映射顺序。
- 如权利要求1或2所述的虚拟映射方法，其中，计算所有候选映射节点的吸引力包括：根据双向广度搜索算法分别计算每个候选映射节点与相邻已映射的每个物理节点之间的吸引力；分别将每个候选映射节点与相邻已映射的每个物理结点之间的吸引力加权求和，作为每个候选映射节点的吸引力。
- 如权利要求3所述的虚拟映射方法，其中，根据双向广度搜索算法分别计算每个候选映射节点与相邻已映射的每个物理节点之间的吸引力，包括：根据双向广度搜索算法分别确定每个候选映射节点与相邻已映射的每个物理节点之间的最短路径；计算每条最短路径上节点之间的吸引力。
- 如权利要求4所述的虚拟映射方法，其中，计算所有候选映射节点的吸引力，包括：候选映射节点吸引力按照如下公式计算：
其中，表示虚拟拓扑网中待映射的虚拟节点；
表示
映射到物理拓扑网中的候选映射节点；
表示
已经映射过的相邻的虚拟节点，
表示节点
已经映射过的第i个相邻虚拟节点；
表示
映射到物理拓扑网中的节点，
表示
映射到物理拓扑网中的节点；
表示节点对
和
之间的吸引力，大小为
p表示物理拓扑网中
和
满足带宽
需求的最短路径；hop表示路径p上节点的跳数；Ave(p)表示路径p的平均带宽，大小为
l是路径p上的一条链路；
是虚拟拓扑网中相邻节点
和
的带宽。
- 一种虚拟映射装置，包括：计算模块，设置为计算物理拓扑网中每个物理节点和虚拟拓扑网中每个虚拟节点的资源值；构建模块，设置为以所述虚拟拓扑网中资源值最大的虚拟节点作为根节点，按照所述虚拟拓扑网资源值的降序排序依次构建广度优先搜索树的每一层待映射节点，以作为虚拟拓扑网中节点的映射顺序；选取模块，设置为选取所述物理拓扑网中资源值最大的前预定个未映射的节点作为当前虚拟节点的候选映射节点；映射模块，设置为计算所有候选映射节点的吸引力，并选取所有候选映射节点中吸引力最大的节点进行映射，以建立虚拟拓扑网与物理拓扑网的映射。
- 如权利要求6所述的虚拟映射装置，其中，所述构建模块包括：选取单元，设置为从所述虚拟拓扑网中选取资源值最大的虚拟节点作为 根节点；构建单元，设置为按照所述虚拟拓扑网资源值的降序排序，从所述根节点开始依次构建广度优先搜索树的每一层待映射节点，以作为虚拟拓扑网中节点的映射顺序。
- 如权利要求6或7所述的虚拟映射装置，其中，所述映射模块包括：计算单元，设置为根据双向广度搜索算法计算每个候选映射节点与相邻已映射的各物理结点之间的吸引力；确定单元，设置为将每个候选映射节点与相邻已映射的各物理节点之间吸引力加权求和，作为每个候选映射节点的吸引力。
- 如权利要求8所述的虚拟映射装置，其中：所述计算单元是设置为根据双向广度搜索算法确定每个候选映射节点与相邻已映射的各物理节点之间的最短路径；计算每条最短路径上节点之间的吸引力。
- 如权利要求9所述的虚拟映射装置，其中，所述映射模块通过如下方式实现计算所有候选映射节点的吸引力：按照如下公式计算候选映射节点的吸引力：候选映射节点吸引力为：
其中，表示虚拟拓扑网中待映射的虚拟节点；
表示
映射到物理拓扑网中的候选映射节点；
表示
已经映射过的相邻的虚拟节点，其中
表示节点
已经映射过的第i个相邻虚拟节点；
表示
映射到物理拓扑网中的节点，其中
表示
映射到物理拓扑网中的节点；
表示节点对
和
之间的吸引力，大小为
p表示物理拓扑网中
和
满足带宽
需求的最短路径；hop表示路径p上节点的跳数；Ave(p)表示路径p的平均带宽，大小为
其中l是路径p上的一条链路；
是虚拟拓扑网中相邻节点
和
的带宽。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610014047.2 | 2016-01-08 | ||
| CN201610014047.2A CN106961343B (zh) | 2016-01-08 | 2016-01-08 | 一种虚拟映射方法及装置 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2017117951A1 true WO2017117951A1 (zh) | 2017-07-13 |
Family
ID=59273176
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2016/088908 WO2017117951A1 (zh) | 2016-01-08 | 2016-07-06 | 一种虚拟映射方法及装置 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN106961343B (zh) |
| WO (1) | WO2017117951A1 (zh) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107666448A (zh) * | 2017-09-30 | 2018-02-06 | 重庆邮电大学 | 一种时延感知下的5g虚拟接入网映射方法 |
| CN108471358A (zh) * | 2018-03-20 | 2018-08-31 | 河北工程大学 | 一种基于最小生成树的虚拟网络保护性映射方法 |
| CN110380906A (zh) * | 2019-07-25 | 2019-10-25 | 哈尔滨工业大学 | 一种大规模多维融合的虚拟网络映射方法 |
| CN111563293A (zh) * | 2020-04-17 | 2020-08-21 | 上海外高桥造船有限公司 | 船舶电缆路径节点的拓扑网络的自动生成方法 |
| CN113193984A (zh) * | 2021-03-31 | 2021-07-30 | 西安交通大学 | 一种空天地一体化网络资源映射方法及系统 |
| CN113300861A (zh) * | 2020-02-24 | 2021-08-24 | 中国电信股份有限公司 | 网络切片配置方法、装置以及存储介质 |
| CN113630328A (zh) * | 2021-07-09 | 2021-11-09 | 西安电子科技大学 | 一种数据中心虚拟网络映射方法及系统 |
| CN114422375A (zh) * | 2021-12-10 | 2022-04-29 | 国网河南省电力公司 | 虚拟网资源分配方法、设备、介质及程序产品 |
| CN114697221A (zh) * | 2022-03-22 | 2022-07-01 | 西安交通大学 | 空天地一体化网络中基于快照树模型的可靠性映射算法 |
| CN114745274A (zh) * | 2022-04-15 | 2022-07-12 | 烽火通信科技股份有限公司 | 一种基于灵活栅格光网络实现虚拟网络映射的方法及装置 |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110661704B (zh) * | 2018-06-30 | 2021-10-26 | 华为技术有限公司 | 转发路径的计算方法及sdn控制器 |
| CN110311862B (zh) * | 2019-06-14 | 2020-09-25 | 北京邮电大学 | 一种服务链映射方法及装置 |
| CN111752707A (zh) * | 2020-06-09 | 2020-10-09 | 深圳供电局有限公司 | 一种基于映射关系的电力通信网资源分配方法 |
| CN115862417B (zh) * | 2023-02-28 | 2023-05-12 | 中电运行(北京)信息技术有限公司 | 一种集成攻防演练学习的虚拟仿真系统及仿真方法 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103179052A (zh) * | 2011-12-20 | 2013-06-26 | 中国科学院声学研究所 | 一种基于接近度中心性的虚拟资源分配方法及系统 |
| CN103457752A (zh) * | 2012-05-30 | 2013-12-18 | 中国科学院声学研究所 | 一种虚拟网络映射方法 |
| CN104320276A (zh) * | 2014-10-28 | 2015-01-28 | 北京邮电大学 | 一种基于割集的虚拟网络映射方法及其系统 |
| CN104917659A (zh) * | 2015-06-02 | 2015-09-16 | 浙江大学 | 一种基于虚拟网络连接特性的虚拟网络映射方法 |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8582423B2 (en) * | 2010-08-04 | 2013-11-12 | Alcatel Lucent | Multi-chassis inter-process communication |
| CN102929989B (zh) * | 2012-10-19 | 2016-01-20 | 南京邮电大学 | 一种地理空间数据在云计算平台上的负载均衡方法 |
| CN103457852B (zh) * | 2013-09-13 | 2016-04-20 | 电子科技大学 | 一种多播虚拟网络的抗毁性映射方法 |
-
2016
- 2016-01-08 CN CN201610014047.2A patent/CN106961343B/zh active Active
- 2016-07-06 WO PCT/CN2016/088908 patent/WO2017117951A1/zh active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103179052A (zh) * | 2011-12-20 | 2013-06-26 | 中国科学院声学研究所 | 一种基于接近度中心性的虚拟资源分配方法及系统 |
| CN103457752A (zh) * | 2012-05-30 | 2013-12-18 | 中国科学院声学研究所 | 一种虚拟网络映射方法 |
| CN104320276A (zh) * | 2014-10-28 | 2015-01-28 | 北京邮电大学 | 一种基于割集的虚拟网络映射方法及其系统 |
| CN104917659A (zh) * | 2015-06-02 | 2015-09-16 | 浙江大学 | 一种基于虚拟网络连接特性的虚拟网络映射方法 |
Non-Patent Citations (1)
| Title |
|---|
| CHENG, XIANG ET AL.: "Virtual Network Embedding Through Topology-Aware Node Ranking", ACM SIGCOMM COMPUTER COMMUNICATION REVIEW, vol. 41, no. 2, 30 April 2011 (2011-04-30), pages 38 - 47, XP058000077 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107666448B (zh) * | 2017-09-30 | 2020-09-15 | 重庆邮电大学 | 一种时延感知下的5g虚拟接入网映射方法 |
| CN107666448A (zh) * | 2017-09-30 | 2018-02-06 | 重庆邮电大学 | 一种时延感知下的5g虚拟接入网映射方法 |
| CN108471358A (zh) * | 2018-03-20 | 2018-08-31 | 河北工程大学 | 一种基于最小生成树的虚拟网络保护性映射方法 |
| CN108471358B (zh) * | 2018-03-20 | 2021-02-12 | 河北工程大学 | 一种基于最小生成树的虚拟网络保护性映射方法 |
| CN110380906A (zh) * | 2019-07-25 | 2019-10-25 | 哈尔滨工业大学 | 一种大规模多维融合的虚拟网络映射方法 |
| CN110380906B (zh) * | 2019-07-25 | 2022-03-04 | 哈尔滨工业大学 | 一种大规模多维融合的虚拟网络映射方法 |
| CN113300861A (zh) * | 2020-02-24 | 2021-08-24 | 中国电信股份有限公司 | 网络切片配置方法、装置以及存储介质 |
| CN113300861B (zh) * | 2020-02-24 | 2022-05-03 | 中国电信股份有限公司 | 网络切片配置方法、装置以及存储介质 |
| CN111563293A (zh) * | 2020-04-17 | 2020-08-21 | 上海外高桥造船有限公司 | 船舶电缆路径节点的拓扑网络的自动生成方法 |
| CN111563293B (zh) * | 2020-04-17 | 2023-03-14 | 上海外高桥造船有限公司 | 船舶电缆路径节点的拓扑网络的自动生成方法 |
| CN113193984B (zh) * | 2021-03-31 | 2022-12-09 | 西安交通大学 | 一种空天地一体化网络资源映射方法及系统 |
| CN113193984A (zh) * | 2021-03-31 | 2021-07-30 | 西安交通大学 | 一种空天地一体化网络资源映射方法及系统 |
| CN113630328A (zh) * | 2021-07-09 | 2021-11-09 | 西安电子科技大学 | 一种数据中心虚拟网络映射方法及系统 |
| CN114422375A (zh) * | 2021-12-10 | 2022-04-29 | 国网河南省电力公司 | 虚拟网资源分配方法、设备、介质及程序产品 |
| CN114422375B (zh) * | 2021-12-10 | 2024-04-30 | 国网河南省电力公司 | 虚拟网资源分配方法、设备、介质及程序产品 |
| CN114697221A (zh) * | 2022-03-22 | 2022-07-01 | 西安交通大学 | 空天地一体化网络中基于快照树模型的可靠性映射算法 |
| CN114745274A (zh) * | 2022-04-15 | 2022-07-12 | 烽火通信科技股份有限公司 | 一种基于灵活栅格光网络实现虚拟网络映射的方法及装置 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106961343B (zh) | 2020-07-17 |
| CN106961343A (zh) | 2017-07-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2017117951A1 (zh) | 一种虚拟映射方法及装置 | |
| CN109495300B (zh) | 一种可靠的sdn虚拟网络映射方法 | |
| CN112738820A (zh) | 一种服务功能链的动态部署方法、装置及计算机设备 | |
| CN108667657B (zh) | 一种面向sdn的基于局部特征信息的虚拟网络映射方法 | |
| US9892532B2 (en) | Apparatus and method for generating a shortest-path tree in a graph | |
| CN108833271B (zh) | 一种电网广域控制业务通信路径选择方法及服务器 | |
| CN108718246B (zh) | 一种面向网络功能虚拟化的资源调度方法和系统 | |
| CN104935476B (zh) | 一种基于sdn的网络流量矩阵测量方法 | |
| CN104917659B (zh) | 一种基于虚拟网络连接特性的虚拟网络映射方法 | |
| CN112560204B (zh) | 一种基于lstm深度学习的光网络路由优化方法及其相关装置 | |
| CN107786449A (zh) | 基于fsr协议的路径选择方法、装置、服务器和存储介质 | |
| CN105591876B (zh) | 一种虚拟网络映射方法 | |
| CN113300861B (zh) | 网络切片配置方法、装置以及存储介质 | |
| CN111181792A (zh) | 基于网络拓扑的sdn控制器部署方法、装置及电子设备 | |
| Nguyen et al. | Rethinking virtual link mapping in network virtualization | |
| CN108243066B (zh) | 低延迟的网络服务请求部署方法 | |
| CN106681795B (zh) | 节点局部拓扑和可利用资源容量的虚拟网络映射方法 | |
| CN108924053B (zh) | 用于多路径路由的多条部分不相交最短路径快速寻找方法 | |
| CN110958192A (zh) | 一种基于虚拟交换机的虚拟数据中心资源分配系统及方法 | |
| JP5049316B2 (ja) | ネットワークトポロジ設計装置、ネットワークトポロジ設計方法、およびプログラム | |
| JP4408000B2 (ja) | 経路探索方法および装置 | |
| JP2010193224A (ja) | ネットワーク設計装置、ネットワーク設計方法、およびプログラム | |
| CN110213149B (zh) | 一种基于节点熵的虚拟网络映射方法 | |
| JP5595342B2 (ja) | 複数経路探索方法及び装置 | |
| CN107995109B (zh) | 路由方法及路由设备 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16883086 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 16883086 Country of ref document: EP Kind code of ref document: A1 |