CN113055218B - Redundancy evaluation method and device for NFV network and computing equipment - Google Patents
Redundancy evaluation method and device for NFV network and computing equipment Download PDFInfo
- Publication number
- CN113055218B CN113055218B CN201911385566.XA CN201911385566A CN113055218B CN 113055218 B CN113055218 B CN 113055218B CN 201911385566 A CN201911385566 A CN 201911385566A CN 113055218 B CN113055218 B CN 113055218B
- Authority
- CN
- China
- Prior art keywords
- sample
- information
- network
- attribute
- node
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
- H04L41/14—Network analysis or design
- H04L41/142—Network analysis or design using statistical or mathematical methods
-
- 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
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- 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
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Algebra (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Probability & Statistics with Applications (AREA)
- Pure & Applied Mathematics (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The embodiment of the invention relates to the technical field of communication, and discloses a method and a device for evaluating redundancy of an NFV network and computing equipment. Wherein, the method comprises the following steps: acquiring network information to be detected, wherein the network information to be detected comprises topological structure information and attribute information; determining the nodes of the network to be detected and the connection relation among the nodes according to the topological structure information; generating an adjacency matrix according to the connection relation between the nodes; determining an attribute characteristic value of each node according to the attribute information; generating a feature matrix according to the attribute feature value of each node; and inputting the adjacency matrix and the characteristic matrix into a redundancy detection model, and obtaining a redundancy detection result output by the redundancy detection model. Through the mode, the embodiment of the invention can conveniently and quickly detect the redundancy of the NFV network.
Description
Technical Field
The embodiment of the invention relates to the technical field of communication, in particular to a method and a device for evaluating redundancy of an NFV network and computing equipment.
Background
Network Function Virtualization (NFV) refers to implementing various Network device Functions on standardized general-purpose IT devices (such as x86 servers, storage and switching devices) by using Virtualization technology. Generally, an NFV network includes a hardware layer, a virtual network element, an NFV management and orchestration system.
At present, the redundancy evaluation of the NFV network mainly depends on expert experience, but because the NFV network has a complex and various structure, the method of manual evaluation is very troublesome and slow.
Disclosure of Invention
In view of the foregoing problems, embodiments of the present invention provide a method, an apparatus, and a computing device for evaluating redundancy of an NFV network, which can conveniently and quickly detect redundancy of the NFV network.
According to an aspect of the embodiments of the present invention, there is provided a method for evaluating redundancy of an NFV network, the method including: acquiring network information to be detected, wherein the network information to be detected comprises topological structure information and attribute information; determining the nodes of the network to be detected and the connection relation among the nodes according to the topological structure information; generating an adjacency matrix according to the connection relation between the nodes; determining an attribute characteristic value of each node according to the attribute information; generating a feature matrix according to the attribute feature value of each node; and inputting the adjacency matrix and the characteristic matrix into a redundancy detection model, and obtaining a redundancy detection result output by the redundancy detection model.
In an optional manner, the generating an adjacency matrix according to a connection relationship between the nodes further includes:
generating the adjacency matrix according to the following formula:
wherein A is the adjacency matrixN is the total number of said nodes, element V ij Used for representing the connection relationship between the ith node and the jth node, if the ith node and the jth node have connection, V ij Is 1, if there is no connection between the ith node and the jth node, then V ij Is 0, wherein i, j and N are integers.
In an alternative mode, the attribute information includes hierarchy information and device category information;
determining an attribute feature value of each node according to the attribute information, further comprising: generating a first attribute code according to the hierarchy information; generating a second attribute code according to the equipment type information; and taking the first attribute code and the second attribute code corresponding to the node as attribute characteristic values of the node.
In an optional manner, the generating a feature matrix according to the attribute feature value of each node further includes:
generating the feature matrix according to the following formula:
wherein X is the feature matrix, N is the total number of the nodes, and element X i1 For representing a first attribute code corresponding to the ith node, element x i2 And the attribute code is used for representing a second attribute code corresponding to the ith node, wherein i and N are integers.
In an optional manner, the method further comprises: obtaining sample network information, wherein the sample network information comprises sample topological structure information and sample attribute information; acquiring sample detection information corresponding to the sample network information; determining sample nodes of the sample network information and a connection relation between the sample nodes according to the sample topological structure information; generating a sample adjacency matrix according to the connection relation between the sample nodes; determining a sample attribute characteristic value of each sample node according to the sample attribute information; generating a sample characteristic matrix according to the sample attribute characteristic value of each sample node; training a preset graph convolution neural network model according to the sample adjacency matrix, the sample characteristic matrix and sample detection information corresponding to the sample network information; and taking the trained preset graph convolution neural network model as the redundancy detection model.
In an alternative mode, the preset graph convolutional neural network model comprises an encoder and a predictor; the training of the preset graph convolution neural network model according to the sample adjacency matrix, the sample characteristic matrix and the sample detection information corresponding to the sample network information further comprises: inputting the sample adjacency matrix, the sample feature matrix, into the encoder, such that the encoder outputs a potential spatial vector representation of the sample network; inputting the potential space vector representation of the sample network into the predictor, and acquiring training detection information output by the predictor; and adjusting the preset graph convolutional neural network model according to the training detection information and the sample detection information corresponding to the sample network information.
In an optional manner, the method further comprises: and if the value of the redundancy detection result is 1, determining that the redundancy detection of the network to be detected passes.
According to another aspect of the embodiments of the present invention, there is provided an apparatus for evaluating redundancy of an NFV network, the apparatus including: the system comprises a detection information acquisition module, a detection information acquisition module and a detection information processing module, wherein the detection information acquisition module is used for acquiring network information to be detected, and the network information to be detected comprises topological structure information and attribute information; a connection relation determining module, configured to determine, according to the topology structure information, a node of the to-be-detected network and a connection relation between the nodes; the adjacency matrix generation module is used for generating an adjacency matrix according to the connection relation among the nodes; an attribute characteristic value determining module, configured to determine an attribute characteristic value of each node according to the attribute information; the characteristic matrix generation module is used for generating a characteristic matrix according to the attribute characteristic value of each node; and the detection module is used for inputting the adjacency matrix and the characteristic matrix into a redundancy detection model and acquiring a redundancy detection result output by the redundancy detection model.
According to still another aspect of an embodiment of the present invention, there is provided a computing device including: a processor, a memory, and a communication interface, the processor, the memory, and the communication interface in communication with each other; the memory is configured to store at least one executable instruction that causes the processor to perform the operations of the method for evaluating redundancy of an NFV network as described above.
According to another aspect of the embodiments of the present invention, there is provided a computer-readable storage medium having at least one executable instruction stored therein, the executable instruction causing a processor to execute the method for evaluating the redundancy of the NFV network as described above.
According to the embodiment of the invention, the network information to be detected is obtained, the network information to be detected comprises topological structure information and attribute information, the nodes of the network to be detected and the connection relation among the nodes are determined according to the topological structure information, the adjacency matrix is generated according to the connection relation among the nodes, the attribute characteristic value of each node is determined according to the attribute information, the characteristic matrix is generated according to the attribute characteristic value of each node, the adjacency matrix and the characteristic matrix are input into the redundancy detection model, and the redundancy detection result output by the redundancy detection model is obtained, so that the problems that the existing NFV network is complex and various and depends on manual detection are solved, and the redundancy of the NFV network can be conveniently and quickly detected.
The foregoing description is only an overview of the technical solutions of the embodiments of the present invention, and the embodiments of the present invention can be implemented according to the content of the description in order to make the technical means of the embodiments of the present invention more clearly understood, and the detailed description of the present invention is provided below in order to make the foregoing and other objects, features, and advantages of the embodiments of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
fig. 1a and fig. 1b are schematic structural diagrams of an NFV network according to an embodiment of the present invention;
fig. 2 is a flowchart illustrating a redundancy evaluation method for an NFV network according to an embodiment of the present invention;
fig. 3 is a flowchart illustrating a redundancy evaluation method for an NFV network according to another embodiment of the present invention;
FIG. 4 is a schematic structural diagram of a preset-map convolutional neural network model according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram illustrating a redundancy evaluation apparatus of an NFV network according to an embodiment of the present invention;
fig. 6 shows a schematic structural diagram of a computing device provided by an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
Generally, an NFV network includes a hardware layer, a virtual network element, an NFV management and orchestration system. The network structure of the NFV site is designed according to a hierarchical architecture, and a telecom cloud hardware resource pool is divided into an access layer, a core layer and an exit layer from bottom to top. Wherein, the access layer is provided with TOR (Top of Rack) and various computing servers and storage devices. The core layer is deployed with an EOR (end of row) router and an EOR pairing router, and is configured with devices such as an inner Firewall, an Intrusion Detection System (IDS), an Intrusion Prevention System (IPS), a website Application protection System (WAF), a load balancer and the like according to requirements, the EOR downward converges all access layer switching devices in a switching network, the EOR performs two-layer convergence, and the EOR pairing router serves as a three-layer gateway and is interconnected with an exit routing device upward. The exit layer multiplexes the existing network bearing and transmission network networking equipment. The exit layer is used as a link for interconnection and intercommunication between the internal network and the external network of the telecommunication cloud station, high-speed interconnection with external equipment is externally completed, interconnection with core layer exchange equipment of the telecommunication cloud station is internally responsible, and forwarding and maintenance of routing information outside the station are completed.
For example, as shown in fig. 1a and 1b, the NFV network can be divided into two scenarios according to security requirements and connection requirements. In fig. 1a, a hardware management TOR connects all network devices in a computing, storage, access layer and core layer in a resource pool, and exit layer devices such as a CE, a network access router, and an EPC firewall; in fig. 1b, the hardware management manages all the network devices and inner firewalls of the computation, storage, access layer and core layer in the resource pool, and the exit layer devices and security devices such as CE, network access router, EPC firewall, etc., and the traffic, storage, management TOR accessed by the computation server and storage server of the DMZ zone (the space between two firewalls is called DMZ) are all set independently.
The redundancy refers to repeated establishment, the networking redundancy refers to establishment of a standby network and related hardware, so that network load is balanced, and when a main network does not operate, the standby network can immediately take over the work of the main network. In the process of networking planning and design, a single point of failure easily occurs when the network structure is unreasonable, that is, a certain node fails to cause the whole network to be broken down. At present, the redundancy evaluation of the NFV network mainly depends on expert experience, but because the NFV network has a complex and various structure, the method of manual evaluation is very troublesome and slow.
Based on this, the embodiment of the invention provides a method, a device and a computing device for evaluating the redundancy of an NFV network, which can conveniently and quickly detect the redundancy of the NFV network.
Specifically, the embodiments of the present invention will be further explained below with reference to the drawings.
It should be understood that the following examples are provided by way of illustration and are not intended to limit the invention in any way to the particular embodiment disclosed.
Fig. 2 is a flowchart illustrating a redundancy evaluation method for an NFV network according to an embodiment of the present invention. The method is applied to a computing device. As shown in fig. 2, the method includes:
The network information to be detected comprises topological structure information and attribute information. The topological structure information refers to the connection relation of each device in the network; the attribute information includes hierarchy information and device type information, the hierarchy information refers to a hierarchy to which the device belongs, and may include, for example, an egress layer, a core layer, and an access layer, and the device type information refers to a type to which the device belongs, and may include, for example, a switch, a router, a firewall, a load balancer, a server, a TOR, an EOR, and the like. Of course, in some other embodiments, the attribute information may further include other characteristic information, which may be set according to actual situations.
The network information to be detected is acquired, the network information to be detected can be manually input, or a structure diagram of the network to be detected is acquired, and the structure diagram of the network to be detected is processed and identified, so that topological structure information and attribute information are acquired.
And step 120, determining the nodes of the network to be detected and the connection relation among the nodes according to the topological structure information.
Wherein, according to the topology structure information, determining the node of the network to be detected may be: determining that one device in the network to be detected is a node, or determining that the same type of device in the same hierarchy in the network to be detected is a node.
Determining the connection relationship among the nodes according to the topological structure information, wherein the determining the connection relationship among the nodes comprises the following steps: and determining the connection relation between each node and other nodes and the connection relation between each node and the node. The connection relationship between each node and itself may be: interconnection among similar devices at the same level, or connection of a single device to itself.
And 130, generating an adjacency matrix according to the connection relation among the nodes.
Wherein, according to the connection relation between each node, generating an adjacency matrix, further comprising: the adjacency matrix is generated according to the following formula:
where A is the adjacency matrix, N is the total number of nodes, and element V ij Used for representing the connection relationship between the ith node and the jth node, if the ith node and the jth node have connection, V ij Is 1, if there is no connection between the ith node and the jth node, then V ij Is 0, wherein i, j and N are integers.
For example, assume that there is device A, B, device A is connected to itself, device B is connected to device A, and device A is denoted as V 1 And the device B is V 2 Then the adjacency matrix is
And step 140, determining the attribute characteristic value of each node according to the attribute information.
Wherein, the attribute information includes hierarchy information and device type information, step 140 specifically includes:
step 141, generating a first attribute code according to the hierarchy information;
step 142, generating a second attribute code according to the equipment type information;
and 143, taking the first attribute code and the second attribute code corresponding to the node as the attribute characteristic value of the node.
Wherein, according to the hierarchy information, generating a first attribute code may be: the codes corresponding to the exit layer, the core layer and the access layer are preset to be 01, 02 and 03 respectively, when the hierarchy information of the node is located in the exit layer, the first attribute code of the node is 01, when the hierarchy information of the node is located in the core layer, the first attribute code of the node is 02, and when the hierarchy information of the node is located in the access layer, the first attribute code of the node is 03. Of course, in some other embodiments, the preset hierarchy and the corresponding relationship of the codes can be freely set according to the actual use.
The generating of the second attribute code according to the device type information may be: the method includes the steps that codes corresponding to a switch, a router, a firewall, a load balancer, a server, a TOR and an EOR are respectively 11, 12, 13, 14, 15, 16 and 17, when a node is the switch, the second attribute code of the node is 11, when the node is the router, the second attribute code of the node is 12, when the node is the firewall, the second attribute code of the node is 13, when the node is the load balancer, the second attribute code of the node is 14, when the node is the server, the second attribute code of the node is 15, when the node is the TOR, the second attribute code of the node is 16, and when the node is the EOR, the second attribute code of the node is 17. Of course, in some other embodiments, the preset correspondence between the device type and the code may be freely set according to actual use.
In step 143, after determining the first attribute code and the second attribute code corresponding to a certain node, the attribute feature value of the node is determined.
And 150, generating a characteristic matrix according to the attribute characteristic value of each node.
Wherein, according to the attribute eigenvalue of each node, generating a characteristic matrix, further comprising: the feature matrix is generated according to the following formula:
wherein X is a feature matrix, N is the total number of nodes, and X is an element i1 For representing a first attribute code corresponding to the ith node, element x i2 And the attribute code is used for representing a second attribute code corresponding to the ith node, wherein i and N are integers.
For example, assuming there is A, B, device A is TOR and is located in the Access stratum, and device B is EOR and is located in the core layer, the feature matrix is
In this embodiment, the attribute feature value includes a first attribute code and a second attribute code, and the feature matrix is an N × 2 matrix. In some other embodiments, when the attribute information further includes some other features, and the corresponding attribute feature values include m attribute codes, the feature matrix is an N × m matrix.
And 160, inputting the adjacency matrix and the characteristic matrix into the redundancy detection model to obtain a redundancy detection result output by the redundancy detection model.
The redundancy detection model can be used for outputting a redundancy detection result according to the adjacency matrix and the feature matrix. The redundancy detection model may be trained by a Graph Convolutional neural network (GCN). The purpose of GCN is to extract the spatial features of the topology, the goal being to learn a mapping of signals or features on graph G ═ V, E, with inputs comprising adjacency matrix a and feature matrix X, the model will produce a node-level output or graph-level output.
Wherein each neural network layer can be written as a non-linear function:
H (l+1) ＝f(H (l) ,A)
wherein H (0) X is input data, H (l) Z is the output data and L is the number of layers of the neural network, and different models are determined by selecting different f-functions and parameters. For example,
wherein, W (l) Is a parameter matrix of the 1 st neural network layer, sigma is a nonlinear activation function (e.g., ReLU), a is a adjacency matrix, and D is a node-degree diagonal matrix of a.
In step 160, after the adjacency matrix and the feature matrix of the network to be detected are obtained, the adjacency matrix and the feature matrix are input into the redundancy detection model, and the redundancy detection model outputs the redundancy detection result, so that the redundancy evaluation of the network to be detected can be performed according to the redundancy detection result.
The redundancy detection result may be a probability value or 0 or 1. The method may further comprise: if the value of the redundancy detection result is 1, determining that the redundancy detection of the network to be detected passes; and if the value of the redundancy detection result is 0, determining that the redundancy detection of the network to be detected does not pass.
According to the embodiment of the invention, the network information to be detected is obtained, the network information to be detected comprises topological structure information and attribute information, the nodes of the network to be detected and the connection relation among the nodes are determined according to the topological structure information, the adjacency matrix is generated according to the connection relation among the nodes, the attribute characteristic value of each node is determined according to the attribute information, the characteristic matrix is generated according to the attribute characteristic value of each node, the adjacency matrix and the characteristic matrix are input into the redundancy detection model, and the redundancy detection result output by the redundancy detection model is obtained, so that the problems that the existing NFV network is complex and various and depends on manual detection are solved, and the redundancy of the NFV network can be conveniently and quickly detected.
Fig. 3 is a flowchart illustrating a redundancy evaluation method for an NFV network according to another embodiment of the present invention. The method is applied to a computing device. As shown in fig. 3, the difference from the above embodiment is that the method further includes:
The sample network may be an NFV network that has been subjected to manual redundancy evaluation. The sample network information comprises sample topological structure information and sample attribute information, wherein the sample topological structure information is the connection relation of each device in the sample network, and the sample attribute information comprises sample hierarchy information and sample device type information.
And step 220, obtaining sample detection information corresponding to the sample network information.
The sample detection information is an evaluation result obtained by evaluating the redundancy of the sample network. The number of the sample networks can be several, and each sample network has corresponding sample detection information.
And step 230, determining the sample nodes of the sample network information and the connection relation among the sample nodes according to the sample topological structure information.
And 240, generating a sample adjacency matrix according to the connection relation among the sample nodes.
And step 250, determining the sample attribute characteristic value of each sample node according to the sample attribute information.
And step 260, generating a sample characteristic matrix according to the sample attribute characteristic value of each sample node.
And 270, training a preset graph convolution neural network model according to the sample adjacency matrix, the sample characteristic matrix and the sample detection information corresponding to the sample network information.
The preset graph convolutional neural network model comprises an encoder and a predictor. Specifically, step 270 includes:
step 271, inputting the sample adjacency matrix and the sample feature matrix into an encoder, so that the encoder outputs a potential space vector representation of the sample network;
step 272, representing the potential space vector of the sample network to be input into the predictor, and acquiring training detection information output by the predictor;
and 273, adjusting the preset graph convolution neural network model according to the training detection information and the sample detection information corresponding to the sample network information.
As shown in fig. 4, the encoder includes an input layer and three Graph convolution layers (Graph Conv), specifically: the first layer is an input layer and is used for inputting a sample adjacency matrix A and a sample characteristic matrix X; the second layer is a graph convolution layer, the number of convolution kernels is 64, and an activation function is set to be ReLU; the third layer is a graph convolution layer, the number of convolution kernels is 64, and an activation function is set to be ReLU; the fourth layer is a graph convolution layer, the number of convolution kernels is 64, the activation function is set to "lambda", and the potential space vector representation Z for the output sample network is Z, where Z is GCN (X, a). The predictor comprises three full-connection layers (Dense) and two drop layers (dropout), and specifically comprises the following steps: the fifth layer is a full connection layer, the number of neurons is set to 64, and an activation function is set to 'ReLU'; the sixth layer is a discarding layer, the discarding probability is set to be 0.5, and when parameters are updated each time in the training process, input neurons are randomly disconnected according to a certain probability (50%), so that overfitting is prevented; the seventh layer is a full connection layer, the number of neurons is set to be 32, and an activation function is set to be 'ReLU'; the eighth layer is a discarding layer, and the discarding probability is set to be 0.5; the ninth layer is a full connection layer, the number of the neurons is set to be 1, and the activation function is set to be sigmoid and used for outputting a networking redundancy detection result. Wherein, the objective function may select a second class logarithmic loss function "binary _ cross", the training round number is set to 2000(epochs is 2000), and the gradient descent optimization algorithm may select an adam optimizer for improving the learning speed of the conventional gradient descent (optimizer is 'adam'). The neural network can find the optimal weight value which enables the target function to be the lowest through gradient descent, and the neural network can learn the weight value automatically through training. Training is carried out through a training set (namely sample network information), so that the smaller the objective function is, the better the objective function is, a verification model (namely a preset graph convolution neural network model) is evaluated by a test set (namely sample detection information corresponding to the sample network information) after each round of training, and the weight of the model, namely the optimal weight value, is derived after the model converges.
And step 280, taking the trained preset graph convolutional neural network model as a redundancy detection model.
The embodiment of the invention obtains sample detection information corresponding to the sample network information by obtaining the sample network information, wherein the sample network information comprises sample topological structure information and sample attribute information, determines sample nodes of the sample network information and the connection relation among the sample nodes according to the sample topological structure information, generates a sample adjacency matrix according to the connection relation among the sample nodes, determines sample attribute characteristic values of the sample nodes according to the sample attribute information, generates a sample characteristic matrix according to the sample attribute characteristic values of the sample nodes, trains a preset graph convolution neural network model according to the sample adjacency matrix, the sample characteristic matrix and the sample detection information corresponding to the sample network information, takes the trained preset graph convolution neural network model as a redundancy detection model, and obtains the redundancy detection model by training the preset graph convolution neural network model, therefore, the redundancy of the network information to be detected can be detected through the redundancy detection model, and the redundancy of the NFV network can be conveniently and quickly detected.
Fig. 5 is a schematic structural diagram illustrating a redundancy evaluation apparatus for an NFV network according to an embodiment of the present invention. As shown in fig. 5, the apparatus 300 includes: the detection method comprises a detection information acquisition module 310, a connection relation determination module 320, an adjacency matrix generation module 330, an attribute characteristic value determination module 340, a characteristic matrix generation module 350 and a detection module 360.
The detection information acquiring module 310 is configured to acquire network information to be detected, where the network information to be detected includes topology structure information and attribute information; the connection relationship determining module 320 is configured to determine the nodes of the network to be detected and the connection relationship between the nodes according to the topology structure information; the adjacency matrix generating module 330 is configured to generate an adjacency matrix according to a connection relationship between the nodes; the attribute feature value determining module 340 is configured to determine an attribute feature value of each node according to the attribute information; the feature matrix generation module 350 is configured to generate a feature matrix according to the attribute feature value of each node; the detection module 360 is configured to input the adjacency matrix and the feature matrix into a redundancy detection model, and obtain a redundancy detection result output by the redundancy detection model.
In an optional manner, the adjacency matrix generation module 330 is specifically configured to: generating the adjacency matrix according to the following formula:
wherein A is the adjacency matrix, N is the total number of the nodes, and element V ij Used for representing the connection relationship between the ith node and the jth node, if the ith node and the jth node have connection, V ij Is 1, if there is no connection between the ith node and the jth node, then V ij Is 0, wherein i, j and N are integers.
In an alternative mode, the attribute information includes hierarchy information and device category information; the attribute feature value determination module 340 is specifically configured to: generating a first attribute code according to the hierarchy information; generating a second attribute code according to the equipment type information; and taking the first attribute code and the second attribute code corresponding to the node as attribute characteristic values of the node.
In an optional manner, the feature matrix generation module 350 is specifically configured to: generating the feature matrix according to the following formula:
wherein X is the feature matrix, N is the total number of the nodes, and element X i1 For representing a first attribute code corresponding to the ith node, element x i2 And the attribute code is used for representing a second attribute code corresponding to the ith node, wherein i and N are integers.
In an optional manner, the apparatus 300 further comprises: and a model training module. The model training module is specifically configured to: obtaining sample network information, wherein the sample network information comprises sample topological structure information and sample attribute information; acquiring sample detection information corresponding to the sample network information; determining sample nodes of the sample network information and a connection relation between the sample nodes according to the sample topological structure information; generating a sample adjacency matrix according to the connection relation between the sample nodes; determining a sample attribute characteristic value of each sample node according to the sample attribute information; generating a sample characteristic matrix according to the sample attribute characteristic value of each sample node; training a preset graph convolution neural network model according to the sample adjacency matrix, the sample characteristic matrix and sample detection information corresponding to the sample network information; and taking the trained preset graph convolution neural network model as the redundancy detection model.
In an alternative mode, the preset graph convolutional neural network model comprises an encoder and a predictor; the model training module specifically comprises: inputting the sample adjacency matrix, the sample feature matrix, into the encoder, such that the encoder outputs a potential spatial vector representation of the sample network; inputting the potential space vector representation of the sample network into the predictor, and acquiring training detection information output by the predictor; and adjusting the preset graph convolutional neural network model according to the training detection information and the sample detection information corresponding to the sample network information.
In an optional manner, the apparatus 300 further comprises: and determining a module. The determination module is to: and if the value of the redundancy detection result is 1, determining that the redundancy detection of the network to be detected passes.
It should be noted that, the redundancy evaluation apparatus for an NFV network according to the embodiments of the present invention is an apparatus capable of executing the redundancy evaluation method for an NFV network, and all embodiments of the redundancy evaluation method for an NFV network are applicable to the apparatus and can achieve the same or similar beneficial effects.
According to the embodiment of the invention, the network information to be detected is obtained, the network information to be detected comprises topological structure information and attribute information, the nodes of the network to be detected and the connection relation among the nodes are determined according to the topological structure information, the adjacency matrix is generated according to the connection relation among the nodes, the attribute characteristic value of each node is determined according to the attribute information, the characteristic matrix is generated according to the attribute characteristic value of each node, the adjacency matrix and the characteristic matrix are input into the redundancy detection model, and the redundancy detection result output by the redundancy detection model is obtained, so that the problems that the existing NFV network is complex and various and depends on manual detection are solved, and the redundancy of the NFV network can be conveniently and quickly detected.
An embodiment of the present invention provides a computer-readable storage medium, where at least one executable instruction is stored in the storage medium, and the executable instruction causes a processor to execute the method for evaluating the redundancy of the NFV network in any method embodiment described above.
According to the embodiment of the invention, the network information to be detected is obtained, the network information to be detected comprises topological structure information and attribute information, the nodes of the network to be detected and the connection relation among the nodes are determined according to the topological structure information, the adjacency matrix is generated according to the connection relation among the nodes, the attribute characteristic value of each node is determined according to the attribute information, the characteristic matrix is generated according to the attribute characteristic value of each node, the adjacency matrix and the characteristic matrix are input into the redundancy detection model, and the redundancy detection result output by the redundancy detection model is obtained, so that the problems that the existing NFV network is complex and various and depends on manual detection are solved, and the redundancy of the NFV network can be conveniently and quickly detected.
An embodiment of the present invention provides a computer program product, which includes a computer program stored on a computer storage medium, where the computer program includes program instructions, and when the program instructions are executed by a computer, the computer executes the redundancy evaluation method for the NFV network in any of the above method embodiments.
According to the embodiment of the invention, the network information to be detected is obtained, the network information to be detected comprises topological structure information and attribute information, the nodes of the network to be detected and the connection relation among the nodes are determined according to the topological structure information, the adjacency matrix is generated according to the connection relation among the nodes, the attribute characteristic value of each node is determined according to the attribute information, the characteristic matrix is generated according to the attribute characteristic value of each node, the adjacency matrix and the characteristic matrix are input into the redundancy detection model, and the redundancy detection result output by the redundancy detection model is obtained, so that the problems that the existing NFV network is complex and various and depends on manual detection are solved, and the redundancy of the NFV network can be conveniently and quickly detected.
Fig. 5 is a schematic structural diagram of a computing device according to an embodiment of the present invention, and the specific embodiment of the present invention does not limit the specific implementation of the computing device.
As shown in fig. 5, the computing device may include: a processor (processor)402, a Communications Interface 404, a memory 406, and a Communications bus 408.
Wherein: the processor 402, communication interface 404, and memory 406 communicate with each other via a communication bus 408. A communication interface 404 for communicating with network elements of other devices, such as clients or other servers. The processor 402 is configured to execute the program 410, and specifically may execute the method for evaluating the redundancy of the NFV network in any of the method embodiments described above.
In particular, program 410 may include program code comprising computer operating instructions.
The processor 402 may be a central processing unit CPU or an application Specific Integrated circuit asic or one or more Integrated circuits configured to implement embodiments of the present invention. The computing device includes one or more processors, which may be the same type of processor, such as one or more CPUs; or may be different types of processors such as one or more CPUs and one or more ASICs.
And a memory 406 for storing a program 410. Memory 406 may comprise high-speed RAM memory, and may also include non-volatile memory (non-volatile memory), such as at least one disk memory.
According to the embodiment of the invention, the network information to be detected is obtained, the network information to be detected comprises topological structure information and attribute information, the nodes of the network to be detected and the connection relation among the nodes are determined according to the topological structure information, the adjacency matrix is generated according to the connection relation among the nodes, the attribute characteristic value of each node is determined according to the attribute information, the characteristic matrix is generated according to the attribute characteristic value of each node, the adjacency matrix and the characteristic matrix are input into the redundancy detection model, and the redundancy detection result output by the redundancy detection model is obtained, so that the problems that the existing NFV network is complex and various and depends on manual detection are solved, and the redundancy of the NFV network can be conveniently and quickly detected.
The algorithms or displays presented herein are not inherently related to any particular computer, virtual system, or other apparatus. Various general purpose systems may also be used with the teachings herein. The required structure for constructing such a system will be apparent from the description above. In addition, embodiments of the present invention are not directed to any particular programming language. It is appreciated that a variety of programming languages may be used to implement the teachings of the present invention as described herein, and any descriptions of specific languages are provided above to disclose the best mode of the invention.
In the description provided herein, numerous specific details are set forth. It is understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the embodiments of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the invention and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be interpreted as reflecting an intention that: that the invention as claimed requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.
Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
Furthermore, those skilled in the art will appreciate that while some embodiments herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the following claims, any of the claimed embodiments may be used in any combination.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names. The steps in the above embodiments should not be construed as limiting the order of execution unless specified otherwise.
Claims (10)
1. A method for evaluating the redundancy of an NFV network, the method comprising:
acquiring network information to be detected, wherein the network information to be detected comprises topological structure information and attribute information;
determining the nodes of the network to be detected and the connection relation among the nodes according to the topological structure information;
generating an adjacency matrix according to the connection relation between the nodes;
determining an attribute characteristic value of each node according to the attribute information;
generating a feature matrix according to the attribute feature value of each node;
and inputting the adjacency matrix and the characteristic matrix into a redundancy detection model, and obtaining a redundancy detection result output by the redundancy detection model.
2. The method of claim 1, wherein generating an adjacency matrix according to the connection relationship between the nodes further comprises: generating the adjacency matrix according to the following formula:
wherein A is the adjacency matrix, N is the total number of the nodes, and element V ij Used for representing the connection relationship between the ith node and the jth node, if the ith node and the jth node have connection, V ij Is 1, if there is no connection between the ith node and the jth node, then V ij Is 0, wherein i, j and N are integers.
3. The method according to claim 1, wherein the attribute information includes hierarchy information and device category information;
determining an attribute feature value of each node according to the attribute information, further comprising:
generating a first attribute code according to the hierarchy information;
generating a second attribute code according to the equipment type information;
and taking the first attribute code and the second attribute code corresponding to the node as attribute characteristic values of the node.
4. The method of claim 3, wherein generating an eigen matrix based on the attribute eigenvalues of each of the nodes further comprises:
generating the feature matrix according to the following formula:
wherein X is the feature matrix, N is the total number of the nodes, and element X i1 For representing a first attribute code corresponding to the ith node, element x i2 And the attribute code is used for representing a second attribute code corresponding to the ith node, wherein i and N are integers.
5. The method of claim 1, further comprising:
obtaining sample network information, wherein the sample network information comprises sample topological structure information and sample attribute information;
acquiring sample detection information corresponding to the sample network information;
determining sample nodes of the sample network information and a connection relation between the sample nodes according to the sample topological structure information;
generating a sample adjacency matrix according to the connection relation between the sample nodes;
determining a sample attribute characteristic value of each sample node according to the sample attribute information;
generating a sample characteristic matrix according to the sample attribute characteristic value of each sample node;
training a preset graph convolution neural network model according to the sample adjacency matrix, the sample characteristic matrix and sample detection information corresponding to the sample network information;
and taking the trained preset graph convolution neural network model as the redundancy detection model.
6. The method of claim 5, wherein the predetermined graph convolutional neural network model comprises an encoder and a predictor;
the training of the preset graph convolution neural network model according to the sample adjacency matrix, the sample characteristic matrix and the sample detection information corresponding to the sample network information further comprises:
inputting the sample adjacency matrix, the sample feature matrix, into the encoder, such that the encoder outputs a potential spatial vector representation of the sample network;
inputting the potential space vector representation of the sample network into the predictor, and acquiring training detection information output by the predictor;
and adjusting the preset graph convolutional neural network model according to the training detection information and the sample detection information corresponding to the sample network information.
7. The method according to any one of claims 1-6, further comprising:
and if the value of the redundancy detection result is 1, determining that the redundancy detection of the network to be detected passes.
8. An apparatus for evaluating redundancy of an NFV network, the apparatus comprising:
the system comprises a detection information acquisition module, a detection information acquisition module and a detection information processing module, wherein the detection information acquisition module is used for acquiring network information to be detected, and the network information to be detected comprises topological structure information and attribute information;
a connection relation determining module, configured to determine, according to the topology structure information, a node of the to-be-detected network and a connection relation between the nodes;
the adjacency matrix generation module is used for generating an adjacency matrix according to the connection relation among the nodes;
an attribute characteristic value determining module, configured to determine an attribute characteristic value of each node according to the attribute information;
the characteristic matrix generation module is used for generating a characteristic matrix according to the attribute characteristic value of each node;
and the detection module is used for inputting the adjacency matrix and the characteristic matrix into a redundancy detection model and acquiring a redundancy detection result output by the redundancy detection model.
9. A computing device, comprising: the system comprises a processor, a memory and a communication interface, wherein the processor, the memory and the communication interface are communicated with each other;
the memory is configured to store at least one executable instruction that causes the processor to perform the operations of the method for evaluating redundancy of an NFV network according to any of claims 1 to 7.
10. A computer-readable storage medium having stored therein at least one executable instruction for causing a processor to execute the method for redundancy evaluation of an NFV network according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911385566.XA CN113055218B (en) | 2019-12-29 | 2019-12-29 | Redundancy evaluation method and device for NFV network and computing equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911385566.XA CN113055218B (en) | 2019-12-29 | 2019-12-29 | Redundancy evaluation method and device for NFV network and computing equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113055218A CN113055218A (en) | 2021-06-29 |
| CN113055218B true CN113055218B (en) | 2022-08-05 |
Family
ID=76507271
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911385566.XA Active CN113055218B (en) | 2019-12-29 | 2019-12-29 | Redundancy evaluation method and device for NFV network and computing equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113055218B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110650046A (en) * | 2019-09-24 | 2020-01-03 | 北京明略软件系统有限公司 | Network node importance scoring model training and importance detecting method and device |
| CN113535399B (en) * | 2021-07-15 | 2023-04-25 | 电子科技大学 | NFV resource scheduling method, device and system |
| CN114518948B (en) * | 2022-02-21 | 2024-09-24 | 南京航空航天大学 | Dynamic perception rescheduling method for large-scale micro-service application and application |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103516733A (en) * | 2012-06-19 | 2014-01-15 | 华为技术有限公司 | Method and apparatus for processing virtual private cloud |
| CN106254178A (en) * | 2016-08-03 | 2016-12-21 | 陈鸣 | A kind of network test platform NFVNTP based on NFV and test method thereof |
| CN106941424A (en) * | 2017-04-26 | 2017-07-11 | 北京东土科技股份有限公司 | A kind of redundancy switching method and equipment based on TTDP agreements |
| CN107220152A (en) * | 2017-06-20 | 2017-09-29 | 郑州云海信息技术有限公司 | A kind of NFV system availabilities method of testing and device |
| CN108132827A (en) * | 2016-11-30 | 2018-06-08 | 华为技术有限公司 | A kind of network slice method for mapping resource, relevant device and system |
| CN109167691A (en) * | 2018-09-27 | 2019-01-08 | 中国船舶工业综合技术经济研究院 | A kind of network redundancy design method of weapon control system |
| CN110581785A (en) * | 2018-06-11 | 2019-12-17 | 中国移动通信集团浙江有限公司 | reliability evaluation method and device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6601237B2 (en) * | 2016-01-27 | 2019-11-06 | 富士通株式会社 | Test apparatus, network system, and test method |
-
2019
- 2019-12-29 CN CN201911385566.XA patent/CN113055218B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103516733A (en) * | 2012-06-19 | 2014-01-15 | 华为技术有限公司 | Method and apparatus for processing virtual private cloud |
| CN106254178A (en) * | 2016-08-03 | 2016-12-21 | 陈鸣 | A kind of network test platform NFVNTP based on NFV and test method thereof |
| CN108132827A (en) * | 2016-11-30 | 2018-06-08 | 华为技术有限公司 | A kind of network slice method for mapping resource, relevant device and system |
| CN106941424A (en) * | 2017-04-26 | 2017-07-11 | 北京东土科技股份有限公司 | A kind of redundancy switching method and equipment based on TTDP agreements |
| CN107220152A (en) * | 2017-06-20 | 2017-09-29 | 郑州云海信息技术有限公司 | A kind of NFV system availabilities method of testing and device |
| CN110581785A (en) * | 2018-06-11 | 2019-12-17 | 中国移动通信集团浙江有限公司 | reliability evaluation method and device |
| CN109167691A (en) * | 2018-09-27 | 2019-01-08 | 中国船舶工业综合技术经济研究院 | A kind of network redundancy design method of weapon control system |
Non-Patent Citations (1)
| Title |
|---|
| 基于TOPSIS的装备保障网络节点;邢彪，曹军海，宋太亮，陈守华，董原生;《装甲兵工程学院学报》;20170630;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113055218A (en) | 2021-06-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113055218B (en) | Redundancy evaluation method and device for NFV network and computing equipment | |
| CN113033811B (en) | Processing method and device for two-quantum bit logic gate | |
| CN112862092B (en) | Training method, device, equipment and medium for heterogeneous graph convolution network | |
| CN113825165B (en) | 5G slice network congestion early warning method and device based on time diagram network | |
| CN113541980A (en) | Network slice fault root cause positioning method and device, computing equipment and storage medium | |
| CN116489038A (en) | Network traffic prediction method, device, equipment and medium | |
| CN113541986B (en) | Fault prediction method and device for 5G slice and computing equipment | |
| WO2018135515A1 (en) | Information processing device, neural network design method, and recording medium | |
| KR20200038072A (en) | Entropy-based neural networks partial learning method and system | |
| CN116055330B (en) | Digital twin network slicing method and device based on knowledge graph | |
| CN113765683A (en) | Cutover alarm shielding method and device for network slices | |
| Javaheripi et al. | Swann: Small-world architecture for fast convergence of neural networks | |
| Raval et al. | Distributed detection in neural network based multihop wireless sensor network | |
| CN112104467B (en) | Cutover operation risk rating method and device and computing equipment | |
| Dragicevic et al. | Sustainable network dynamics | |
| Jitendra et al. | Analysis of Airline Connectivity System using Graph Theory | |
| Lai et al. | Spatial-Temporal Correlation-based Prediction Model for Node and Link Residual Resources in NFV Networks | |
| CN113825151B (en) | Method and device for predicting relationship between slice alarms based on relationship graph convolution network | |
| CN116827810A (en) | Network link state prediction method, training method, device and equipment of model | |
| Jo et al. | Genetic algorithm for service function chaining in NFV | |
| CN113766532A (en) | Isolation evaluation method and device for network slice example | |
| CN113810212B (en) | Root cause positioning method and device for 5G slice user complaints | |
| Ding et al. | A high order neural network to solve n-queens problem | |
| CN118802596A (en) | Government cloud service system migration method and device apparatus, medium, and program product | |
| CN116562390A (en) | Multi-aggregation node federal knowledge distillation learning method, system, equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |