CN112398674A - Method and device for generating VNFD configuration template for describing virtual network functions - Google Patents
Method and device for generating VNFD configuration template for describing virtual network functions Download PDFInfo
- Publication number
- CN112398674A CN112398674A CN201910760929.7A CN201910760929A CN112398674A CN 112398674 A CN112398674 A CN 112398674A CN 201910760929 A CN201910760929 A CN 201910760929A CN 112398674 A CN112398674 A CN 112398674A
- Authority
- CN
- China
- Prior art keywords
- vnfd
- text
- template
- vnf deployment
- vnf
- 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.)
- Granted
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
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
- H04L41/084—Configuration by using pre-existing information, e.g. using templates or copying from other elements
- H04L41/0843—Configuration by using pre-existing information, e.g. using templates or copying from other elements based on generic templates
-
- 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
- H04L41/08—Configuration management of networks or network elements
- H04L41/0893—Assignment of logical groups to network elements
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a method and a device for generating a VNFD configuration template for describing virtual network functions, wherein the method comprises the following steps: obtaining a VNF deployment text, and determining a VNFD configuration template corresponding to the VNF deployment text according to a pre-constructed template generator. By using the method and the device, the VNFD configuration template corresponding to the VNF deployment text can be automatically generated through the pre-constructed template generator when the VNF deployment text is given, so that the generation efficiency of the VNFD configuration template is greatly improved, the deployment efficiency of Virtual Network Function (VNF) of NFV is improved, and the VNF is rapidly deployed.
Description
Technical Field
The invention relates to the technical field of internet, in particular to a method and a device for generating a VNFD configuration template for virtual network function description, electronic equipment and a storage medium.
Background
Currently, in the internet, each network device function is connected in a fixed manner, and corresponding network services are provided to the outside in a unified manner.
In practical applications, each Network device Function depends on corresponding dedicated Network hardware, so that in order to make the Network device Function not depend on the dedicated Network hardware any more, thereby reducing the investment cost of operating equipment, and accelerating the deployment of new Network device functions, operators tend to abandon heavy and expensive dedicated Network hardware, so as to implement the Network device Function by Network Function Virtualization (NFV), that is, various Network device functions are implemented in standardized general Network devices (such as servers, storage and switching devices) by using Virtualization technology.
Further, in NFV, to implement a Network device Function, a Network Function module and a Virtual Network Function Description (VNFD) (i.e., a configuration template describing deployment and operation behaviors of a virtualized Network Function module) need to be created for the Network device Function.
The existing method for generating the VNFD configuration template is to manually complete the production according to the actual requirements of the network function module.
However, manually generating the VNFD configuration template is time-consuming, labor-consuming, inefficient, and prone to errors, which results in failure to deploy the network function module.
Disclosure of Invention
In view of the above, the present invention is proposed to provide a method and an apparatus for generating a VNFD configuration template, an electronic device, and a storage medium, which overcome or at least partially solve the above problems.
According to an aspect of the present invention, a method for generating a VNFD configuration template for virtual network function description, the method includes:
acquiring a VNF deployment text of a virtual network function;
and determining a VNFD configuration template corresponding to the VNF deployment text according to a pre-constructed template generator.
According to another aspect of the present invention, there is provided an apparatus for generating a VNFD configuration template, the apparatus including:
the acquisition module is used for acquiring a VNF deployment text of a virtual network function;
and determining a VNFD configuration template corresponding to the VNF deployment text according to a pre-constructed template generator.
According to another aspect of the present invention, there is provided an electronic apparatus including: the system comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete mutual communication through the communication bus;
the memory is configured to store at least one executable instruction that causes the processor to:
acquiring a VNF deployment text of a virtual network function;
and determining a VNFD configuration template corresponding to the VNF deployment text according to a pre-constructed template generator.
According to yet another aspect of the present invention, there is provided a storage medium having stored therein at least one executable instruction, the executable instruction causing a processor to:
acquiring a VNF deployment text of a virtual network function;
and determining a VNFD configuration template corresponding to the VNF deployment text according to a pre-constructed template generator.
According to the method and the device for generating the VNFD configuration template of the description of the virtual network functions, provided by the invention, the method comprises the following steps: obtaining a VNF deployment text, and determining a VNFD configuration template corresponding to the VNF deployment text according to a pre-constructed template generator. By using the method and the device, the VNFD configuration template corresponding to the VNF deployment text can be automatically generated through the pre-constructed template generator when the VNF deployment text is given, so that the generation efficiency of the VNFD configuration template is greatly improved, the deployment efficiency of Virtual Network Function (VNF) of NFV is improved, and the VNF is rapidly deployed.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages 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:
figure 1 illustrates a flow diagram of a method of generating a VNFD configuration template according to one embodiment of the invention;
FIG. 2 shows a schematic diagram of a generator building block according to one embodiment of the invention;
figure 3 shows a schematic diagram of an apparatus for generating a VNFD configuration template according to an embodiment of the invention;
fig. 4 shows a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may 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 disclosure to those skilled in the art.
Fig. 1 shows a flow chart of a method for generating a VNFD configuration template according to an embodiment of the invention. As shown in fig. 1, the method comprises the steps of:
s101: and acquiring a VNF deployment text of a virtual network function.
In practical applications, to implement a Network device Function in the NFV, a Network Function module and a Virtual Network Function Description (VNFD) (i.e., a configuration template describing deployment and operation behaviors of a virtualized Network Function module) need to be created for the Network device Function.
It should be noted here that NFV refers to the implementation of various network device functions in standardized general network devices (servers, storage and switching devices) by using virtualization technology, and after NFV is adopted, on one hand, network device functions do not depend on dedicated hardware, and standard network devices are low in cost, and it is expected to save equipment investment cost for operators, and on the other hand, through software and hardware decoupling and function abstraction, resources can be fully and flexibly shared, so that rapid development and deployment of new services are realized, and automatic deployment, elastic expansion, fault isolation, self-healing and the like are performed based on actual service requirements.
In addition, VNFD refers to a configuration template that describes the deployment and operational behavior of a virtualized network function module, which is used for the running of the virtualized network function module and for lifecycle management of the virtualized network function module instance. The quality of VNFD production directly impacts VNF instantiation deployment. VNFD typically only requires modification of blueprints and pans files.
Further, in the process of generating the VNFD configuration template, first, a VNF deployment text needs to be acquired.
It should be noted that, because the VNFD generally only needs to modify the blueprints and the pans files, in this embodiment of the present specification, the content recorded in the VNF deployment file is all related content in the blueprints and the pans files, and specifically, the operation and maintenance personnel fills the VNF deployment file according to actual needs of the network function module.
In this specification embodiment, the VNF deployment text may include: at least one of identification information of the VNFD configuration template, network information of the VNFD configuration template, VNF information, and virtual deployment unit VDU information.
It should be noted that, in this embodiment of the present specification, the identification information of the VNFD configuration template may specifically be: VNFD (three elements, Vendor, VNFD _ id, VNFD _ version, which is a unique mark of VNFD file): VNFD _ name (VNFD name), vendor (vendor name of VNF), VNFD _ id (identifier of VNFD), VNFD _ version (version of VNFD), description (description of VNF), date _ created (creation time of VNFD), icon (picture used by VNF in graphical user interface);
the network information of the VNFD configuration template may specifically be: vnfd.net.network and vnfd.net.subnet;
the VNF information may specifically be: vnfd.node.vnf: VNF _ name (name of VNF), blueprint _ name (for association with the reference of the plan), VNF _ type (type of VNF), VNF _ version (version of VNF), max _ vm _ num (maximum number of virtual machines deployed by VNF), priority (specifying VNF priority), user _ data (customization data), default _ user _ embedded _ in _ VNF (indicating that a default user name and password for accessing VNF need to be carried when the template is added), software _ packages, VNF _ uri (a. declaring each life cycle of MML command execution under VNFD; b. declaring REST message uri of MANO and opposite end), status _ query (indicating that message is sent to the uri regularly, if failure occurs, app chain-breaking alarm);
the VDU information of the virtual deployment unit may specifically be: providing an image of a virtual machine (the configuration is created in a class mode or an instance mode when the virtual machine is created), a deployment order of the virtual machine (Priority, which refers to deployment Priority of different virtual machines in a VNF), an inject _ files (a network element needs to know communication IP, port, user name and password of a MANO, and in addition, the network element needs other information such as IP, MAC, gateway and the like, and the information is injected to the virtual machine by means of injection), an image/container _ file (a software compression packet which defines dependence needs to be uploaded on a page of the MANO), and a storage _ type (a magnetic array storage network _ image and a local image storage local _ image).
In addition, in this embodiment of the present specification, the VNF deployment text provided by the operation and maintenance personnel needs to include the following contents in table 1:
table 1S 102: and determining a VNFD configuration template corresponding to the VNF deployment text according to a pre-constructed template generator.
Further, in the embodiments of the present description, after obtaining the VNF deployment text, a VNFD configuration template corresponding to the VNF deployment text needs to be determined according to a pre-constructed template generator.
It should be noted that, since the VNFD configuration template corresponding to the VNF deployment text needs to be determined according to a pre-constructed template generator, the template generator needs to be pre-constructed before the template generator is used to generate the VNFD configuration template corresponding to the VNF deployment text.
The embodiment of the present specification provides an implementation manner of constructing a template generator, which is specifically as follows:
acquiring a template discriminator to be constructed, a template generator and a first training data set, wherein the first training data set comprises: the method comprises the steps that a first VNF deploys a training text and a corresponding first VNFD configuration template, a template discriminator and a template generator to be constructed are initialized, a second VNFD configuration template corresponding to the first VNF deploys the training text in a first training data set is determined through the initialized template generator to be constructed, the first VNF deploys the training text and the corresponding second VNFD configuration template as a second training data set, the training text and the corresponding first VNFD configuration template and the corresponding second VNFD configuration template are deployed according to the first VNF, a first output value corresponding to the first VNFD configuration template and a second output value corresponding to the second VNFD configuration template are constructed according to the first VNF, the initialized template discriminator to be constructed is trained, and the template generator is constructed according to the trained template discriminator.
It should be noted here that only the parameters in the template discriminator to be constructed are unknown, and the parameters need to be determined by training data, and others are known, and similarly, only the parameters in the template generator to be constructed are unknown, and others are known. The first VNF deployment training text in the first training data set and the corresponding first VNFD configuration template are true and correct, and specifically, the first training data set is obtained by collecting the VNF deployment text generated in the actual application and the corresponding true and correct VNFD configuration template.
Since the template generator to be constructed is only an initialization mode at this time, and the parameters in the template generator are not optimal, the second VNFD configuration template corresponding to the first VNF deployment training text in the first training data set determined by the template generator to be constructed is wrong. At this point, a correct training data set and an incorrect training data set are generated.
It should be noted that, as shown in fig. 2, the generator is composed of an encoder (encoder) and a decoder (decoder), and includes 1 embedded layer, 12 hidden layers, and 1 output layer. In the embodiment of the present specification, the neuron selects a long-short term memory (LSTM) neuron, and the LSTM can remember long-term information by controlling the time for storing values in the cache, which is suitable for long-sequence learning.
Here, it is also to be noted that the embedding layer (embedding): the input data dimension is set to the size N1 of the VNF deployment text dataset text dictionary and the output is set to the size 128 dimension that requires converting words into vector space. The role of this layer is to perform vector mapping (word mappings) on each word in the input text, i.e. to convert the sequence of integers for each word in the text into a vector of fixed shape 128 dimensions;
hiding the layer: 64 LSTM neurons are arranged in 12 hidden layers, and the activation function of each layer is set to be 'relu';
output layer (fully connected Dense layer): the number of all-connected neurons containing the Dense is set to be the same as the dimension of the output sequence, the activation function is set to be 'softmax', the softmax output result is input into the loss function, and the output shape is converted into the dimension of final output.
In addition, when the encoded VNF deployment text is input, the length of each VNF deployment text in the data set needs to be filled with L before the model is input for training.
In an embodiment of this specification, the template discriminator to be constructed after training initialization may specifically be as follows, according to the first VNF deployment training text, the first VNFD configuration template and the second VNFD configuration template corresponding to the first VNF deployment training text, and the first output value corresponding to the first VNFD configuration template and the second output value corresponding to the second VNFD configuration template:
determining, by a random discard layer, whether to input the VNF deployment training text and its corresponding VNFD configuration template into an embedding layer, upon determining to input the VNF deployment training texts and their corresponding VNFD configuration templates into an embedding layer, the embedding layer, via the embedded layer, determining a multi-dimensional vector VNF deployment training text and a VNFD configuration template corresponding to the multi-dimensional vector VNF deployment training text according to the VNF deployment training text and the VNFD configuration template corresponding to the VNF deployment training text, extracting the multi-dimensional vector VNF deployment training text and the text features corresponding to the VNFD configuration templates through the convolution layer, determining and reserving the text features with the maximum feature values through the pooling layer, determining the VNF deployment training text and a third output value corresponding to the VNFD configuration template corresponding to the VNF deployment training text according to the text feature with the maximum feature value, and training a template discriminator according to the third output value, the first output value and the second output value.
It should be noted that the VNF deployment training text and the corresponding VNFD configuration template thereof may be a first VNF deployment training text and a corresponding first VNFD configuration template thereof, or may be the first VNF deployment training text and a corresponding second VNFD configuration template thereof. The length of the sequence input into the random discard layer is L + M, wherein the longest length of the VNF deployment text in the total training data set is taken as the length L of the coding sequence, and the longest length of the VNFD deployment text in the total training data set is taken as the length M of the coding sequence.
It should be further noted that the discriminator is a classifier composed of a convolutional neural network, and the specific details are as follows:
random discard layer (dropout): the rejection probability is set to be 0.5, the input neurons are randomly disconnected according to a certain probability (50%) when parameters are updated every time in the training process, and the Dropout layer is used for preventing overfitting;
embedding layer (embedding): the input is set to the size of the total data text dictionary, N2, and the output is set to the size 128 dimension needed to convert words to vector space. Converting the integer sequence of each word in the VNF deployment requirement text and the corresponding VNFD configuration to a fixed-shape 128-dimensional vector;
convolutional layer (Conv 1D): the number of convolution kernels is 48 (i.e., the output dimension), the spatial window length of the convolution kernel is set to 2 (i.e., the convolution kernel reads 2 words at a time in succession), and the activation function is set to "relu". Extracting text features by utilizing the convolutional layer;
maximum pooling layer (MaxPooling 1D): the size of the pooling window is set to be 2, the maximum pooling layer reserves the maximum value in the characteristic values extracted by the convolution kernel, and other characteristic values are discarded completely;
planarization layer (flatten) this layer is used to "flatten" the input, converting the three-dimensional input into two dimensions, often used in the transition from a convolutional layer to a fully-connected layer.
Full connection layer: contains 16 neurons, with the activation function set to "relu";
output layer (sense full interconnect layer): 1 sense neuron is contained, the activation function is set to be 'sigmoid', and the output value is 1 or 0. And outputting the sigmoid to a loss function.
In addition, in this embodiment of the present specification, a training text and a corresponding VNFD configuration template need to be deployed according to a third output value and the VNF, and parameters of the template discriminator are adjusted through an objective function.
It should be noted here that the objective function is
Wherein θ is a model parameter, ci is an ith VNF deployment text, xi is a VNFD configuration template corresponding to the ith VNF deployment text, D (ci, xi) is a third output value of the output of the discriminator, and when D (ci, xi) is a positive value, the discriminator parameter is updated to increase Pθ(xi|ci) When D (ci, xi) is negative, the discriminator parameter is updated to decrease Pθ(xi|ci)。
Further, in the embodiments of the present description, after the training of the classifier is completed, the template generator needs to be constructed according to the trained template classifier, which is specifically as follows:
obtaining a third training data set, wherein the third training data set comprises: the method comprises the steps that a training text is deployed on a second VNF, a VNFD configuration template corresponding to the second VNF deployment text is generated by the template generator to be created, a fourth output value corresponding to the second VNF deployment text and the VNFD configuration template corresponding to the second VNF deployment text is determined through the template discriminator, and training parameters in the template generator are adjusted according to the fourth output value, the first output value and the second output value.
Thus far, the construction of the template generator has been completed, and in the embodiment of the present specification, the construction of the whole template generator is roughly: model parameters of the generator G and the discriminator D are initialized first, and then in the process of training the generator and the discriminator, each training cycle is as follows:
(1) fixing the parameters of the generator, training the discriminator, updating the parameters of the discriminator: randomly taking out a VNF deployment text C and a corresponding VNFD configuration X from a first training data set, randomly taking out a VNF deployment text C 'from the first training data set, inputting the VNF deployment text C' into a generator G (C '), generating a corresponding VNFD configuration X', forming a second training data set, inputting the VNF deployment text serving as a condition and the corresponding VNFD configuration into a discriminator, learning by the discriminator to give a higher score to the VNF deployment text and the corresponding real VNFD configuration pair, and giving a lower score to the VNF deployment text and the corresponding generated VNFD configuration pair, thereby gradually improving the discrimination capability of the discriminator;
(2) fixing the parameters of the arbiter, training the generator, updating the parameters of the generator: obtaining a third training data set, wherein the third training data set comprises: the method comprises the steps that a training text is deployed on a second VNF, a VNFD configuration template corresponding to the second VNF deployment text is generated by the template generator to be created, a fourth output value corresponding to the second VNF deployment text and the VNFD configuration template corresponding to the second VNF deployment text is determined through the template discriminator, and training parameters in the template generator are adjusted according to the fourth output value, the first output value and the second output value.
It should be noted that, the objective of the generator learning is to make the generated VNFD configuration closer to the actual VNFD configuration as better, that is, the closer the fourth output value to the first output value is as better, the closer the generated VNFD configuration is to the actual VNFD configuration, so as to maximize the score output by the discriminator, and the score can be regarded as the return in the reinforcement learning, thereby gradually improving the generating capability of the generator.
Finally, the model is trained until the discriminator cannot distinguish whether the VNFD configuration template corresponding to the VNF deployment text is real or generated by the generator, the model weight is derived after convergence, and the VNFD can be automatically generated by the trained generator.
In addition, during the training process of the model, the acquired training data set can be represented as follows:
S＝{(c1,x1),(c2,x2),(c3,x3),…,(cn,xn) Where S denotes a training data set, c1To cnRepresenting VNF deployment text, x1To xnThe VNFD configuration template corresponding to the VNF deployment text is represented.
Subsequently, the constructed template generator may be directly used to determine a VNFD configuration template corresponding to the VNF deployment text, specifically, the VNF deployment text is subjected to integer serialization to generate an integer serialization VNF deployment text, the integer serialization VNF deployment text is converted into a multidimensional vector VNF deployment text, a text feature corresponding to the multidimensional vector VNF deployment text is extracted through an encoder included in the template generator, and the VNFD configuration template corresponding to the VNF deployment text is determined through a decoder included in the template generator according to the text feature corresponding to the multidimensional vector VNF deployment text.
It should be noted that, specifically, the integer serialization of the VNF deployment text is to use a unified letter case, convert a capital letter into a small letter, and convert each word in the data into an integer sequence, for example: [ "qci": 40, "imsi": 105, "info": 8, "update": 278, "on": 89, "agent": 164, "modify": 59, "the": 21, "type": 303, "storage": 231,...].
By the method, the VNF deployment text is given, and the VNFD configuration template corresponding to the VNF deployment text can be automatically generated through the pre-constructed template generator, so that the generation efficiency of the VNFD configuration template is greatly improved, the deployment efficiency of Virtual Network Function (VNF) of NFV is improved, and the VNF is rapidly deployed.
Finally, the generated VNFD configuration template is imported into a Virtualized Network Function Manager (VNFM) of the MANO, and the VNFM deploys the VNF according to the VNFD configuration template.
It should be noted that, when the VNFM is a function module for performing lifecycle management of a virtualized network function module, importing a VNFD file into the VNFM may implement VNF deployment through a C10 interface between the VNFM and the VNF.
To sum up, in the embodiment of the present description, a deep learning framework is used to build a conditional countermeasure generation network model, where the model is composed of a generator and a discriminator, a VNFD generator is built using an encoder-decoder architecture, and is responsible for generating a VNFD configuration template for specifying a VNF deployment description, and a discriminator composed of a convolutional neural network is built at the same time and is responsible for judging how much difference exists between a result generated by the generator and a true VNFD configuration template, and the generator is adjusted according to an output of the discriminator, so that the generator can generate a correct VNFD configuration template corresponding to the VNF deployment description. Therefore, the manufacturing efficiency of the VNFD is greatly improved, and the VNF deployment efficiency of the NFV is further improved.
Based on the above method for generating a VNFD configuration template described in this application, an embodiment of the present application provides an apparatus for generating a VNFD configuration template described in this application, and as shown in fig. 3, the apparatus includes:
an obtaining module 301, configured to obtain a virtual network function VNF deployment text;
a determining module 302, configured to determine, according to a pre-constructed template generator, a VNFD configuration template corresponding to the VNF deployment text.
The VNF deployment text comprises: at least one of identification information of the VNFD configuration template, network information of the VNFD configuration template, VNF information, and virtual deployment unit VDU information.
The determining module 302 is specifically configured to perform integer serialization on the VNF deployment text to generate an integer serialized VNF deployment text, convert the integer serialized VNF deployment text into a multidimensional vector VNF deployment text, extract text features corresponding to the multidimensional vector VNF deployment text through an encoder included in a template generator, and determine, according to the text features corresponding to the multidimensional vector VNF deployment text, a VNFD configuration template corresponding to the VNF deployment text through a decoder included in the template generator.
The device further comprises:
a building module 303, configured to obtain a template discriminator to be built, a template generator, and a first training data set, where the first training data set includes: a first VNF deploys a training text and a corresponding first VNFD configuration template; determining a second VNFD configuration template corresponding to a first VNF deployment training text in a first training data set through a template generator to be constructed, and taking the first VNF deployment training text and the second VNF deployment training template corresponding to the first VNF deployment training text as a second training data set; training the template discriminator to be constructed according to the first VNF deployment training text, the first VNFD configuration template and the second VNFD configuration template corresponding to the first VNF deployment training text, and the first output value corresponding to the first VNFD configuration template and the second output value corresponding to the second VNFD configuration template; and constructing a template generator according to the trained template discriminator.
The building module 303 is specifically configured to determine, through a random discard layer, whether to input the VNF deployment training text and the VNFD configuration template corresponding to the VNF deployment training text into an embedding layer; when the VNF deployment training text and the VNFD configuration template corresponding to the VNF deployment training text are determined to be input to the embedding layer, the multi-dimensional vector VNF deployment training text and the VNFD configuration template corresponding to the VNF deployment training text are determined through the embedding layer according to the VNF deployment training text and the VNFD configuration template corresponding to the VNF deployment training text; extracting the multi-dimensional vector VNF deployment training text and text features corresponding to the VNFD configuration templates through a convolution layer; determining and reserving the text features with the maximum feature values through a pooling layer; determining the VNF deployment training text and a third output value corresponding to the VNFD configuration template corresponding to the VNF deployment training text according to the text feature with the maximum feature value; and training a template discriminator according to the third output value, the first output value and the second output value.
The building module 303 is specifically configured to adjust a parameter of the template discriminator through an objective function according to the third output value, the VNF deployment training text, and the VNFD configuration template corresponding to the VNF deployment training text.
The building module 303 is specifically configured to obtain a third training data set, where the third training data set includes: the second VNF deploys training texts; the template generator to be created generates a VNFD configuration template corresponding to the second VNF deployment text; determining, by the template discriminator, the second VNF deployment text and a fourth output value corresponding to the VNFD configuration template corresponding thereto; and adjusting the training parameters in the template generator according to the fourth output value, the first output value and the second output value.
The embodiment of the present application further provides a non-volatile computer storage medium, where the computer storage medium stores at least one executable instruction, and the computer executable instruction may execute the method for generating the VNFD configuration template described in any method embodiment above.
Fig. 4 is a schematic structural diagram of an electronic 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 electronic device.
As shown in fig. 4, the server 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 may specifically execute relevant steps in the above-described method for generating a VNFD configuration template.
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 electronic device comprises one or more processors, which can 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.
The program 410 may specifically be configured to cause the processor 402 to perform the following operations:
acquiring a VNF deployment text of a virtual network function;
and determining a VNFD configuration template corresponding to the VNF deployment text according to a pre-constructed template generator.
Optionally, the program 410 may also be for causing the processor 402 to:
the VNF deployment text comprises: at least one of identification information of the VNFD configuration template, network information of the VNFD configuration template, VNF information, and virtual deployment unit VDU information.
Optionally, the program 410 may also be for causing the processor 402 to:
the VNF deployment texts are subjected to integer serialization to generate integer serialization VNF deployment texts, the integer serialization VNF deployment texts are converted into multi-dimensional vector VNF deployment texts, text features corresponding to the multi-dimensional vector VNF deployment texts are extracted through an encoder included in a template generator, and VNFD configuration templates corresponding to the VNF deployment texts are determined through a decoder included in the template generator according to the text features corresponding to the multi-dimensional vector VNF deployment texts.
Optionally, the program 410 may also be for causing the processor 402 to:
acquiring a template discriminator to be constructed, a template generator and a first training data set, wherein the first training data set comprises: a first VNF deploys a training text and a corresponding first VNFD configuration template; determining a second VNFD configuration template corresponding to a first VNF deployment training text in a first training data set through a template generator to be constructed, and taking the first VNF deployment training text and the second VNF deployment training template corresponding to the first VNF deployment training text as a second training data set; training the template discriminator to be constructed according to the first VNF deployment training text, the first VNFD configuration template and the second VNFD configuration template corresponding to the first VNF deployment training text, and the first output value corresponding to the first VNFD configuration template and the second output value corresponding to the second VNFD configuration template; and constructing a template generator according to the trained template discriminator.
Optionally, the program 410 may also be for causing the processor 402 to:
determining whether the VNF deployment training text and the VNFD configuration template corresponding to the VNF deployment training text are input into an embedding layer or not through a random abandoning layer; when the VNF deployment training text and the VNFD configuration template corresponding to the VNF deployment training text are determined to be input to the embedding layer, the multi-dimensional vector VNF deployment training text and the VNFD configuration template corresponding to the VNF deployment training text are determined through the embedding layer according to the VNF deployment training text and the VNFD configuration template corresponding to the VNF deployment training text; extracting the multi-dimensional vector VNF deployment training text and text features corresponding to the VNFD configuration templates through a convolution layer; determining and reserving the text features with the maximum feature values through a pooling layer; determining the VNF deployment training text and a third output value corresponding to the VNFD configuration template corresponding to the VNF deployment training text according to the text feature with the maximum feature value; and training a template discriminator according to the third output value, the first output value and the second output value.
Optionally, the program 410 may also be for causing the processor 402 to:
and according to the third output value, the VNF deployment training text and the corresponding VNFD configuration template, and adjusting the parameters of the template discriminator through an objective function.
Optionally, the program 410 may also be for causing the processor 402 to:
obtaining a third training data set, wherein the third training data set comprises: the second VNF deploys training texts; the template generator to be created generates a VNFD configuration template corresponding to the second VNF deployment text; determining, by the template discriminator, the second VNF deployment text and a fourth output value corresponding to the VNFD configuration template corresponding thereto; and adjusting the training parameters in the template generator according to the fourth output value, the first output value and the second output value.
The algorithms and displays presented herein are not inherently related to any particular computer, virtual machine, 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. Moreover, the present invention is 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 invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure 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 described 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.
The various component embodiments of the invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that a microprocessor or Digital Signal Processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components of the generation apparatus of the virtual network function description VNFD configuration template according to embodiments of the present invention. The present invention may also be embodied as apparatus or device programs (e.g., computer programs and computer program products) for performing a portion or all of the methods described herein. Such programs implementing the present invention may be stored on computer-readable media or may be in the form of one or more signals. Such a signal may be downloaded from an internet website or provided on a carrier signal or in any other form.
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.
Claims (10)
1. A method for generating a VNFD configuration template for Virtual Network Function Description (VNFD), comprising the following steps:
acquiring a VNF deployment text of a virtual network function;
and determining a VNFD configuration template corresponding to the VNF deployment text according to a pre-constructed template generator.
2. The method of claim 1, the VNF deployment text comprising: at least one of identification information of the VNFD configuration template, network information of the VNFD configuration template, VNF information, and virtual deployment unit VDU information.
3. The method according to claim 1, wherein the step of determining the VNFD configuration template corresponding to the VNF deployment text according to a pre-constructed template generator specifically comprises:
performing integer serialization on the VNF deployment text to generate an integer serialized VNF deployment text;
converting the integer-serialized VNF deployment text into a multi-dimensional vector VNF deployment text;
extracting text features corresponding to the multi-dimensional vector VNF deployment text through an encoder included by a template generator;
and determining a VNFD configuration template corresponding to the VNF deployment text through a decoder included in a template generator according to the text features corresponding to the multi-dimensional vector VNF deployment text.
4. The method of claim 1, constructing a template generator, specifically comprising:
acquiring a template discriminator to be constructed, a template generator and a first training data set, wherein the first training data set comprises: a first VNF deploys a training text and a corresponding first VNFD configuration template;
determining a second VNFD configuration template corresponding to a first VNF deployment training text in a first training data set through a template generator to be constructed, and taking the first VNF deployment training text and the second VNF deployment training template corresponding to the first VNF deployment training text as a second training data set;
training the template discriminator to be constructed according to the first VNF deployment training text, the first VNFD configuration template and the second VNFD configuration template corresponding to the first VNF deployment training text, and the first output value corresponding to the first VNFD configuration template and the second output value corresponding to the second VNFD configuration template;
and constructing a template generator according to the trained template discriminator.
5. The method according to claim 4, wherein the training of the template discriminator according to the first VNF deployment training text, the first VNFD configuration template and the second VNFD configuration template corresponding to the first VNF deployment training text, and the first output value corresponding to the first VNFD configuration template and the second output value corresponding to the second VNFD configuration template specifically comprises:
determining whether the VNF deployment training text and the VNFD configuration template corresponding to the VNF deployment training text are input into an embedding layer or not through a random abandoning layer;
when the VNF deployment training text and the VNFD configuration template corresponding to the VNF deployment training text are determined to be input to the embedding layer, the multi-dimensional vector VNF deployment training text and the VNFD configuration template corresponding to the VNF deployment training text are determined through the embedding layer according to the VNF deployment training text and the VNFD configuration template corresponding to the VNF deployment training text;
extracting the multi-dimensional vector VNF deployment training text and text features corresponding to the VNFD configuration templates through a convolution layer;
determining and reserving the text features with the maximum feature values through a pooling layer;
determining the VNF deployment training text and a third output value corresponding to the VNFD configuration template corresponding to the VNF deployment training text according to the text feature with the maximum feature value;
and training a template discriminator according to the third output value, the first output value and the second output value.
6. The method of claim 5, further comprising:
and according to the third output value, the VNF deployment training text and the corresponding VNFD configuration template, and adjusting the parameters of the template discriminator through an objective function.
7. The method of claim 4, wherein the constructing the template generator according to the trained template discriminator specifically comprises:
obtaining a third training data set, wherein the third training data set comprises: the second VNF deploys training texts;
the template generator to be created generates a VNFD configuration template corresponding to the second VNF deployment text;
determining, by the template discriminator, the second VNF deployment text and a fourth output value corresponding to the VNFD configuration template corresponding thereto;
and adjusting the training parameters in the template generator according to the fourth output value, the first output value and the second output value.
8. An apparatus for generating a VNFD configuration template for virtual network function description, comprising:
the acquisition module is used for acquiring a VNF deployment text of a virtual network function;
and the determining module is used for determining the VNFD configuration template corresponding to the VNF deployment text according to a pre-constructed template generator.
9. An electronic device, comprising: the system comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete mutual communication through the communication bus;
the memory is used for storing at least one executable instruction, and the executable instruction causes the processor to execute the operation corresponding to the generation method of the Virtual Network Function Description (VNFD) configuration template as claimed in any one of claims 1-7.
10. A storage medium having stored therein at least one executable instruction to cause a processor to perform operations corresponding to the method of generating a virtual network function description, VNFD, configuration template as recited in any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910760929.7A CN112398674B (en) | 2019-08-16 | 2019-08-16 | Method and device for generating VNFD configuration template for describing virtual network functions |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910760929.7A CN112398674B (en) | 2019-08-16 | 2019-08-16 | Method and device for generating VNFD configuration template for describing virtual network functions |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112398674A true CN112398674A (en) | 2021-02-23 |
| CN112398674B CN112398674B (en) | 2022-08-23 |
Family
ID=74603175
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910760929.7A Active CN112398674B (en) | 2019-08-16 | 2019-08-16 | Method and device for generating VNFD configuration template for describing virtual network functions |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112398674B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112953751A (en) * | 2019-12-11 | 2021-06-11 | 中国移动通信有限公司研究院 | Template processing method, device and equipment and computer readable storage medium |
| CN115225498A (en) * | 2021-03-29 | 2022-10-21 | 中国移动通信集团广东有限公司 | Office data configuration method and device |
| CN116132289A (en) * | 2022-09-27 | 2023-05-16 | 马上消费金融股份有限公司 | Information configuration method, device, equipment and medium |
| WO2024104312A1 (en) * | 2022-11-18 | 2024-05-23 | 华为技术有限公司 | Method and apparatus for deploying virtualized network function |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104468688A (en) * | 2013-09-13 | 2015-03-25 | 株式会社Ntt都科摩 | Method and apparatus for network virtualization |
| CN106161173A (en) * | 2015-04-15 | 2016-11-23 | 中兴通讯股份有限公司 | A kind of virtual network function that realizes disposes the method and device of specification configuration |
| WO2017080391A1 (en) * | 2015-11-09 | 2017-05-18 | 中兴通讯股份有限公司 | Network service deployment method and device |
| CN107209686A (en) * | 2015-01-23 | 2017-09-26 | 日本电气株式会社 | Network function virtual management and method of combination, equipment and program |
| CN108370341A (en) * | 2016-03-02 | 2018-08-03 | 华为技术有限公司 | Resource allocation method, virtual network function manager and Element management system |
| CN108762768A (en) * | 2018-05-17 | 2018-11-06 | 烽火通信科技股份有限公司 | Network Intelligent Service dispositions method and system |
| CN109639479A (en) * | 2018-12-07 | 2019-04-16 | 北京邮电大学 | Based on the network flow data Enhancement Method and device for generating confrontation network |
| US20190149425A1 (en) * | 2017-11-16 | 2019-05-16 | Verizon Patent And Licensing Inc. | Method and system for virtual network emulation and self-organizing network control using deep generative models |
| US20190155284A1 (en) * | 2017-11-20 | 2019-05-23 | At&T Intellectual Property I, L.P. | Object modeling with adversarial learning |
-
2019
- 2019-08-16 CN CN201910760929.7A patent/CN112398674B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104468688A (en) * | 2013-09-13 | 2015-03-25 | 株式会社Ntt都科摩 | Method and apparatus for network virtualization |
| CN107209686A (en) * | 2015-01-23 | 2017-09-26 | 日本电气株式会社 | Network function virtual management and method of combination, equipment and program |
| CN106161173A (en) * | 2015-04-15 | 2016-11-23 | 中兴通讯股份有限公司 | A kind of virtual network function that realizes disposes the method and device of specification configuration |
| WO2017080391A1 (en) * | 2015-11-09 | 2017-05-18 | 中兴通讯股份有限公司 | Network service deployment method and device |
| CN108370341A (en) * | 2016-03-02 | 2018-08-03 | 华为技术有限公司 | Resource allocation method, virtual network function manager and Element management system |
| US20190149425A1 (en) * | 2017-11-16 | 2019-05-16 | Verizon Patent And Licensing Inc. | Method and system for virtual network emulation and self-organizing network control using deep generative models |
| US20190155284A1 (en) * | 2017-11-20 | 2019-05-23 | At&T Intellectual Property I, L.P. | Object modeling with adversarial learning |
| CN108762768A (en) * | 2018-05-17 | 2018-11-06 | 烽火通信科技股份有限公司 | Network Intelligent Service dispositions method and system |
| CN109639479A (en) * | 2018-12-07 | 2019-04-16 | 北京邮电大学 | Based on the network flow data Enhancement Method and device for generating confrontation network |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112953751A (en) * | 2019-12-11 | 2021-06-11 | 中国移动通信有限公司研究院 | Template processing method, device and equipment and computer readable storage medium |
| CN112953751B (en) * | 2019-12-11 | 2023-11-28 | 中国移动通信有限公司研究院 | Template processing method, device, equipment and computer readable storage medium |
| CN115225498A (en) * | 2021-03-29 | 2022-10-21 | 中国移动通信集团广东有限公司 | Office data configuration method and device |
| CN116132289A (en) * | 2022-09-27 | 2023-05-16 | 马上消费金融股份有限公司 | Information configuration method, device, equipment and medium |
| WO2024104312A1 (en) * | 2022-11-18 | 2024-05-23 | 华为技术有限公司 | Method and apparatus for deploying virtualized network function |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112398674B (en) | 2022-08-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112398674B (en) | Method and device for generating VNFD configuration template for describing virtual network functions | |
| CN106980868A (en) | Embedded space for the image with multiple text labels | |
| CN110597719B (en) | Image clustering method, device and medium for adaptation test | |
| US11244466B2 (en) | Automated capacity management using artificial intelligence techniques | |
| CN110728295A (en) | Semi-supervised landform classification model training and landform graph construction method | |
| CN111523324A (en) | Training method and device for named entity recognition model | |
| CN113541980A (en) | Network slice fault root cause positioning method and device, computing equipment and storage medium | |
| CN116881429B (en) | Multi-tenant-based dialogue model interaction method, device and storage medium | |
| CN108563951A (en) | Method for detecting virus and device | |
| CN112202584B (en) | Alarm correlation method, device, computing equipment and computer storage medium | |
| CN112669215A (en) | Training text image generation model, text image generation method and device | |
| WO2021042895A1 (en) | Neural network-based verification code identification method and system, and computer device | |
| WO2018101958A1 (en) | Fuzzy input for autoencoders | |
| CN110135428A (en) | Image segmentation processing method and device | |
| CN111445545B (en) | Text transfer mapping method and device, storage medium and electronic equipment | |
| CN115713669A (en) | Image classification method and device based on inter-class relation, storage medium and terminal | |
| CN113591881A (en) | Intention recognition method and device based on model fusion, electronic equipment and medium | |
| CN114299370A (en) | Internet of things scene perception method and device based on cloud edge cooperation | |
| US11588753B2 (en) | Methods and systems for generating deployment architecture and template | |
| CN109933969B (en) | Verification code identification method and device, electronic equipment and readable storage medium | |
| CN113747480A (en) | Processing method and device for 5G slice fault and computing equipment | |
| CN113762978B (en) | Complaint delimiting method and device for 5G slicing user and computing equipment | |
| CN116798052B (en) | Training method and device of text recognition model, storage medium and electronic equipment | |
| CN114079958B (en) | Disaster recovery pool selection method and device for 5G slice instance and computing equipment | |
| CN112101394B (en) | Provider domain deployment method, device, computing equipment and computer storage 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 |