CN113054651B - Network topology optimization method, device and system - Google Patents
Network topology optimization method, device and system Download PDFInfo
- Publication number
- CN113054651B CN113054651B CN201911368600.2A CN201911368600A CN113054651B CN 113054651 B CN113054651 B CN 113054651B CN 201911368600 A CN201911368600 A CN 201911368600A CN 113054651 B CN113054651 B CN 113054651B
- Authority
- CN
- China
- Prior art keywords
- topology
- network
- samples
- network topology
- evaluation
- 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
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The application discloses a network topology optimization method, which comprises the following steps: acquiring a first network topology and a flow model of a target network; according to the first network topology and the flow model, determining a first topology evaluation value corresponding to the first network topology through a target topology evaluation which is predetermined based on the graph neural network and/or the artificial topology evaluation; and if the first topology evaluation value does not meet the first preset condition, determining a second network topology according to the first network topology, wherein a second topology evaluation value corresponding to the second network topology meets the first preset condition, and the second network topology is used for network topology adjustment of the target network, so that the network topology of the target network is adjusted to the second network topology. According to the technical scheme, the topological optimization problem of the network is converted into the mathematical optimization problem taking the topological evaluation as the objective function, so that the topological optimization result is easy to quantify and evaluate, the calculation complexity in the topological optimization process is obviously reduced, and the network topological optimization problem is solved systematically.
Description
Technical Field
The present application relates to the field of communications technologies, and in particular, to a method, an apparatus, and a system for network topology optimization.
Background
The increasing contradiction between the digital communication service demand and the laggard communication network performance will restrict the development progress of the times of internet of everything, for example, in application scenarios such as Virtual Reality (VR) and internet of things, the problems of communication delay, traffic congestion and the like will become more and more prominent. In order to improve the transmission capabilities of a communication network, in addition to improving device performance, the communication network topology needs to be carefully designed and adjusted. In essence, the network topology and the routing algorithm determine the flow direction of the traffic flow in the network, and therefore determine key network performance indexes such as traffic rate, delay and network congestion. The current network topology optimization is based on empirical rules and simulation technology, the calculation complexity is high, and a systematic optimization scheme is lacked.
Disclosure of Invention
The embodiment of the application provides a network topology optimization method, so that the result of topology optimization is easy to quantify and evaluate, the computational complexity in the topology optimization process can be obviously reduced, and the problem of network topology optimization is solved systematically.
In order to achieve the above object, the embodiments of the present application provide the following technical solutions:
a first aspect of the present application provides a method for optimizing a network topology, including: acquiring a first network topology and a flow model of a target network, wherein the first network topology is a current network topology structure of the target network, namely the current network topology of the target network comprises forwarding nodes, the positions and the connection relations of the forwarding nodes, capacity data and other topology information; determining a first topology evaluation value corresponding to a first network topology through a predetermined target topology evaluation according to the first network topology and a flow model, wherein the target topology evaluation is determined based on a graph neural network and/or an artificial topology evaluation, the graph neural network is obtained based on a plurality of topology element samples and index value training of a network performance index corresponding to each topology element sample, the graph neural network is used for indicating a relation between the network topology and the network performance index, the target topology evaluation is a topology evaluation for the target network, the topology evaluation can be used for measuring network communication performance of the target network, such as communication performance of bearing capacity, traffic transmission delay, transmission rate, congestion and the like, the graph neural network is a deep learning neural network with input as network topology (or a network topology graph), the artificial topology evaluation is a manually given topology evaluation, and the principle of the artificial topology evaluation is based on mathematical theories such as graph theory and the like; and if the first topology evaluation value does not meet the first preset condition, determining a second network topology according to the first network topology, wherein a second topology evaluation value corresponding to the second network topology meets the first preset condition, and the second network topology is used for network topology adjustment of the target network, so that the network topology of the target network is adjusted to the second network topology.
From the first aspect, the topology optimization problem of the network is converted into the mathematical optimization problem taking the topology evaluation as the objective function by using the topology evaluation instead of simulation, so that the result of the topology optimization is easy to quantify and evaluate, the computational complexity in the topology optimization process can be obviously reduced, and the network topology optimization problem can be solved systematically.
Optionally, with reference to the first aspect, in a first possible implementation manner of the first aspect, before determining, according to the first network topology and the traffic model, a first topology evaluation value corresponding to the first network topology through a predetermined target topology evaluation, the method further includes: acquiring a graph neural network and/or artificial topology evaluation; and determining a target topology evaluation according to the graph neural network and/or the artificial topology evaluation.
Optionally, with reference to the first possible implementation manner of the first aspect, in a second possible implementation manner of the first aspect, the obtaining a graph neural network includes: acquiring a plurality of topological meta-samples and index values of network performance indexes corresponding to the topological meta-samples; and taking the topological element sample as input, and taking the index value of the network performance index corresponding to the topological element sample as output, and training the graph neural network.
Optionally, with reference to the second possible implementation manner of the first aspect, in a third possible implementation manner of the first aspect, the obtaining index values of a plurality of topology metadata samples and a network performance index corresponding to each topology metadata sample includes: acquiring the scale of the topological elements, wherein the scale of the topological elements is used for indicating the number of nodes of a topological element sample; generating a plurality of topological element samples according to the topological element scale; and acquiring an index value of the network performance index of each topological element sample.
Optionally, with reference to the first aspect and any one of the first to third possible implementation manners of the first aspect, in a fourth possible implementation manner of the first aspect, if the first topology evaluation value does not satisfy the preset condition, determining a second network topology according to the first network topology includes: determining a plurality of first samples according to the first network topology, wherein the plurality of first samples are topology structures of which the topology evaluation values meet a second preset condition in the plurality of first topology structures, and the plurality of first topology structures are topology structures which have the connectivity property of the first network topology in the plurality of topology structures randomly generated based on the first network topology; determining a plurality of second samples according to the plurality of first samples, wherein the plurality of second samples are topological structures with the connection property of the first network topology in the plurality of second topological structures, and the plurality of second topological structures are generated based on every two samples in the plurality of first samples; determining a plurality of third samples according to the plurality of second samples, wherein the plurality of third samples are topological structures with the connectivity property of the first network topology in the plurality of third topological structures, and the plurality of third topological structures are generated by randomly changing part or all of the second samples in the plurality of second samples; determining a target sample according to the plurality of third samples, wherein the target sample is the third sample with the smallest topological evaluation value in the plurality of third samples; and if the topology evaluation value of the target sample meets the first preset condition, determining that the target sample is a second network topology.
Optionally, with reference to the first aspect and any one of the first to the fourth possible implementation manners of the first aspect, in a fifth possible implementation manner of the first aspect, after determining, according to the first network topology and the traffic model, a first topology evaluation value corresponding to the first network topology through a predetermined target topology evaluation, the method further includes: and if the first topology evaluation value meets the first preset condition, determining that the network topology of the target network is the first network topology.
A second aspect of the present application provides a network topology optimizing apparatus, including: the first acquisition module is used for acquiring a first network topology and a flow model of a target network; the first determining module is used for determining a first topology evaluation value corresponding to the first network topology through predetermined target topology evaluation according to the first network topology and the flow model acquired by the first acquiring module, wherein the target topology evaluation is determined based on a graph neural network and/or artificial topology evaluation, the graph neural network is obtained based on a plurality of topological element samples and index values of network performance indexes corresponding to the topological element samples through training, and the graph neural network is used for indicating the relationship between the network topology and the network performance indexes; and the second determining module is used for determining a second network topology according to the first network topology when the first topology evaluation value determined by the first determining module does not meet the first preset condition, wherein a second topology evaluation value corresponding to the second network topology meets the first preset condition, and the second network topology is used for adjusting the network topology of the target network so as to adjust the network topology of the target network to the second network topology.
Optionally, with reference to the second aspect, in a first possible implementation manner of the second aspect, the apparatus further includes: the second acquisition module is used for acquiring the neural network of the graph and/or the artificial topology evaluation; and the third determining module is used for determining the target topology evaluation according to the graph neural network and/or the artificial topology evaluation acquired by the second acquiring module.
Optionally, with reference to the first possible implementation manner of the second aspect, in a second possible implementation manner of the second aspect, the second obtaining module is configured to obtain a plurality of topology metadata samples and an index value of a network performance index corresponding to each topology metadata sample; and taking the topological element sample as input, and taking the index value of the network performance index corresponding to the topological element sample as output, and training the graph neural network.
Optionally, with reference to the second possible implementation manner of the second aspect, in a third possible implementation manner of the second aspect, the second obtaining module is configured to obtain a topology element size, where the topology element size is used to indicate a node number of a topology element sample; generating a plurality of topological element samples according to the topological element scale; and acquiring an index value of the network performance index of each topological element sample.
Optionally, with reference to the second aspect and any one of the first to third possible implementation manners of the second aspect, in a fourth possible implementation manner of the second aspect, the second determining module is configured to determine, according to the first network topology, a plurality of first samples when the first topology evaluation value determined by the first determining module does not satisfy a first preset condition, where the plurality of first samples are topology structures of which topology evaluation values satisfy a second preset condition among the plurality of first topology structures, and the plurality of first topology structures are topology structures that have connectivity properties of the first network topology among the plurality of topology structures randomly generated based on the first network topology; determining a plurality of second samples according to the plurality of first samples, wherein the plurality of second samples are topological structures with the connection property of the first network topology in the plurality of second topological structures, and the plurality of second topological structures are generated based on every two samples in the plurality of first samples; determining a plurality of third samples according to the plurality of second samples, wherein the plurality of third samples are topological structures with the connectivity property of the first network topology in the plurality of third topological structures, and the plurality of third topological structures are generated by randomly changing part or all of the second samples in the plurality of second samples; determining a target sample according to the plurality of third samples, wherein the target sample is the third sample with the smallest topological evaluation value in the plurality of third samples; and if the topology evaluation value of the target sample meets the first preset condition, determining that the target sample is a second network topology.
Optionally, with reference to the second aspect and any one of the first to the fourth possible implementation manners of the second aspect, in a fifth possible implementation manner of the second aspect, the second determining module is further configured to determine that the network topology of the target network is the first network topology when the first topology evaluation value determined by the first determining module satisfies the first preset condition.
A third aspect of the present application provides a network topology optimization device comprising a processor and a memory. The memory is used for storing computer readable instructions (or referred to as a computer program) which are read by the processor to implement the method provided by the foregoing first aspect or any one of the implementation manners of the first aspect.
In some implementations, the network topology optimization device also includes a transceiver to receive and transmit data.
A fourth aspect of the present application provides a computer storage medium, which may be non-volatile. The computer storage medium has stored therein computer readable instructions that, when executed by a processor, implement the method of the first aspect or any possible implementation of the first aspect.
A fifth aspect of the present application provides a computer storage medium, which may be non-volatile. The computer storage medium has stored therein computer readable instructions that, when executed by a processor, implement the method of the first aspect or any possible implementation of the first aspect.
According to the embodiment of the application, the topology evaluation is used for replacing simulation, the topology optimization problem of the network is converted into the mathematical optimization problem with the topology evaluation as the objective function, so that the result of the topology optimization is easy to quantify and evaluate, the calculation complexity in the topology optimization process can be obviously reduced, and the network topology optimization problem is solved systematically.
Drawings
Fig. 1 is a schematic architecture diagram of an application system provided in an embodiment of the present application;
fig. 2 is a schematic diagram of an embodiment of a network topology optimization method provided by an embodiment of the present application;
fig. 3 is a schematic diagram of another embodiment of a network topology optimization method provided in an embodiment of the present application;
FIG. 4 is a schematic diagram of an embodiment of a method for acquiring the neural network in the embodiment of the present application;
FIG. 5 is a schematic diagram of an embodiment of a GA topology iterative algorithm provided by an embodiment of the present application;
fig. 6 is a schematic structural diagram of a network topology optimizing device according to an embodiment of the present application;
fig. 7 is a schematic structural diagram of an apparatus provided in an embodiment of the present application.
Detailed Description
Embodiments of the present application will be described below with reference to the accompanying drawings, and it is to be understood that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. As can be known to those skilled in the art, with the evolution of computing frameworks and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.
The terms "first," "second," and the like in the description and in the claims of the present application and in the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It will be appreciated that the data so used may be interchanged under appropriate circumstances such that the embodiments described herein may be practiced otherwise than as specifically illustrated or described herein. Moreover, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or modules is not necessarily limited to those steps or modules explicitly listed, but may include other steps or modules not expressly listed or inherent to such process, method, article, or apparatus. The naming or numbering of the steps appearing in the present application does not mean that the steps in the method flow must be executed in the chronological/logical order indicated by the naming or numbering, and the named or numbered flow steps may be changed in execution order according to the technical purpose to be achieved, as long as the same or similar technical effects are achieved. The division of the modules presented in this application is a logical division, and in practical applications, there may be another division, for example, multiple modules may be combined or integrated into another system, or some features may be omitted, or not executed, and in addition, the shown or discussed coupling or direct coupling or communication connection between each other may be through some interfaces, and the indirect coupling or communication connection between the modules may be in an electrical or other similar form, which is not limited in this application. Moreover, the modules or sub-modules described as separate components may or may not be physically separated, may or may not be physical modules, or may be distributed in a plurality of circuit modules, and some or all of the modules may be selected according to actual needs to achieve the purpose of the present application.
The topology optimization of the current network is mainly realized based on experience rules and simulation technology, and a set of special simulation system needs to be constructed for simulating the service processing of a real network. Specifically, the simulation system takes network topology information, a traffic model, and network element configuration information as inputs, and simulates a real network to process a service flow through simulation, so as to obtain network performance indexes of the network, such as network performance indexes of load capacity, service flow transmission delay, transmission rate, congestion, and the like. For the network topology with network performance not up to the standard, the network topology needs to be adjusted according to expert experience, and then the adjusted network topology is subjected to simulation again until the final network performance index reaches the standard. The applicant finds that the method for realizing network topology optimization based on the simulation technology depends on the driving of simulation results, and has the following problems: firstly, in practical application, a large amount of network element configuration information irrelevant to network topology needs to be introduced into a simulation technology, and unnecessary difficulty is objectively increased for topology optimization of the network topology; meanwhile, because the configuration information of some network configurations is not easy to obtain, the result of simulation output is difficult to match with the output of an actual scene. Secondly, the simulation computation complexity is high, the simulation in a large-scale topological scene is usually above an hour level, and the method cannot be applied to a large-scale online optimization scene. Thirdly, the adjustment of the network topology is based on the experience rule, and the problem scene which can be solved by the method has strong limitation because the expansibility of the adjustment mode of the experience rule is poor; meanwhile, because empirical rules are difficult to quantify, the network topology adjusting scheme output by the method is difficult to quantitatively evaluate, and therefore an optimal or suboptimal network topology optimizing scheme is difficult to obtain.
Based on the above problems, embodiments of the present application provide a network topology optimization method, which converts a topology optimization problem of a network into a mathematical optimization problem using topology evaluation as a target function by using topology evaluation instead of simulation, so that a result of topology optimization is easy to quantify and evaluate, and the computational complexity in a topology optimization process can be significantly reduced, thereby systematically solving the network topology optimization problem. The embodiment of the application also provides a corresponding network topology optimization device and a system. The details will be described below separately.
The embodiment of the present application first provides an application system architecture for network topology optimization, please refer to fig. 1.
Fig. 1 is a schematic architecture diagram of an application system according to an embodiment of the present application.
Referring to fig. 1, an application system architecture for network topology optimization provided in the embodiment of the present application may include: a network management system 101, a network topology optimization tool 102, a Software Defined Network (SDN) controller 103, and one or more switches 104 controlled by SDN controller 103.
The network topology optimization method provided by the embodiment of the application can be implemented by the network topology optimization tool 102. It should be noted that the network topology optimization tool 102 in the embodiment of the present application may be a software system for implementing the network topology optimization method, and the software system may be directly deployed in the network management system 101; the network topology optimization tool 102 may also be independent of the network management system 101, and may be deployed in a third-party system to perform communication interfacing with the network management system 101, which is not limited in this embodiment of the present application.
In the embodiment of the present application, the network management system 101 refers to a communication network management system having functions of storing a network state and managing network devices. The network topology optimization tool 102 may obtain a network topology and a traffic model of a target network through the network management system 101, then perform topology optimization on the network topology of the target network based on the network topology and the traffic model, and finally output the optimized network topology to the network management system 101, so that the network management system 101 issues the optimized network topology to the SDN controller 103, and adjust a topology structure of the network topology through the SDN controller 103, thereby implementing topology optimization on the network topology of the target network. In this embodiment of the present application, the target network may be different types of networks such as a digital communication network, a wireless physical network, and a logical network, and the network type and scale of the target network are not limited in this embodiment of the present application.
Based on the application system, the network topology optimization method provided in the embodiment of the present application will be described in detail below, please refer to fig. 2.
Fig. 2 is a schematic diagram of an embodiment of a network topology optimization method provided in an embodiment of the present application.
Referring to fig. 2, an embodiment of a network topology optimization method provided in the embodiment of the present application may include:
201. a first network topology and a traffic model of a target network are obtained.
In the embodiment of the application, a network topology optimization tool firstly obtains a first network topology and a flow model of a target network through a network management system. In the embodiment of the application, the first network topology is used to indicate a topology structure of the current network topology of the target network, that is, topology information such as forwarding nodes, positions and connection relationships of the forwarding nodes, capacity data, and the like included in the current network topology of the target network. In the embodiment of the present application, the traffic model is used to indicate service flow information carried by a target network. Specifically, the traffic model may be a mathematical model describing the traffic flow, and may be obtained by summarizing the characteristics of the actual traffic flow.
202. According to the first network topology and the flow model, a first topology evaluation value corresponding to the first network topology is determined through predetermined target topology evaluation, wherein the target topology evaluation is determined based on a graph neural network and/or artificial topology evaluation, the graph neural network is obtained based on a plurality of topology element samples and index values of network performance indexes corresponding to the topology element samples through training, and the graph neural network is used for indicating the relationship between the network topology and the network performance indexes.
In the embodiment of the application, after acquiring the first network topology and the traffic model of the target network, the network topology optimization tool determines a first topology evaluation value corresponding to the first network topology through a predetermined target topology evaluation.
In the embodiment of the present application, the topology evaluation is a metric of the network topology, and is used to measure a certain capability or property of the network topology. The topology evaluation may be in the form of a function with the network topology as input. Therefore, the topology evaluation may also be referred to as a topology evaluation function or a topology evaluation index. When the input is determined, the value corresponding to the topology evaluation is the topology evaluation value. In the embodiment of the present application, the target topology evaluation is a topology evaluation for the target network, and the topology evaluation may be a topology evaluation for measuring network communication performance of the target network, such as bearer capacity, traffic flow transmission delay, transmission rate, congestion, and the like.
In the embodiment of the application, the target topology evaluation can be determined according to a graph neural network and/or a manual topology evaluation. The graph neural network refers to a deep learning neural network with input of network topology (or network topology graph). The neural network is obtained by training an index value of a network performance index corresponding to a large number of topological element samples and each topological element sample, and essentially can be a function which takes the network topology as input and the network performance index as output and is used for indicating the relationship between the network topology and the network performance index. In the embodiment of the application, the manual topology evaluation refers to the manual given topology evaluation, and the principle of the manual topology evaluation is based on mathematical theories such as graph theory and the like.
203. And if the first topology evaluation value does not meet the first preset condition, determining a second network topology according to the first network topology, wherein a second topology evaluation value corresponding to the second network topology meets the first preset condition, and the second network topology is used for network topology adjustment of the target network, so that the network topology of the target network is adjusted to the second network topology.
In the embodiment of the application, after determining a first topology evaluation value corresponding to a first network topology through a predetermined target topology evaluation, a network topology optimization tool first determines whether the first topology evaluation value meets a first preset condition. In the embodiment of the application, the setting of the first preset condition is used for judging whether the network performance corresponding to the network topology reaches the standard or not, when the topology evaluation value of the network topology meets the first preset condition, the network performance of the network topology reaches the standard, and when the topology evaluation value of the network topology does not meet the first preset condition, the network performance of the network topology does not reach the standard, and optimization is required.
In the embodiment of the application, when the network topology optimization tool judges that the first topology evaluation value does not meet the first preset condition, that is, when it is determined that the network performance of the first network topology does not meet the standard, the network topology optimization tool performs topology optimization of the network topology according to the first network topology, so that a second network topology in which the second topology evaluation value meets the first preset condition is obtained. The second network topology is used for outputting to the network management system, so that the network management system issues the second network topology to the SDN controller, and the SDN controller adjusts the network topology of the target network to the second network topology. In the embodiment of the present application, the network topology optimization tool may perform topology optimization of the network topology according to the principle of Genetic Algorithm (GA), which will be described in detail later in the present application.
According to the embodiment of the application, the topology evaluation is used for replacing simulation, the topology optimization problem of the network is converted into the mathematical optimization problem with the topology evaluation as the objective function, so that the result of the topology optimization is easy to quantify and evaluate, the calculation complexity in the topology optimization process can be obviously reduced, and the network topology optimization problem is solved systematically.
Fig. 3 is a schematic diagram of another embodiment of a network topology optimization method provided in an embodiment of the present application.
As shown in fig. 3, another embodiment of the network topology optimization method provided in the embodiment of the present application may include:
301. and obtaining a graph neural network and artificial topology evaluation.
In the embodiment of the application, before performing topology optimization on an online network topology, offline target topology evaluation generation needs to be performed first. In the process of generating the offline target topology evaluation, the network topology optimization tool needs to train the graph neural network and construct the artificial topology evaluation.
Firstly, introducing the artificial topology evaluation in the embodiment of the application:
the manual topology evaluation in the embodiment of the application refers to a part directly generated by expert input in the topology evaluation. Taking the topology optimization for reducing the time delay of a peer to peer (PoP) network as an example, the manual topology evaluation in the embodiment of the present application may be represented by the following formula: sigma i∈E (k 0 +k 1 l i +k 2 l i w i )+∑ i∈Ec k 3
Where E is the set of modified topological connections, i is the identity of the topological connection, N C A set of nodes in the topological nodes that satisfy a degree of nodes greater than a threshold is identified. k is a radical of 0 ,k 1 ,k 2 ,k 3 The parameters to be determined for the system, in particular,
∑ i∈E k 0 is the sum of the number of changes in the connection multiplied by k 0 Wherein k is 0 Is a parameter to be determined, and is relevant to a specific example;
∑ i∈E k 1 l i is the sum of the lengths of the modified edges of the connection multiplied by k 1 Wherein k is 1 Is a parameter to be determined, and is related to a specific example;
∑ i∈E k 2 l i w i is the capacity of the modified edge of the connection multiplied by k 2 Wherein k is 2 Is a parameter to be determined, and is related to a specific example;
∑ i∈Ec k 3 is the number of points with the node degree larger than a certain threshold value in the network topology multiplied by k 3 Wherein k is 3 Are parameters to be determined, and are relevant to the specific examples.
In practical application, an expert is required to input the parameter k 0 ,k 1 ,k 2 ,k 3 To confirm the concrete form of the expression, this section is therefore called artificial topology evaluation. The principle of artificial topology evaluation is based on classical measurement methods of topology in graph theory. Therein, sigma i∈ E k 0 、∑ i∈E k 1 l i Sum Σ i∈E k 2 l i w i Measuring the cost, sigma, of topology optimization from three aspects, namely the number of changes of the connection, the length of the changed edge of the connection and the capacity of the changed edge of the connection i∈Ec k 3 The structural complexity of the network topology is measured.
The graph neural network in the embodiment of the application is used for indicating the relationship between the network topology and the network performance index, and is obtained by training the index values of the network performance index corresponding to a plurality of topological element samples and each topological element sample. Specifically, the embodiment of the present application provides a method for obtaining a graph neural network, please refer to fig. 4.
As shown in fig. 4, the method for acquiring a graph neural network provided in the embodiment of the present application may include:
3011. and acquiring the scale of the topological element, wherein the scale of the topological element is used for indicating the number of nodes of the topological element sample.
In the embodiment of the application, a network topology optimization tool firstly determines the topology element scale of the topology elements used for the graph neural network training, wherein the topology element scale refers to the number of forwarding nodes contained in each topology element sample. In the embodiment of the present application, the size of the topology element may be determined and input by an expert, for example, the size of the topology element may be between 10 and 15.
3012. And generating a plurality of topological element samples according to the topological element scale.
In the embodiment of the application, after determining the scale of the topological element, the network topology optimization tool randomly generates a topological element sample by taking the scale of the topological element as input. In the embodiment of the present application, full connectivity is maintained, that is, a plurality of topology element samples having at least one communication path between every two forwarding nodes are learning samples for performing graph neural network training.
3013. And calculating the index value of the network performance index of each topological element sample.
In the embodiment of the application, after determining a plurality of topology element samples, the network topology optimization tool respectively obtains the index value of the network performance index corresponding to each topology element sample. It should be noted that, in the embodiment of the present application, the topology learning is a frame-type algorithm, and different results are obtained by selecting different indexes in practice. Taking the purpose of topology optimization to reduce communication delay as an example, the network performance index used can be represented by the following formula:
∑ i∈E (k 0 +k 1 l i +k 2 l i w i )+∑ i∈Ec k 3 -LLPD
the meaning of the first half of the above formula is explained in the manual topology evaluation section, and LLPD can be expressed by the following formula:
wherein, poP pairs refer to node pairs in the PoP network. In the PoP network topology, the shortest path exists between each node pair, and the shortest path consists of a plurality of connecting edges in the network topology; APA refers to the proportion of all the connections of this shortest path where there are alternative paths. Therefore, LLPD refers to the ratio of the number of pops of the APA pending parameter a to the total number of pops, which is used to measure the replaceability of a connection path in a network topology. In some applications, a may take on a value of 0.7.
3014. And training the graph neural network according to the plurality of topological element samples and the index value of the network performance index corresponding to each topological element sample.
In the embodiment of the application, after the network topology optimization tool obtains a plurality of topological element samples and the index value of the network performance index corresponding to each topological element sample, the training of the graph neural network is performed according to the plurality of topological element samples and the index value of the network performance index corresponding to each topological element sample. The graph neural network in the embodiment of the application comprises four layers, namely an input layer, a graph convolution neural network layer, a full connection layer and an output layer.
It should be noted that, for the topology optimization for reducing the delay of the PoP network, the manual topology evaluation and the acquisition method of the graph neural network may be different for other types of networks and different optimization effects. Therefore, in the practical application process, the network topology optimization tool may also acquire the artificial topology evaluation or the graph neural network by using other acquisition manners, which is not limited in the embodiment of the present application.
302. And determining target topology evaluation according to the graph neural network and the artificial topology evaluation.
In the embodiment of the application, after the graph neural network and the artificial topology evaluation are respectively obtained, the network topology optimization tool combines the graph neural network and the artificial topology evaluation to generate the target topology evaluation.
In the embodiment of the present application, the function expression of the target topology evaluation TopologyMetrics (τ) is as follows:
TopologyMetrics(τ)＝ArtificialMetrics(τ)+F(g,τ)
wherein: τ is a adjacency matrix of the network topology, i.e., a mathematical representation of the network topology; articicialMetrics (tau) is a functional expression of artificial topology evaluation; g (-) is a graph neural network model obtained by topological element learning, the essence of the graph neural network model is also a function expression taking network topology as input, and F (g, tau) is the expression of the network topology under g (-).
If the current network topology is tau, the topology is optimizedThe scale is larger than the scale of the topology elements, so that the current topology needs to be divided into the topology elements according to the scale of the topology elements
The specific form of F (g, τ) in the present embodiment may be:
303. a first network topology and a traffic model of a target network are obtained.
The embodiment of the present application can be understood with reference to step 201 in fig. 2, and is not described herein again.
It should be noted that, in the embodiment of the present application, the sequence of step 303 and the sequence from step 301 to step 302 is not limited.
304. And determining a first topology evaluation value corresponding to the first network topology through the predetermined target topology evaluation according to the first network topology and the traffic model.
The embodiment of the present application can be understood with reference to step 202 in fig. 2, and is not described herein again.
305. And if the first topology evaluation value does not meet the first preset condition, determining a second network topology according to the first network topology, wherein a second topology evaluation value corresponding to the second network topology meets the first preset condition, and the second network topology is used for adjusting the network topology of the target network so as to adjust the network topology of the target network to the second network topology.
In the embodiment of the application, after determining a first topology evaluation value corresponding to a first network topology through a predetermined target topology evaluation, a network topology optimization tool first determines whether the first topology evaluation value meets a first preset condition. The first preset condition is set for judging whether the network performance corresponding to the network topology meets the standard or not, when the topology evaluation value of the network topology meets the first preset condition, the network performance of the network topology meets the standard, and when the topology evaluation value of the network topology does not meet the first preset condition, the network performance of the network topology does not meet the standard, and optimization is needed.
In the embodiment of the present application, the network topology optimization tool may perform topology optimization of the network topology according to the principle of the GA algorithm. Specifically, the embodiment of the present application provides a GA topology iterative algorithm, which can implement optimization of a network topology, and please refer to fig. 5 for understanding.
Fig. 5 is a schematic diagram of an embodiment of a GA topology iteration algorithm provided in an embodiment of the present application.
As shown in fig. 5, an embodiment of the GA topology iterative algorithm provided in this application may include:
the network topology optimization tool can perform topology optimization of the network topology according to the following GA topology iterative algorithm:
3051. and initializing the network topology.
In the embodiment of the application, when the network topology optimization tool determines that the first topology evaluation value corresponding to the first network topology does not meet the first preset condition, the network topology is initialized for the first network topology.
Specifically, the network topology initialization may include the following two steps: (1) The network topology optimization tool randomly adds, reduces or changes a plurality of connecting edges on the basis of the first network topology, thereby generating a plurality of topological structures; (2) The network topology optimization tool determines a plurality of first topologies from the plurality of randomly generated topologies, wherein the plurality of first topologies have connectivity properties of the first network topology.
In the embodiment of the present application, the topology having connectivity property of the first network topology means that the topology has at least connectivity of the first network topology. Specifically, in addition to the most basic connectivity of the first network topology, the connectivity of the topology may be better than that of the first network topology, which is not limited in the embodiments of the present application.
3052. The topology is preserved.
In the embodiment of the application, after a plurality of first topological structures are obtained by initializing a first network topology through a network topology, a network topology optimization tool calculates a topology evaluation value corresponding to each of the plurality of first topological structures through target topology evaluation, then reserves the plurality of topological structures of which the topology evaluation values meet a second preset condition, and deletes the topological structures which do not meet the second preset condition, so as to obtain a plurality of first samples. For example, the second preset condition may refer to that the topology evaluation values in the plurality of first topologies are arranged in order from small to large, and belong to the top 10% of the topologies.
3053. Topological crossover.
In this embodiment of the present application, after determining a plurality of first samples, the network topology optimization tool performs topology intersection on the plurality of first samples, specifically, the method may include the following three steps: (1) The network topology optimization tool extracts two first samples in the plurality of first samples in a traversal mode to form a topology pair; (2) Generating a new topology structure according to each topology pair, wherein the generation rule can be that the connectivity between two nodes of the new topology structure is randomly replaced by the connectivity between two nodes corresponding to a certain first sample in the topology pair, so as to obtain a plurality of new topology structures; (3) And reserving the topology structure with the connectivity property of the first network topology in the plurality of new topology structures, thereby obtaining a plurality of second samples.
3054. And (6) topology variation.
In this embodiment of the application, after obtaining the plurality of second samples, the network topology optimization tool performs topology variation on the plurality of second samples, which may specifically include the following steps: (1) Randomly selecting part or all of the second samples from the plurality of second samples; (2) Randomly changing a plurality of edges of the part or all of the second samples so as to obtain a plurality of third topological structures; (3) And reserving a topology structure with the connectivity property of the first network topology in the third topologies to obtain a third samples.
3055. A target sample is determined.
In the embodiment of the application, after determining a plurality of third samples, the network topology optimization tool determines a target sample from the plurality of third samples, where the target sample is a third sample with a smallest topology evaluation value in the plurality of third samples.
Specifically, after obtaining a plurality of third samples, the network topology optimization tool calculates a topology evaluation value of each of the plurality of third samples, and then determines, as the target sample, the third sample with the smallest topology evaluation value.
In the embodiment of the application, if the topology evaluation value of the target sample meets the first preset condition, the target sample can be determined to be a second network topology, and the second network topology is output to the network management system, so that the network management system issues the second network topology to the SDN controller, and the SDN controller adjusts the network topology of the target network to the second network topology.
It should be noted that, when the topology evaluation value of the target sample does not satisfy the first preset condition, the network topology optimization tool performs iterative computation on the first network topology again according to the computation method in the above step 3051 to step 3055 until the final topology evaluation value of the target sample satisfies the first preset condition, so as to obtain a second network topology, and outputs the second network topology to the network management system, so that the network management system issues the second network topology to the SDN controller, and the SDN controller adjusts the network topology of the target network to the second network topology.
306. And if the first topology evaluation value meets the first preset condition, determining that the network topology of the target network is the first network topology.
In the embodiment of the application, if the network topology optimization tool determines the first topology evaluation value corresponding to the first network topology through the predetermined target topology evaluation and determines that the first topology evaluation value satisfies the first preset condition, the network topology of the target network is determined to be the first network topology, that is, the network topology of the target network does not need to be optimized.
According to the embodiment of the application, the topology evaluation is used for replacing simulation, the topology optimization problem of the network is converted into the mathematical optimization problem with the topology evaluation as the objective function, so that the result of the topology optimization is easy to quantify and evaluate, the calculation complexity in the topology optimization process can be obviously reduced, and the network topology optimization problem is solved systematically.
The network topology optimization method in the embodiment of the present application is introduced above, and a network topology optimization device provided in the embodiment of the present application is introduced next, please refer to fig. 6.
Fig. 6 is a schematic structural diagram of a network topology optimizing device according to an embodiment of the present application.
Referring to fig. 6, a network topology optimizing apparatus 60 provided in the embodiment of the present application includes:
a first obtaining module 601, configured to obtain a first network topology and a traffic model of a target network;
a first determining module 602, configured to determine, according to the first network topology and the traffic model acquired by the first acquiring module 601, a first topology evaluation value corresponding to the first network topology through a predetermined target topology evaluation, where the target topology evaluation is determined based on a graph neural network and/or an artificial topology evaluation, the graph neural network is obtained based on a plurality of topology element samples and an index value of a network performance index corresponding to each topology element sample, and the graph neural network is used to indicate a relationship between the network topology and the network performance index.
A second determining module 603, configured to determine a second network topology according to the first network topology when the first topology evaluation value determined by the first determining module 602 does not meet the first preset condition, where a second topology evaluation value corresponding to the second network topology meets the first preset condition, and the second network topology is used to adjust the network topology of the target network, so that the network topology of the target network is adjusted to the second network topology.
According to the network topology optimization device provided by the embodiment of the application, the topology optimization problem of the network is converted into the mathematical optimization problem taking the topology evaluation as the objective function by using the topology evaluation instead of simulation, so that the result of the topology optimization is easy to quantify and evaluate, the calculation complexity in the topology optimization process can be obviously reduced, and the network topology optimization problem is solved systematically.
Optionally, as an embodiment, the network topology optimizing device 60 provided in this embodiment of the present application may further include: a second obtaining module 604, configured to obtain the graph neural network and/or the artificial topology evaluation; a third determining module 605, configured to determine the target topology evaluation according to the graph neural network and/or the artificial topology evaluation acquired by the second acquiring module 604.
Optionally, as an embodiment, the second obtaining module 604 is configured to obtain the plurality of topology element samples and an index value of a network performance indicator corresponding to each topology element sample; and taking the topological element sample as input and the index value of the network performance index corresponding to the topological element sample as output to train the graph neural network.
Optionally, as an embodiment, the second obtaining module 604 is configured to obtain a size of a topology element, where the size of the topology element is used to indicate a number of nodes of the topology element sample; generating the plurality of topological element samples according to the topological element scale; and acquiring an index value of the network performance index of each topological element sample.
Optionally, as an embodiment, the second determining module 603 is configured to determine, when the first topology evaluation value determined by the first determining module does not satisfy a first preset condition, a plurality of first samples according to the first network topology, where the plurality of first samples are a topology structure of a plurality of first topology structures whose topology evaluation value satisfies a second preset condition, and the plurality of first topology structures are topology structures having connectivity properties of the first network topology among a plurality of topology structures randomly generated based on the first network topology; determining a plurality of second samples from the plurality of first samples, the plurality of second samples being a topology of a plurality of second topologies that has connectivity properties of a first network topology, the plurality of second topologies being generated based on every two samples of the plurality of first samples; determining a plurality of third samples according to the plurality of second samples, wherein the plurality of third samples are topological structures with the connectivity property of the first network topology in a plurality of third topological structures, and the plurality of third topological structures are generated by randomly changing part or all of the second samples in the plurality of second samples; determining a target sample according to the plurality of third samples, wherein the target sample is the third sample with the smallest topological evaluation value in the plurality of third samples; and if the topology evaluation value of the target sample meets the first preset condition, determining that the target sample is the second network topology.
Optionally, as an embodiment, the second determining module 603 is further configured to determine that the network topology of the target network is the first network topology when the first topology evaluation value determined by the first determining module 602 satisfies a first preset condition.
It should be understood that the first obtaining module 601, the first determining module 602, the second determining module 603, the second obtaining module 604, and the third determining module 605 in the embodiments of the present application may be implemented by a processor or a processor-related circuit component.
As shown in fig. 7, the embodiment of the present application further provides an apparatus 70, where the apparatus 70 includes a processor 710, a memory 720 and a transceiver 730, where the memory 720 stores instructions or programs, and the processor 710 is configured to execute the instructions or programs stored in the memory 720. When the instructions or programs stored in the memory 720 are executed, the processor 710 is configured to perform the operations performed by the first obtaining module 601, the first determining module 602, the second determining module 603, the second obtaining module 604, and the third determining module 605 in the above embodiments.
It should be understood that the device 70 according to the embodiment of the present application may correspond to a network topology optimization tool in the network topology optimization method according to the embodiment of the present application, and operations and/or functions of each module in the device 70 are respectively for implementing corresponding flows of each method in fig. 2 to fig. 5, and are not described herein again for brevity.
Optionally, an embodiment of the present application further provides a chip system, where the chip system includes a processor, and is used to support the device 70 to implement the network topology optimization method. In one possible design, the system-on-chip further includes a memory. The memory is used for storing program instructions and data necessary for the terminal equipment. The chip system may be formed by a chip, and may also include a chip and other discrete devices, which is not specifically limited in this embodiment of the present application.
It is understood that the Processor in the embodiments of the present Application may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a transistor logic device, a hardware component, or any combination thereof. The general purpose processor may be a microprocessor, but may be any conventional processor.
The method steps in the embodiments of the present application may be implemented by hardware, or may be implemented by software instructions executed by a processor. The software instructions may be comprised of corresponding software modules that may be stored in Random Access Memory (RAM), flash Memory, read-Only Memory (ROM), programmable ROM (PROM), erasable PROM (EPROM), electrically EPROM (EEPROM), registers, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. Of course, the storage medium may also be integral to the processor. The processor and the storage medium may reside in an ASIC. In addition, the ASIC may reside in a network device or a terminal device. Of course, the processor and the storage medium may reside as discrete components in a network device or a terminal device.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are performed in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer program or instructions may be stored in or transmitted over a computer-readable storage medium. The computer readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server that integrates one or more available media. The usable medium may be a magnetic medium, such as a floppy disk, a hard disk, a magnetic tape; or an optical medium, such as a DVD; it may also be a semiconductor medium, such as a Solid State Disk (SSD).
In the embodiments of the present application, unless otherwise specified or conflicting with respect to logic, the terms and/or descriptions in different embodiments have consistency and may be mutually cited, and technical features in different embodiments may be combined to form a new embodiment according to their inherent logic relationship.
In the present application, "at least one" means one or more, "a plurality" means two or more. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a alone, A and B together, and B alone, wherein A and B may be singular or plural. In the description of the text of the present application, the character "/" generally indicates that the former and latter associated objects are in an "or" relationship; in the formula of the present application, the character "/" indicates that the preceding and following related objects are in a relationship of "division".
It is to be understood that the various numbers or letter designations referred to in the embodiments of the present application are merely for descriptive convenience and are not intended to limit the scope of the embodiments of the present application. The sequence numbers of the above-mentioned processes do not mean the execution sequence, and the execution sequence of the processes should be determined by their functions and inherent logic.
The network topology optimization method, device and system provided by the embodiment of the present application are introduced in detail, a specific example is applied in the description to explain the principle and the implementation of the present application, and the description of the embodiment is only used to help understand the method and the core idea of the present application; meanwhile, for a person skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.
Claims (14)
1. A method for optimizing network topology, comprising:
acquiring a first network topology and a flow model of a target network;
determining a first topology evaluation value corresponding to the first network topology through a predetermined target topology evaluation according to the first network topology and the traffic model, wherein the target topology evaluation is determined based on a graph neural network and an artificial topology evaluation, the graph neural network is obtained based on a plurality of topological element samples and index values of network performance indexes corresponding to the topological element samples through training, and the graph neural network is used for indicating a relation between the network topology and the network performance indexes;
and if the first topology evaluation value does not meet a first preset condition, determining a second network topology according to the first network topology, wherein a second topology evaluation value corresponding to the second network topology meets the first preset condition, and the second network topology is used for network topology adjustment of the target network, so that the network topology of the target network is adjusted to the second network topology.
2. The method according to claim 1, wherein before determining a first topology evaluation value corresponding to the first network topology through a predetermined target topology evaluation according to the first network topology and the traffic model, the method further comprises:
acquiring the graph neural network and/or the artificial topology evaluation;
and determining the target topology evaluation according to the graph neural network and/or the artificial topology evaluation.
3. The method of claim 2, wherein the obtaining the graph neural network comprises:
acquiring the plurality of topological meta-samples and index values of network performance indexes corresponding to the topological meta-samples;
and taking the topological element sample as input and the index value of the network performance index corresponding to the topological element sample as output to train the graph neural network.
4. The method according to claim 3, wherein the obtaining the index values of the plurality of topology meta-samples and the network performance index corresponding to each topology meta-sample comprises:
acquiring a topological element scale, wherein the topological element scale is used for indicating the number of nodes of the topological element sample;
generating a plurality of topological element samples according to the topological element scale;
and acquiring an index value of the network performance index of each topological element sample.
5. The method according to any one of claims 1 to 4, wherein if the first topology evaluation value does not satisfy a preset condition, determining a second network topology according to the first network topology includes:
determining a plurality of first samples according to the first network topology, wherein the plurality of first samples are topology structures of which topology evaluation values meet a second preset condition in the plurality of first topology structures, and the plurality of first topology structures are topology structures which have connectivity properties of the first network topology in the plurality of topology structures randomly generated based on the first network topology;
determining a plurality of second samples from the plurality of first samples, the plurality of second samples being ones of a plurality of second topologies that possess connectivity properties of the first network topology, the plurality of second topologies being generated based on every two samples of the plurality of first samples;
determining a plurality of third samples according to the plurality of second samples, wherein the plurality of third samples are topological structures with the connectivity property of the first network topology, and the plurality of third topological structures are generated by randomly changing part or all of the second samples in the plurality of second samples;
determining a target sample according to the plurality of third samples, wherein the target sample is the third sample with the smallest topological evaluation value in the plurality of third samples;
and if the topology evaluation value of the target sample meets the first preset condition, determining that the target sample is the second network topology.
6. The method according to claim 1, wherein after determining a first topology evaluation value corresponding to the first network topology by a predetermined target topology evaluation according to the first network topology and the traffic model, the method further comprises:
and if the first topology evaluation value meets the first preset condition, determining that the network topology of the target network is the first network topology.
7. An apparatus for optimizing a network topology, comprising:
the first acquisition module is used for acquiring a first network topology and a flow model of a target network;
a first determining module, configured to determine, according to the first network topology and the traffic model acquired by the first acquiring module, a first topology evaluation value corresponding to the first network topology through a predetermined target topology evaluation, where the target topology evaluation is determined based on a graph neural network and an artificial topology evaluation, the graph neural network is obtained based on a plurality of topology element samples and an index value of a network performance index corresponding to each topology element sample, and the graph neural network is used for indicating a relationship between a network topology and a network performance index;
a second determining module, configured to determine a second network topology according to the first network topology when the first topology evaluation value determined by the first determining module does not meet a first preset condition, where a second topology evaluation value corresponding to the second network topology meets the first preset condition, and the second network topology is used for network topology adjustment of the target network, so that the network topology of the target network is adjusted to the second network topology.
8. The apparatus of claim 7, further comprising:
the second acquisition module is used for acquiring the graph neural network and/or the artificial topology evaluation;
and the third determining module is used for determining the target topology evaluation according to the graph neural network and/or the artificial topology evaluation acquired by the second acquiring module.
9. The apparatus of claim 8,
the second obtaining module is configured to obtain the plurality of topology meta-samples and an index value of a network performance index corresponding to each topology meta-sample; and taking the topological element sample as input and the index value of the network performance index corresponding to the topological element sample as output to train the graph neural network.
10. The apparatus of claim 9,
the second obtaining module is configured to obtain a topology element scale, where the topology element scale is used to indicate the number of nodes in the topology element sample; generating the plurality of topological element samples according to the topological element scale; and acquiring an index value of the network performance index of each topology meta-sample.
11. The apparatus according to any one of claims 7 to 10,
the second determining module is configured to determine, when the first topology evaluation value determined by the first determining module does not satisfy a first preset condition, a plurality of first samples according to the first network topology, where the plurality of first samples are topology structures whose topology evaluation values satisfy a second preset condition among a plurality of first topology structures, and the plurality of first topology structures are topology structures that have connectivity properties of the first network topology among a plurality of topology structures randomly generated based on the first network topology; determining a plurality of second samples from the plurality of first samples, the plurality of second samples being ones of a plurality of second topologies that possess connectivity properties of the first network topology, the plurality of second topologies being generated based on every two samples of the plurality of first samples; determining a plurality of third samples according to the plurality of second samples, wherein the plurality of third samples are topological structures with connectivity properties of the first network topology in a plurality of third topological structures, and the plurality of third topological structures are generated by randomly changing part or all of the second samples in the plurality of second samples; determining a target sample according to the plurality of third samples, wherein the target sample is the third sample with the smallest topological evaluation value in the plurality of third samples; and if the topology evaluation value of the target sample meets the first preset condition, determining that the target sample is the second network topology.
12. The apparatus of claim 7,
the second determining module is further configured to determine that the network topology of the target network is the first network topology when the first topology evaluation value determined by the first determining module satisfies a first preset condition.
13. A computer device comprising a processor, a memory;
the memory is for storing computer readable instructions or a computer program, the processor being for reading the computer readable instructions to implement the method of any one of claims 1-6.
14. A computer-readable storage medium comprising computer program instructions which, when run on a computer, cause the computer to perform the method of any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911368600.2A CN113054651B (en) | 2019-12-26 | 2019-12-26 | Network topology optimization method, device and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911368600.2A CN113054651B (en) | 2019-12-26 | 2019-12-26 | Network topology optimization method, device and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113054651A CN113054651A (en) | 2021-06-29 |
| CN113054651B true CN113054651B (en) | 2023-03-31 |
Family
ID=76505489
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911368600.2A Active CN113054651B (en) | 2019-12-26 | 2019-12-26 | Network topology optimization method, device and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113054651B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113691454B (en) * | 2021-10-25 | 2022-02-08 | 广州鲁邦通物联网科技股份有限公司 | Industrial Internet of things data acquisition and management method and system |
| CN114500333A (en) * | 2021-12-29 | 2022-05-13 | 中国电信股份有限公司 | Network architecture measurement method and device, electronic equipment and storage medium |
| CN114567563B (en) * | 2022-03-31 | 2024-04-12 | 北京邮电大学 | Training method of network topology model, and reconstruction method and device of network topology |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105743691A (en) * | 2014-12-29 | 2016-07-06 | 瞻博网络公司 | Network topology optimization |
| CN108370370A (en) * | 2015-12-14 | 2018-08-03 | 西门子股份公司 | System and method for passively assessing industrial security boundary |
| CN109685279A (en) * | 2018-12-29 | 2019-04-26 | 广东电网有限责任公司清远英德供电局 | A kind of Complicated Distribution Network PQM optimization method based on topology degradation |
| CN110149225A (en) * | 2019-05-13 | 2019-08-20 | 中国联合网络通信集团有限公司 | Network topology structure appraisal procedure and device |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2536606A (en) * | 2014-11-28 | 2016-09-28 | Aria Networks Ltd | Network topology optimization |
| GB2536608A (en) * | 2014-11-28 | 2016-09-28 | Aria Networks Ltd | Optimizing the topology of a network with variable traffic demands |
-
2019
- 2019-12-26 CN CN201911368600.2A patent/CN113054651B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105743691A (en) * | 2014-12-29 | 2016-07-06 | 瞻博网络公司 | Network topology optimization |
| CN108370370A (en) * | 2015-12-14 | 2018-08-03 | 西门子股份公司 | System and method for passively assessing industrial security boundary |
| CN109685279A (en) * | 2018-12-29 | 2019-04-26 | 广东电网有限责任公司清远英德供电局 | A kind of Complicated Distribution Network PQM optimization method based on topology degradation |
| CN110149225A (en) * | 2019-05-13 | 2019-08-20 | 中国联合网络通信集团有限公司 | Network topology structure appraisal procedure and device |
Non-Patent Citations (1)
| Title |
|---|
| 非均匀分簇无线传感器网络拓扑控制仿真;刘志龙等;《计算机仿真》;20190415(第04期);260-264页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113054651A (en) | 2021-06-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113054651B (en) | Network topology optimization method, device and system | |
| US20190394132A1 (en) | System and Method for Network Slicing for Service-Oriented Networks | |
| CN109413175B (en) | Information processing method and device and electronic equipment | |
| US20120072475A1 (en) | Measuring "closeness" in a network | |
| US20170012848A1 (en) | Parallelized network traffic flow availability simulation using stochastic process and traffic engineering algorithms | |
| CN113422695B (en) | Optimization method for improving robustness of topological structure of Internet of things | |
| CN113572697A (en) | Load balancing method based on graph convolution neural network and deep reinforcement learning | |
| CN110162692B (en) | User label determination method and device, computer equipment and storage medium | |
| CN115358487A (en) | Federal learning aggregation optimization system and method for power data sharing | |
| CN113992259A (en) | Method for constructing time slot resource expansion diagram | |
| CN111181792A (en) | SDN controller deployment method and device based on network topology and electronic equipment | |
| US20170371936A1 (en) | Sample size estimator | |
| Tembine | Mean field stochastic games: Convergence, Q/H-learning and optimality | |
| CN104168158A (en) | Betweenness-centrality-based method of performing network analysis on multidimensional network | |
| KR101806628B1 (en) | Method for constructing fused regression network and fused analysis system thereof | |
| CN111598390B (en) | Method, device, equipment and readable storage medium for evaluating high availability of server | |
| CN114116740A (en) | Method and device for determining contribution degree of participants in joint learning | |
| CN114298319A (en) | Method and device for determining joint learning contribution value, electronic equipment and storage medium | |
| CN114401192A (en) | Multi-SDN controller collaborative training method | |
| CN113408140A (en) | Power communication network capacity expansion simulation verification method, device, equipment and medium | |
| CN109981327B (en) | Method and system for predicting traffic | |
| CN108768735B (en) | Bipartite graph sampling method and device for test bed topological structure | |
| CN112925197A (en) | Finite time-based dichotomous consistent tracking method for multi-agent system | |
| CN111917589A (en) | Power communication network resource backup method and related device | |
| KR102067013B1 (en) | Device for collective machine learning of things and method thereof |
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 |