CN111327494B - Multi-domain SDN network flow situation assessment method and system - Google Patents
Multi-domain SDN network flow situation assessment method and system Download PDFInfo
- Publication number
- CN111327494B CN111327494B CN202010092967.2A CN202010092967A CN111327494B CN 111327494 B CN111327494 B CN 111327494B CN 202010092967 A CN202010092967 A CN 202010092967A CN 111327494 B CN111327494 B CN 111327494B
- Authority
- CN
- China
- Prior art keywords
- index
- domain
- network traffic
- sdn network
- weight
- 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
- 238000000034 method Methods 0.000 title claims abstract description 51
- 230000008859 change Effects 0.000 claims abstract description 27
- 238000004364 calculation method Methods 0.000 claims abstract description 24
- 241000282461 Canis lupus Species 0.000 claims description 50
- 241000282421 Canidae Species 0.000 claims description 25
- 239000011159 matrix material Substances 0.000 claims description 19
- 238000009826 distribution Methods 0.000 claims description 14
- 238000005457 optimization Methods 0.000 claims description 13
- 230000008569 process Effects 0.000 claims description 10
- 230000006870 function Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 7
- 238000000513 principal component analysis Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 description 9
- 238000011156 evaluation Methods 0.000 description 8
- 238000004891 communication Methods 0.000 description 7
- 238000007726 management method Methods 0.000 description 7
- 230000001186 cumulative effect Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000009827 uniform distribution Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0823—Errors, e.g. transmission errors
- H04L43/0829—Packet loss
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0852—Delays
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
- H04L43/0882—Utilisation of link capacity
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
- H04L43/0888—Throughput
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
- H04L43/0894—Packet rate
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate
Landscapes
- Engineering & Computer Science (AREA)
- Environmental & Geological Engineering (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a multi-domain SDN network traffic situation assessment method and a system, comprising the following steps: acquiring a multi-domain SDN network traffic situation influencing factor, and constructing a multi-domain SDN network traffic situation index set; calculating an index initial weight; determining a weight change interval based on the index initial weight; optimizing the initial weight of the index based on the weight change interval to obtain the optimal weight of the index; weighting calculation is carried out on the index optimal weight and the index value to obtain a multi-domain SDN network flow situation value; and determining the current period multi-domain SDN network traffic state by combining the multi-domain SDN network traffic situation level table and the current period multi-domain SDN network traffic situation value.
Description
Technical Field
The disclosure relates to the technical field of network traffic situation assessment, in particular to a multi-domain SDN network traffic situation assessment method and system.
Background
The statements in this section merely mention background art related to the present disclosure and do not necessarily constitute prior art.
With the continuous increase of the scale of the internet, the number of users is continuously increased, and the traditional network architecture exposes various problems, such as distributed deployment, poor controllability, poor mobility and the like. These drawbacks severely limit the rapid development of network technology, and therefore, a network architecture that breaks this situation is needed. Software-defined networking (SDN) has also emerged. The essence of a software-defined network is to implement decoupling of the control plane and the data plane so that the network administrator can control the entire network through software programming. The advantages compared with the traditional network are mainly represented by the following two aspects:
1) SDN separates the control plane and data plane of the current network, the data plane is simply forwarded by the actual network equipment (such as a switch and a router), and the control plane is regulated and controlled by a logically centralized management controller (or called a network operating system), so that policy execution, optimization and network configuration are simplified.
2) SDN improves the programmability of the network system by providing a powerful and extensible programming interface for developers. For the upper layer, a developer does not need to pay attention to the hardware details of the bottom layer, and only needs to flexibly design own application on the basis of the northbound interface; for the controller, different network devices are compatible through a south-oriented interface, so that logic of an upper layer application can be realized. SDN separates the forwarding function and the control function of the network, and has more obvious advantages in the aspects of data forwarding, network management and the like.
Network traffic management has been the focus of SDN research. With the continuous development of the SDN, frequent information interaction causes the network traffic to proliferate, increases the complexity of the network operation condition, and researchers propose to apply traffic management technologies such as traffic engineering, anomaly detection, traffic statistics and the like to the SDN. However, it has drawbacks such as limitations and non-comprehensiveness in the description of the flow. Network traffic situation assessment SDN network traffic states enhance the global expressive power of network resources. The network traffic situation assessment refers to extracting, fusing and analyzing a large amount of obtained heterogeneous data on the premise of sensing relevant network traffic elements in a certain space-time environment, and quantifying the current network traffic state according to a specific situation assessment algorithm. The network traffic situation assessment can not only avoid the problem of single data source, but also make up the defect that the acquired information is mutually independent and cannot be effectively associated and analyzed, and in addition, the network traffic situation assessment method can assess the traffic state from the global angle and calculate the current network traffic situation value by utilizing an advanced calculation technology. Network traffic situation assessment SDN network traffic state enables a network manager to know the current traffic situation in time by analyzing network global traffic information, so that the network long-term running state is represented, important basis is provided for daily network maintenance and later network resource allocation, and the network management is very important.
In the process of implementing the present disclosure, the inventor finds that the following technical problems exist in the prior art: at present, network traffic situation assessment is focused on at a primary stage; the traffic situation of the multi-domain SDN cannot be accurately estimated.
Disclosure of Invention
The purpose of the present disclosure is to solve the above problems, and provide a method and a system for evaluating traffic situation of a multi-domain SDN network, so as to accurately evaluate traffic situation of the multi-domain SDN network.
In order to achieve the above purpose, the present disclosure adopts the following technical scheme:
in a first aspect, the present disclosure provides a multi-domain SDN network traffic situation assessment method;
a multi-domain SDN network traffic situation assessment method comprises the following steps:
acquiring a multi-domain SDN network traffic situation influencing factor, and constructing a multi-domain SDN network traffic situation index set;
calculating an index initial weight; determining a weight change interval based on the index initial weight;
optimizing the initial weight of the index based on the weight change interval to obtain the optimal weight of the index;
weighting calculation is carried out on the index optimal weight and the index value to obtain a multi-domain SDN network flow situation value;
and determining the current period multi-domain SDN network traffic state by combining the multi-domain SDN network traffic situation level table and the current period multi-domain SDN network traffic situation value.
In a second aspect, the present disclosure provides a multi-domain SDN network traffic situation assessment server;
a multi-domain SDN network traffic situation assessment server comprising:
an index set construction module configured to: acquiring a multi-domain SDN network traffic situation influencing factor, and constructing a multi-domain SDN network traffic situation index set;
an index initial weight calculation module configured to: calculating an index initial weight; determining a weight change interval based on the index initial weight;
an index optimal weight calculation module configured to: optimizing the initial weight of the index based on the weight change interval to obtain the optimal weight of the index;
a weight calculation module configured to: weighting calculation is carried out on the index optimal weight and the index value to obtain a multi-domain SDN network flow situation value;
a multi-domain SDN network traffic state output module configured to: and determining the current period multi-domain SDN network traffic state by combining the multi-domain SDN network traffic situation level table and the current period multi-domain SDN network traffic situation value.
In a third aspect, the present disclosure also provides an electronic device comprising a memory and a processor, and computer instructions stored on the memory and running on the processor, which when executed by the processor, perform the method of the first aspect.
In a fourth aspect, the present disclosure also provides a computer readable storage medium storing computer instructions which, when executed by a processor, perform the method of the first aspect.
In a fifth aspect, the present disclosure provides a multi-domain SDN network traffic situation assessment system;
a multi-domain SDN network traffic situation assessment system comprising: a multi-domain SDN network traffic situation assessment server according to the third aspect; the system further comprises:
a multi-domain SDN network, the multi-domain SDN network comprising a plurality of SDN networks;
each SDN network is communicated with a corresponding intra-domain controller, and the intra-domain controller is responsible for collecting network traffic situation influencing factors in the corresponding SDN network; the intra-domain controller uploads the acquired network flow situation influencing factors to a database;
network traffic situation influencing factors between the SDN network and other SDN networks are collected by an inter-domain controller, and the inter-domain controller also uploads the collected network traffic situation influencing factors to a database;
the multi-domain SDN network traffic situation assessment server extracts network traffic situation influencing factors from the database, processes the extracted network traffic situation influencing factors and determines the current period multi-domain SDN network traffic state.
The beneficial effects of the present disclosure are:
(1) A multi-domain SDN network traffic situation assessment method is provided. Calculating initial index weight, determining a weight change interval on the basis of obtaining an index initial weight result, optimizing the index weight, calculating the optimal weight of each index, obtaining a corresponding flow situation value of the SDN network by adopting a weighted comprehensive evaluation method, and determining the network flow state in the current period by combining a multi-domain SDN network flow situation level table.
(2) And designing a network traffic situation assessment model by combining the characteristics of the multi-domain SDN. The multi-domain SDN network adopts a vertical architecture, a database is introduced as a main mode of information storage for saving network situation index information and historical processing results with large data volume, and meanwhile, a network situation assessment server is introduced for assessing network traffic situations in the current period for reducing communication expenses and calculation expenses.
(3) And analyzing flow situation influence factors of the multi-domain SDN network, and constructing a flow situation assessment index system. In order to comprehensively reflect the traffic situation of the SDN network, situation influencing factors are analyzed from four aspects of an SDN application plane, a control plane, a data plane and a transmission process, and an SDN network traffic situation index set is constructed.
Drawings
Fig. 1 is a multi-domain SDN network traffic situation assessment model diagram according to an embodiment of the disclosure;
fig. 2 is a multi-domain SDN network traffic situation index system according to an embodiment of the disclosure;
fig. 3 is a flowchart of initial weight calculation based on PCA traffic situation index according to an embodiment of the present disclosure;
FIG. 4 is a Gaussian distribution diagram of a first embodiment of the present disclosure;
fig. 5 is a flowchart of a gray wolf algorithm-based weight optimization according to a first embodiment of the present disclosure.
Detailed Description
The disclosure is further described below with reference to the drawings and examples.
An embodiment one provides a multi-domain SDN network traffic situation assessment method;
a multi-domain SDN network traffic situation assessment method comprises the following steps:
s1: acquiring a multi-domain SDN network traffic situation influencing factor, and constructing a multi-domain SDN network traffic situation index set;
s2: calculating an index initial weight; determining a weight change interval based on the index initial weight;
s3: optimizing the initial weight of the index based on the weight change interval to obtain the optimal weight of the index;
s4: weighting calculation is carried out on the index optimal weight and the index value to obtain a multi-domain SDN network flow situation value;
s5: and determining the current period multi-domain SDN network traffic state by combining the multi-domain SDN network traffic situation level table and the current period multi-domain SDN network traffic situation value.
Further, the acquiring the traffic situation influencing factors of the multi-domain SDN network is acquired from four aspects of an application plane, a control plane, a data plane and a transmission process of the multi-domain SDN network.
Further, the calculating the index initial weight is to use a principal component analysis algorithm to calculate the index initial weight.
It should be understood that, when the main component analysis algorithm is used to calculate the initial weight of the index, when evaluating the traffic situation of the SDN network, the contribution degree of each index to the traffic situation is different, and different weight grades need to be assigned to the evaluation index according to the importance degree of each index factor.
Further, determining a weight change interval based on the index initial weight; the weight change interval is determined by using Gaussian distribution.
Further, the initial weight of the index is optimized based on the weight change interval to obtain the optimal weight of the index; the method is to optimize the initial weight of the index by adopting a gray wolf optimization algorithm to obtain the optimal weight of the index.
It should be understood that the set of traffic situation indicators of the constructed multi-domain SDN network is noted as:
X t ＝{x 1 (t),x 2 (t),...,x m (t)}，
wherein ,Xt A situation index set, x representing the moment t i (t) represents each individual situation index value at time t, m being the number of situation indices in the set, i=1, 2.
It should be understood that for the research of the SDN traffic situation index set, the selected index should have stronger integrity, reflect the overall running state of the network as much as possible, and consider the influence of the data from multiple sources on the network traffic situation assessment result. According to the multi-domain SDN architecture, situation indexes are respectively extracted from an application plane, a control plane, a data plane and a transmission process, and particularly multi-source data consideration is carried out from traffic conditions and equipment states of an application program, a switch, a controller and a link, wherein a multi-domain SDN network traffic situation index set is shown in fig. 2.
It should be appreciated that as the network scale increases, due to limited performance of controllers, a single domain SDN having a single controller is difficult to meet network requirements, and multiple controllers working cooperatively need to be deployed to form a multi-domain SDN network. The monitoring and analysis of network traffic are always key points of SDN research, and in order to accurately and comprehensively evaluate the condition of multi-domain SDN network traffic, the disclosure provides a network traffic situation awareness method for multi-domain SDN.
The multi-domain SDN network traffic situation influencing factors acquired from an application plane of the multi-domain SDN network comprise: application presence rate, controller presence rate, switch presence rate, link presence rate, or REST request rate.
The multi-domain SDN network traffic situation influencing factors acquired from a control plane of the multi-domain SDN network comprise: cross-domain request rate, packet in rate, or controller presence rate.
The multi-domain SDN network traffic situation influencing factors acquired from the data plane of the multi-domain SDN network comprise: switch packet reception, switch packet transmission rate, or switch online rate.
The multi-domain SDN network traffic situation influencing factors obtained from the transmission process of the multi-domain SDN network comprise: bit error rate, bandwidth utilization, maximum transmission time delay, packet loss rate, throughput, or link presence rate.
The indexes related to the multi-domain SDN flow situation index set comprise:
(1) Application/controller/switch/link presence ratio: the number of active applications/controllers/switches/links in the current network is the ratio of the total number of applications/controllers/switches/links.
(2) REST request rate: and the REST packet number of all application programs interacted with the controller in unit time. The controller provides REST API for the application plane, SDN application communicates with the controller by sending REST request, and monitors and optimizes the whole network.
(3) Cross-domain request rate: inter-domain controller and intra-domain controller interaction cross-domain request packet number per unit time. The intra-domain controller is responsible for communication and management of an intra-domain network managed by the intra-domain controller, when a cross-domain request arrives, the intra-domain controller sends the cross-domain request to the inter-domain controller, and the inter-domain controller calculates a cross-domain path based on global topology information.
(4) Packet_in rate: the packet_in Packet number of all switches and controllers interacting in a unit time is shown as formula (1). After the new flow arrives at the switch, if there is no flow item matched with the flow item on the switch, the switch forwards the data Packet to the controller for processing through the packet_in message.
Wherein T is a sampling period; packet_in-num is the number of packet_in packets in the period.
(5) Switch packet transmission/reception rate: the number of packets sent/received per unit time for all ports of the switch.
(6) Throughput: number of successfully transmitted packets per unit time. The greater the throughput, the better the quality of service of the network.
(7) Packet loss amount: the number of lost packets per unit time. The packet loss can represent the problems in the network to a certain extent, and the smaller the packet loss rate is, the better the service quality of the network is.
(8) Maximum transmission delay: the maximum value of the elapsed time of the data packet from the source to the sink. The transmission quality of the network can be well characterized by measuring the maximum time delay, and when the time delay is too large, the network is indicated to be congested on certain links or nodes, and the loss of the data packet is more easily caused by the large time delay.
(9) Bandwidth utilization: the average value of the ratio of the bandwidth used by the link to the bandwidth in the sampling period. For characterizing the ability of the link to transmit data.
(10) Error amount: number of erroneously transmitted packets per unit time. The smaller the bit error amount, the better the quality of service of the network.
It should be understood that in the dynamic operation process of the multi-domain SDN network, the explanation of the global state of the network cannot be performed through a single index, such as throughput, delay, packet loss rate, and the like. In order to evaluate the traffic situation of the multi-domain SDN network, after the network traffic situation index set is established, data mining is further needed to quantitatively describe the current traffic state of the network.
Further, the index optimal weight and the index value are subjected to weighted calculation to obtain a multi-domain SDN network flow situation value; the method is specifically to obtain a multi-domain SDN network flow situation value by adopting a weighted comprehensive evaluation method.
It should be understood that in S5, the multi-domain SDN network traffic state of the current period is determined by combining the multi-domain SDN network traffic state level table and the current period multi-domain SDN network traffic state value; the method specifically comprises the following steps:
under different network operation states, the network traffic situation corresponds to different division results. For SDN network traffic scenarios, a class 4 network morphology may be classified as shown in table 1. And determining the flow state of the SDN according to the flow situation value of the multi-domain SDN at the current moment.
Table 1 Multi-Domain SDN network traffic situation level Table
It should be understood that the calculating the index initial weight by using the principal component analysis algorithm specifically includes:
principal component analysis is a method of mathematically reducing the dimensions of data. The basic idea is to try to combine a plurality of indexes X with certain relativity 1 ，X 2 ，…，X P (e.g. P indexes) are recombined into a group of less number of mutually uncorrelated comprehensive indexes F m Instead of the original index. Then the comprehensive index should be extracted to reflect the original variable X to the greatest extent P The represented information can ensure that the new indexes are kept independent of each other (the information is not overlapped).
Set F 1 Principal component index representing the first linear combination of primary variables:
F 1 ＝a 11 X 1 +a 21 X 2 +...+a p1 X p ,
from mathematical knowledge, the amount of information extracted by each principal component can be measured by its variance, which variance Var (F 1 ) The larger the representation F 1 The more information is contained. It is often desirable for the first principal component F 1 The amount of information contained is maximum, so F is selected from all linear combinations 1 Should be X 1 ，X 2 ，…，X P The variance is the largest among all the linear combinations of (a), so is called F 1 Is the first principal component.
If the first principal component is insufficient to represent the information of the original P indexes, the second principal component index F is selected 2 F to effectively reflect the original information 1 Existing information need not be present in F 1 In, i.e. F 2 And F is equal to 1 To remain independent and uncorrelated, expressed in mathematical language, is its covariance Cov (F 1 ,F 1 ) =0, so F 2 Is with F 1 Uncorrelated X 1 ，X 2 ，…，X P The variance is the largest among all the linear combinations of (a), so is called F 2 F as the second main component, and so on 1 、F 2 、……、F m Is the original variable index X 1 ，X 2 ，…，X P The first, second, … … and m-th main components have the corresponding relation shown in the formula (2).
Further, the calculating the index initial weight by using the principal component analysis algorithm specifically includes:
s201: constructing a sample set; the sample set comprises a plurality of indexes of which the initial weights of the indexes are to be calculated;
s202: carrying out standardization treatment on the sample set;
s203: calculating the correlation coefficient between each index after the normalization processing to obtain a correlation coefficient matrix:
s204: calculating the characteristic variance contribution rate of each component according to the correlation coefficient matrix;
s205: k components with the set threshold value according to the characteristic variance contribution rate are taken as main components;
s206: calculating a component load matrix of each main component and each index according to the obtained plurality of main components;
s207: calculating the comprehensive score coefficient of each index on the main component according to the component load matrix of each main component and each index;
s208: and calculating the initial weight of each index according to the comprehensive score coefficient of each index on the main component.
Further, the specific step S201 includes: a sample data set X, which contains n sample variable indexes, each index collects m pieces of sample data, and for positive indexes in the sample data as original values and negative indexes as opposite numbers of the index values, the sample set is represented by a matrix of formula (3):
further, the step S202: the sample set is subjected to standardization processing, and the specific steps comprise:
the Z-score is adopted to carry out standardization processing on the data, and the method has the advantages that the information of the sample data is fully utilized, the standardized data can be objectively obtained, and a standardized formula is shown as a formula (4).
wherein ,
for the average value of each index in the sample set, the average value of the jth index is shown as the formula (5)
Further, the step S203: calculating the correlation coefficient between each index to obtain a correlation coefficient matrix; the method comprises the following specific steps:
the correlation coefficient matrix of the sample can express the correlation among the indexes, and the correlation coefficient matrix is shown as a formula (6):
wherein ,rij The correlation coefficient representing the ith and jth index variables is defined as shown in formula (7), and r ij ＝r ji ，r ii ＝1，
For the mean value of each sample data, the mean value of the ith sample data is shown as formula (8):
further, the step S204: calculating the characteristic variance contribution rate of each component according to the correlation coefficient matrix; the method comprises the following specific steps:
solving the characteristic root of the correlation coefficient matrix R, and sorting the characteristic root according to a descending order rule, namely lambda 1 ≥λ 2 ≥…≥λ n Not less than 0; the single characteristic variance contribution rate of the component is obtained according to the characteristic root, as shown in a formula (9):
further, the specific step S205 includes: the main components are selected:
the single contribution rates are sequentially added from large to small to obtain the accumulated variance contribution rate, and k components with the contribution rate being more than 85% are used as main components, as shown in the formula (10):
further, the specific step S206 includes: calculating a component load matrix of the main component and each index according to the obtained main component:
the magnitude of the influence coefficient of each variable index on the principal component can be expressed by the load of the variable index on the principal component. Calculating corresponding unit feature vectors according to the size of k:
the unit feature matrix formed by the method is a component load matrix, as shown in formula (11). />
wherein ,Sij The load of the ith index on the jth principal component.
Further, the specific step S207 includes: calculating the comprehensive score coefficient of each index on the main component:
score contribution rate H of each index on principal component ij As shown in the formula (12), the comprehensive score coefficient U of each index i As shown in formula (13).
Further, the specific step S208 includes: calculating the weight value of each index:
normalizing the comprehensive score coefficient of each index to obtain a weight value P of each index i As shown in formula (14):
the flow of initial weight calculation based on PCA traffic situation index is shown in FIG. 3.
Further, the specific step of determining the weight change interval by using the gaussian distribution comprises the following steps:
and adopting [ w-3 sigma, w+3 sigma ] as a variation interval of the weight, wherein w represents the weight, and sigma represents the variance.
It should be understood that the specific steps of determining the weight change interval by using the gaussian distribution include:
after the initial value of the weight is determined, in order to reach the optimal weight, the weight needs to be searched continuously in a certain weight interval, and the reasonable determination of the range of index weight change is very important to the weight optimization result.
Interval of cumulative variance contribution rate from formula [0.85,1 ]]I.e.
When the cumulative variance contribution rate is higher, the number of the selected main components is larger, and when the cumulative variance contribution rate reaches 1, all the components are selected. Extracting a section for setting the cumulative variance contribution rate to [0.85,1 ]]The index weight under each component combination is calculated, and k index weight distribution schemes exist, and each index has k weights which obey Gaussian distribution. Assume that consumer satisfaction studies are performed on an index (physical store, reputation, enterprise image, service) reflecting a certain market performance, under which characteristic values of the first component, the second component, the third component, and the fourth component are 2.721, 1.111, 0.94, and 0.227, respectively, and the individual characteristic variance contribution rates are 54.426%, 22.22%, 18.802%, and 4.552%, respectively. Selecting the interval of the cumulative variance contribution rate as [0.85,1 ]]The composition of the components is respectively [ a first component, a second component and a third component ]]And [ first component, second component, third component, fourth component ]]And two combinations, and respectively calculating index weights under the two combinations.
In order to reasonably give the change interval of each index weight, a 3 sigma characterization method is adopted to judge the weight distribution result. As described above, under different weight allocation schemes, the weights of the indexes all follow gaussian distribution, and the gaussian distribution diagram is shown in fig. 4, and the calculation process is as follows:
wherein ,
gaussian distribution, c, for the ith index i and σi Mean and variance of ith index weight, w, under k weight allocation schemes ij For the weight of the ith index under the jth allocation scheme, k is the number of weight allocation schemes, and [ w-3σ, w+3σ ] is adopted]As a variation interval of the weight.
It is to be understood that the initial weight of the index is optimized by adopting a gray wolf optimization algorithm to obtain the optimal weight of the index; the gray wolf algorithm is used as a novel intelligent optimization algorithm for optimizing by simulating the grade system and food hunting of the gray wolf society. The gray wolf population has a strict pyramid type social grade system in the hunting process. In the whole wolf group, the wolf with the highest grade is alpha and is responsible for decision making and management of the whole wolf group in the hunting process; beta wolves and delta wolves are sub-optimal fitness groups which assist alpha wolves in managing the whole wolf group and have decision weights in hunting process; the remaining wolf individuals are defined as ω, assisting α, β, δ in attacking the prey.
The algorithm is specifically described as follows:
on the D-dimensional search space, the position of each wolf is a vector:
X i (t)＝(X i,1 (t),X i,2 (t),…,X i,D (t))，
where i=1, 2,3, …, N represents that the wolf group consists of N wolves. Grey wolf X i When the position is updated, firstly, three wolves X with optimal positions are calculated best (t)＝{X α (t),X β (t),X δ The distance of (t) } is shown in the formula (18):
wherein ,Xi And (t) represents the position of the ith wolf at time t. D (D) α 、D β 、D δ Respectively represent the gray wolves X at the current moment i Distance from alpha, beta, delta, C k ＝2·r 1 ，r 1 Is a random number between 0 and 1, and the value of k is 1,2 and 3.
Then X i By the formula
Updating the position of the self, which represents approaching to the optimal three wolves in the wolf group, namely the optimal solution direction.
Wherein X (t+1) represents the final movement result of the wolf, X' 1 (t)、X′ 2 (t)、X′ 3 (t) respectively the wolves X i The vector to be moved to α, β, δ is calculated as shown in formula (19):
wherein ,Ak ＝2a·r 2 -a, k=1, 2,3, where r 2 Is a random number between 0 and 1. a is a convergence factor, and is introduced for better constraint parameter optimization, and the calculation formula is a=2-2.t/T max T is the current iteration number, T max For the maximum number of iterations, it can be seen that as the number of iterations increases, the value of the convergence factor decreases linearly from 2 to 0, resulting in |A k The value of i decreases from greater than 1 to less than 1, thereby controlling the direction of optimization. The automatic adjustment of the convergence factor can achieve a faster convergence effect, meanwhile, when the convergence factor is large, the local optimal solution can be jumped out, and when the convergence factor is small, the oscillation back and forth near the global optimal solution can be avoided.
Weight optimization: the gray wolf algorithm is applied to the solving process of the multi-domain SDN flow situation index system weight, 14 indexes are obtained in total, and the change range of each index is determined. Defining the final index weight as
The weight change interval corresponding to the jth index is +.>
Different evaluation results can be obtained corresponding to different weight combinations, weights are substituted into a formula to calculate the evaluation results of different sample data under the weights, the variance of the evaluation results is calculated, the variance is maximized as a fitness function, and the following function maximum problem is solved through a wolf algorithm:
max f＝D(Z) (20)
wherein D (Z) represents the flow situation evaluation result variance of the m sample data; z represents the flow situation assessment result matrix of m sample data, and Z i Representing the final evaluation result of the ith sample; w (w) j A weight representing a j-th index in the index system;
and />
Is the weight w j Upper and lower bounds of the ripple, j=1, 2, …, n.
Further, as shown in fig. 5, the initial weights of the indexes are optimized by adopting a wolf optimization algorithm to obtain the optimal weights of the indexes; the method comprises the following specific steps:
s501: initializing gray wolf algorithm parameters;
s502: calculating the fitness value of each wolf in the wolf group;
s503: updating the individual position of the wolves according to the fitness value of each wolf in the wolf group;
s504: repeating S502-S503 to update the parameter iteration until convergence or until the iteration times are met, and recording the optimal individual with the maximum final fitness function as a weight distribution result.
Further, the step S501: initializing gray wolf algorithm parameters; the method comprises the following specific steps:
determining the number N of individual wolves, the maximum iteration number L and an initial population position matrix D, D= { pos (i) } of a wolf optimization algorithm i∈N . The individuals in the population are the numerical values of a group of weights, an initial population is generated by adopting a random uniform distribution function, the variation range of a weight interval is limited when the initial population is generated, the sum of the weights is adjusted to be 1, and the individuals which do not meet the weight variation interval after the weights are removed are circulated until the population scale is N.
Further, in S502, calculating an fitness value of each wolf in the wolf group; the method comprises the following specific steps:
and calculating the fitness value of each wolf in the wolf population. For the wolf i, decomposing the position pos (i) of the wolf into corresponding index weights w 1 ，w 2 ，…，w 14 And (5) bringing the weighted population into a gray wolf algorithm fitness calculation formula, and calculating the fitness value of each individual.
Further, the step S503: updating the individual position of the wolves according to the fitness value of each wolf in the wolf group; the method comprises the following specific steps:
comparing individual fitness values of the wolves, and sequentially storing the maximum 3 fitness values as f α 、f β and fδ Updating individual positions pos (i) of the wolves according to a position updating formula, and reassigning weights by adopting a normalization weight method in order to meet the condition that the sum of the weights after the position updating is still 1, as shown in a formula (24).
wherein ,wi (i=1, 2, …, 14) is the individual position of the wolf after being updated by the position formula, w i ' (i=1, 2, …, 14) is the normalized individual position of the wolf.
An embodiment two provides a multi-domain SDN network traffic situation assessment server;
a multi-domain SDN network traffic situation assessment server comprising:
an index set construction module configured to: acquiring a multi-domain SDN network traffic situation influencing factor, and constructing a multi-domain SDN network traffic situation index set;
an index initial weight calculation module configured to: calculating an index initial weight; determining a weight change interval based on the index initial weight;
an index optimal weight calculation module configured to: optimizing the initial weight of the index based on the weight change interval to obtain the optimal weight of the index;
a weight calculation module configured to: weighting calculation is carried out on the index optimal weight and the index value to obtain a multi-domain SDN network flow situation value;
a multi-domain SDN network traffic state output module configured to: and determining the current period multi-domain SDN network traffic state by combining the multi-domain SDN network traffic situation level table and the current period multi-domain SDN network traffic situation value.
In a third embodiment, the present embodiment further provides an electronic device, including a memory, a processor, and computer instructions stored on the memory and running on the processor, where the computer instructions, when executed by the processor, perform the method of the first embodiment.
In a fourth embodiment, a computer readable storage medium is provided, where the computer readable storage medium stores computer instructions that, when executed by a processor, perform the method of the first embodiment.
An embodiment five provides a multi-domain SDN network traffic situation assessment system;
a multi-domain SDN network traffic situation assessment system comprising: the multi-domain SDN network traffic situation assessment server described in the third embodiment; the system further comprises:
a multi-domain SDN network, the multi-domain SDN network comprising a plurality of SDN networks;
each SDN network is communicated with a corresponding intra-domain controller, and the intra-domain controller is responsible for collecting network traffic situation influencing factors in the corresponding SDN network; the intra-domain controller uploads the acquired network flow situation influencing factors to a database;
network traffic situation influencing factors between the SDN network and other SDN networks are collected by an inter-domain controller, and the inter-domain controller also uploads the collected network traffic situation influencing factors to a database;
the multi-domain SDN network traffic situation assessment server extracts network traffic situation influencing factors from the database, processes the extracted network traffic situation influencing factors and determines the current period multi-domain SDN network traffic state.
In order to timely and effectively master the situation of the whole network traffic situation, the present disclosure designs a multi-domain SDN network traffic situation assessment model, which is shown in fig. 1 and is composed of three parts of a plurality of SDN domains, a database and a situation assessment server.
(1) The multi-domain SDN network employs a vertical architecture. The system comprises an inter-domain controller and an intra-domain controller, wherein the intra-domain controller only controls internal communication, the inter-domain communication is controlled by the inter-domain controller, and the inter-domain controller grasps a global network view.
(2) In order to save network situation index information and historical processing results with large data volume, a database is introduced as a main mode of information storage, so that the system has the bearing capacity on a large amount of information. The controller collects information such as equipment states and flow in the multi-domain SDN network respectively, wherein intra-domain information is collected by the intra-domain controller, and inter-domain information is collected by the inter-domain controller; and writing the information sub-table into a database, and simultaneously, obtaining flow situation index data by the database in a uniform resource arrangement mode and writing the flow situation index data into the database.
(3) In order to reduce communication overhead and computation overhead, a network situation assessment server is introduced. The method is in charge of carrying out current period network flow situation assessment by using current period flow situation index data in the database, and returning the final assessment result to the inter-domain controller and writing the final assessment result into the database.
SDN has the characteristic that control layer and forwarding layer are separated, can provide clearer global field of view for the network technically, makes the acquisition of a large amount of real-time and historical data possible. As network scale increases, a single common SDN controller is difficult to meet network demands due to limited performance of the controllers, requiring deployment of multiple controllers that work cooperatively. The SDN multi-domain controller cooperative work mode is divided into a horizontal architecture and a vertical architecture. The network domains in the horizontal architecture have equal relation, and each network domain is controlled by a respective control plane and exchanges information with each other to cooperate with inter-domain communication, such as a distributed cluster. The vertical architecture includes inter-domain controllers that manage only internal communications and intra-domain controllers that manage global network views.
While the specific embodiments of the present disclosure have been described above with reference to the drawings, it should be understood that the present disclosure is not limited to the embodiments, and that various modifications and changes can be made by one skilled in the art without inventive effort on the basis of the technical solutions of the present disclosure while remaining within the scope of the present disclosure.
Claims (9)
1. A multi-domain SDN network traffic situation assessment method is characterized by comprising the following steps:
acquiring a multi-domain SDN network traffic situation influencing factor, and constructing a multi-domain SDN network traffic situation index set;
the indexes related to the multi-domain SDN flow situation index set comprise: (1) application and controller presence; (2) REST request rate; (3) a cross-domain request rate; calculating an index initial weight; determining a weight change interval based on the index initial weight;
optimizing the initial weight of the index based on the weight change interval to obtain the optimal weight of the index; optimizing the index initial weight by adopting a gray wolf optimization algorithm to obtain an index optimal weight;
the method comprises the following specific steps:
s501: initializing gray wolf algorithm parameters;
s502: calculating the fitness value of each wolf in the wolf group;
s503: updating the individual position of the wolves according to the fitness value of each wolf in the wolf group;
comparing individual fitness values of the wolves, and sequentially storing the maximum 3 fitness values as
、/>
and />
Position of individual gray wolves according to the position update formula>
Updating, namely, reassigning weights by adopting a normalized weight method in order to meet the condition that the sum of the weights after the position updating is still 1, as shown in a formula (24);
wherein ,
for the position of the individual gray wolves updated by the position formula,/for the individual gray wolves>
The position of the individual gray wolves after normalization;
s504: repeating S502-S503 to update parameter iteration until convergence or until the iteration times are met, and recording the optimal individual with the maximum final fitness function as a weight distribution result;
weighting calculation is carried out on the index optimal weight and the index value to obtain a multi-domain SDN network flow situation value;
combining the multi-domain SDN network traffic situation level table and the current period multi-domain SDN network traffic situation value to determine the current period multi-domain SDN network traffic state;
network traffic situation influencing factors between the SDN network and other SDN networks are collected by the inter-domain controllers, and the inter-domain controllers upload the collected network traffic situation influencing factors to the database.
2. The method of claim 1, wherein calculating the initial weights of the indicators is performed using a principal component analysis algorithm.
3. The method of claim 1, wherein the weight change interval is determined based on an initial weight of the index; the weight change interval is determined by using Gaussian distribution.
4. The method of claim 1, wherein the initial weights of the indicators are optimized based on the weight change interval to obtain optimal weights of the indicators.
5. The method of claim 2, wherein the calculating the index initial weight using the principal component analysis algorithm comprises:
s201: constructing a sample set; the sample set comprises a plurality of indexes of which the initial weights of the indexes are to be calculated;
s202: carrying out standardization treatment on the sample set;
s203: calculating the correlation coefficient between each index after the normalization processing to obtain a correlation coefficient matrix:
s204: calculating the characteristic variance contribution rate of each component according to the correlation coefficient matrix;
s205: based on the characteristic variance contribution rate and the set threshold
The individual components are used as main components;
s206: calculating a component load matrix of each main component and each index according to the obtained plurality of main components;
s207: calculating the comprehensive score coefficient of each index on the main component according to the component load matrix of each main component and each index;
s208: and calculating the initial weight of each index according to the comprehensive score coefficient of each index on the main component.
6. A multi-domain SDN network traffic situation assessment server is characterized by comprising:
an index set construction module configured to: acquiring a multi-domain SDN network traffic situation influencing factor, and constructing a multi-domain SDN network traffic situation index set;
the indexes related to the multi-domain SDN flow situation index set comprise: (1) application and controller presence; (2) REST request rate; (3) a cross-domain request rate;
an index initial weight calculation module configured to: calculating an index initial weight; determining a weight change interval based on the index initial weight;
an index optimal weight calculation module configured to: optimizing the initial weight of the index based on the weight change interval to obtain the optimal weight of the index; optimizing the index initial weight by adopting a gray wolf optimization algorithm to obtain an index optimal weight;
the method comprises the following specific steps:
s501: initializing gray wolf algorithm parameters;
s502: calculating the fitness value of each wolf in the wolf group;
s503: updating the individual position of the wolves according to the fitness value of each wolf in the wolf group;
comparing individual fitness values of the wolves, and sequentially storing the maximum 3 fitness values as
、/>
and />
Position of individual gray wolves according to the position update formula>
Updating, namely, reassigning weights by adopting a normalized weight method in order to meet the condition that the sum of the weights after the position updating is still 1, as shown in a formula (24);
wherein ,
for the position of the individual gray wolves updated by the position formula,/for the individual gray wolves>
The position of the individual gray wolves after normalization;
s504: repeating S502-S503 to update parameter iteration until convergence or until the iteration times are met, and recording the optimal individual with the maximum final fitness function as a weight distribution result;
a weight calculation module configured to: weighting calculation is carried out on the index optimal weight and the index value to obtain a multi-domain SDN network flow situation value;
a multi-domain SDN network traffic state output module configured to: combining the multi-domain SDN network traffic situation level table and the current period multi-domain SDN network traffic situation value to determine the current period multi-domain SDN network traffic state;
network traffic situation influencing factors between the SDN network and other SDN networks are collected by the inter-domain controllers, and the inter-domain controllers upload the collected network traffic situation influencing factors to the database.
7. An electronic device comprising a memory and a processor and computer instructions stored on the memory and running on the processor, which when executed by the processor, perform the method of any one of claims 1-5.
8. A computer readable storage medium storing computer instructions which, when executed by a processor, perform the method of any of claims 1-5.
9. A multi-domain SDN network traffic situation assessment system is characterized by comprising: the multi-domain SDN network traffic situation assessment server of claim 6; the system further comprises:
a multi-domain SDN network, the multi-domain SDN network comprising a plurality of SDN networks;
each SDN network is communicated with a corresponding intra-domain controller, and the intra-domain controller is responsible for collecting network traffic situation influencing factors in the corresponding SDN network; the intra-domain controller uploads the acquired network flow situation influencing factors to a database;
network traffic situation influencing factors between the SDN network and other SDN networks are collected by an inter-domain controller, and the inter-domain controller also uploads the collected network traffic situation influencing factors to a database;
the multi-domain SDN network traffic situation assessment server extracts network traffic situation influencing factors from the database, processes the extracted network traffic situation influencing factors and determines the current period multi-domain SDN network traffic state.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010092967.2A CN111327494B (en) | 2020-02-14 | 2020-02-14 | Multi-domain SDN network flow situation assessment method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010092967.2A CN111327494B (en) | 2020-02-14 | 2020-02-14 | Multi-domain SDN network flow situation assessment method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111327494A CN111327494A (en) | 2020-06-23 |
| CN111327494B true CN111327494B (en) | 2023-05-16 |
Family
ID=71171023
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010092967.2A Active CN111327494B (en) | 2020-02-14 | 2020-02-14 | Multi-domain SDN network flow situation assessment method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111327494B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2619909A (en) * | 2022-06-10 | 2023-12-27 | Vodafone Group Services Ltd | A method of managing network performance and/or configuration data in a telecommunications network |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104506385A (en) * | 2014-12-25 | 2015-04-08 | 西安电子科技大学 | Software defined network security situation assessment method |
| CN105491013A (en) * | 2015-11-20 | 2016-04-13 | 电子科技大学 | Multi-domain network security situation perception model and method based on SDN |
| CN109547431A (en) * | 2018-11-19 | 2019-03-29 | 国网河南省电力公司信息通信公司 | A kind of network security situation evaluating method based on CS and improved BP |
-
2020
- 2020-02-14 CN CN202010092967.2A patent/CN111327494B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104506385A (en) * | 2014-12-25 | 2015-04-08 | 西安电子科技大学 | Software defined network security situation assessment method |
| CN105491013A (en) * | 2015-11-20 | 2016-04-13 | 电子科技大学 | Multi-domain network security situation perception model and method based on SDN |
| CN109547431A (en) * | 2018-11-19 | 2019-03-29 | 国网河南省电力公司信息通信公司 | A kind of network security situation evaluating method based on CS and improved BP |
Non-Patent Citations (1)
| Title |
|---|
| SDN网络流量态势评估及预测技术研究;胡孟婷;《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》;20191215;第I139-44页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111327494A (en) | 2020-06-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Machine learning for networking: Workflow, advances and opportunities | |
| CN111526096B (en) | Intelligent identification network state prediction and congestion control system | |
| WO2020248228A1 (en) | Computing node load prediction method in a hadoop platform | |
| CN116070886B (en) | Multidimensional adaptive regulation and control method and system for water resource system | |
| CN111371644B (en) | Multi-domain SDN network traffic situation prediction method and system based on GRU | |
| Dalmazo et al. | Performance analysis of network traffic predictors in the cloud | |
| CN113114722A (en) | Virtual network function migration method based on edge network | |
| JP2018207241A (en) | Management device, management method, and management program | |
| Sacco et al. | Rope: An architecture for adaptive data-driven routing prediction at the edge | |
| CN116881744B (en) | Operation and maintenance data distribution method, device, equipment and medium based on Internet of things | |
| CN116743635B (en) | Network prediction and regulation method and network regulation system | |
| CN117539726B (en) | Energy efficiency optimization method and system for green intelligent computing center | |
| de Gyvés Avila et al. | Fuzzy logic based QoS optimization mechanism for service composition | |
| CN112202619A (en) | Intelligent cloud computing network flow adjusting and optimizing system and method | |
| CN111327494B (en) | Multi-domain SDN network flow situation assessment method and system | |
| Mahan et al. | A novel resource productivity based on granular neural network in cloud computing | |
| CN105933138B (en) | Space-time dimension combined cloud service credibility situation assessment and prediction method | |
| Wang et al. | A study of situation awareness-based resource management scheme in cloud environment | |
| Ahmed et al. | Network-aware SDN load balancer with deep active learning based intrusion detection model | |
| CN117692460A (en) | Server cluster control method and system | |
| CN114462506B (en) | Communication network auxiliary planning method supporting preference strategy and application thereof | |
| CN115174263B (en) | Attack path dynamic decision method and device | |
| CN114254734A (en) | Traffic matrix modeling method supporting deterministic application | |
| Sun et al. | Optimizing task-specific timeliness with edge-assisted scheduling for status update | |
| Frank et al. | Improving a smart environment with wireless network user load prediction |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230725 Address after: Room 1607, 16th Floor, Hongri Da, 200 meters north of the intersection of Beijing Road and Shuhe Road, Liuqing Street, Lanshan District, Linyi City, Shandong Province, 276000 Patentee after: Shandong Biaobiao Niu Network Technology Co.,Ltd. Address before: 264005 191 Binhai Middle Road, Laishan District, Yantai, Shandong Patentee before: SHANDONG INSTITUTE OF BUSINESS AND TECHNOLOGY |