CN104935476B - A kind of network traffics matrix measuring method based on SDN - Google Patents
A kind of network traffics matrix measuring method based on SDN Download PDFInfo
- Publication number
- CN104935476B CN104935476B CN201510260772.3A CN201510260772A CN104935476B CN 104935476 B CN104935476 B CN 104935476B CN 201510260772 A CN201510260772 A CN 201510260772A CN 104935476 B CN104935476 B CN 104935476B
- Authority
- CN
- China
- Prior art keywords
- matrix
- stream
- node
- rule
- sdn switch
- 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.)
- Expired - Fee Related
Links
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
Abstract
The invention belongs to network management and network measure field, there is provided a kind of network traffics matrix measuring method based on SDN, meet that stream polymerization feasible constraints and network measure resource-constrained constrain to reach, the purpose of reducing the evaluated error of traffic matrix.Destination node identical stream is set in the present invention all to condense together, and convergence matrix is designed by depolymerization and routing rule;Using two-step method come estimated flow matrix, the first step is the most routing rule of one aggregated flow of selection (being referred to as old rule) to produce new rule, so that the number difference of old regular and new regular aggregated flow is minimum, then a traffic matrix is estimated using matrix completing technology;Second step is that the preceding k stream that maximum is therefrom selected based on the traffic matrix estimated in the first step carrys out direct measurement;The difficulty each flowed present invention reduces the cycle of measurement and direct measurement, and meet stream polymerization feasible constraints, evaluated error is than relatively low.
Description
Technical field
The invention belongs to network management and network measure field, particularly deploys and supports the SDN of OpenFlow agreements to hand over
The network management and measurement changed planes.
Background technology
Network measure refers in accordance with certain methods and techniques, and sign network state is tested using software or hardware toolses
With the index of performance, a series of movable general names such as network topology structure, nodal community and traffic characteristics are obtained.Network measure
The time delays of main contents including network, packet loss, available bandwidth, network topology and network traffics matrix etc..Network manager
In order to effectively manage network, it is necessary to the running status of awareness network in time.Therefore, the net of network operation state index is measured
Network e measurement technology becomes one of core technology of network management.Wherein, what traffic matrix represented is any two section in network
Uninterrupted between point pair.Traffic matrix is the key input information of many network management tasks, such as traffic engineering, network rule
Draw, network performance diagnostic analysis and charge on traffic etc..Network traffics matrix measurement is all uninterrupteds to being carried in network
Measure and estimate, so as to the traffic matrix that network manager can obtain according to measurement, rational planning and arrangement network
The route of stream, so as to improve the utilization ratio of the service quality of business and Internet resources.
Traditional network traffics matrix measuring method mainly includes the direct method of measurement and the estimation technique.The direct method of measurement it is basic
Principle is that network probe is arranged on network node or link, directly to obtain required parameter.But direct measurement
There is many difficulties in traffic matrix, first, the number of stream is typically bigger, it is impossible to which direct measurement goes out the big of each stream
Small, the cost so spent is too big;Secondly, measurement needs the measurement hardware resource of specialty, it is impossible on each network node
These expensive resources are installed, expended too big;The phase between heterogeneous networks is needed finally, due to network is isomery, during measurement
Mutually cooperation, but for some reason (such as network security), it is difficult to ensure that can be cooperated between each network, this is with regard to nothing
Method ensures the accuracy of measurement.
The basic ideas of the estimation technique are to obtain easily to measure or spend the small measurement of cost by mode measured directly
Information, reuse these information and estimate traffic matrix, the estimation technique, which is called, does the indirect method of measurement.In the estimation technique, it can utilize
Some additional informations in network come estimated flow matrix, such as link load, route matrix.Link load represents two to be connected
Flow on inter-node link, it is the uninterrupted that the flow in network converges according to route matrix on link.Link load
It can be obtained by using snmp protocol, route matrix can be by collecting the configuration information of interior routing protocol or passing through
The link-state information of interaction obtains between collect & route device.Therefore, traffic matrix estimation is in link load and route square
Under the premise of battle array is known, the estimate of traffic matrix is estimated.In a network, the number of stream is typically much deeper than the number of link,
I.e. route matrix is low-rank, can not accurately estimate traffic matrix.In order to improve the estimation accuracy of traffic matrix, it is necessary to
More additional informations are obtained, but due to the limitation of legacy network control plane, it is very tired to obtain more additional informations
Difficult.
In recent years, in order to strengthen the management and control ability to network, research circle proposes SDN framework.SDN is by the control of network
Face processed and data surface are thoroughly separated, and chain of command centralization is run on network controller, and the presence that data surface then disperses
In in each equipment.SDN this separation design also brings benefit for the measurement of network traffics matrix.First, it is centralized
Control plane there is global network view, can uniformly allotting network resource, dynamically configure flow table.Secondly, distribution
Some counters for stream statistics are provided in the data surface on the network equipment, we can be flow using these counters
The estimation of matrix provides more inputs.Therefore, studying the measurement of the network traffics matrix based on SDN can have to network manager
The control of effect ground and management network are necessary.
At present, academicly there is the research of some network traffics matrix measuring methods based on SDN.ProgMe is proposed
A kind of programmable network flow programming method structure, the traffic statistics value in the flow set that user specifies can be collected.
OpenSketch proposes a kind of gage frame based on Hash, can dynamically configure each framework.But ProgMe and
OpenSketch assumes that all interchangers all support the measurement hardware of specialty, and this is clearly impossible.And both sides
Method measures all just for some specific flow set, such as Heavy Hitter, and the measurement of their traffic matrix is all
It is unsuitable.
The method that Open-TM, DCM and iSTAMP are all based on SDN measurement traffic matrixs.Open-TM and DCM pass through
Monitor each stream and carry out the size that direct measurement each flows, so as to obtain traffic matrix.It is clear that Open-TM and DCM extension
Property it is all very poor, because the number that flows in a network is general very big, but required measurement resource is limited again, so at all
Traffic matrix can not possibly be obtained by size that direct measurement each flows.In order to meet to measure the constraint of resource, and improve
The accuracy of measurement, iSTAMP methods measure the preceding k stream of maximum and the size after some stream convergences, then estimate stream first
Moment matrix.ISTAMP seems to have made balance between the two in Internet resources and measurement accuracy, but it still has following lack
Fall into:First, it is that forward table is searched based on the matching principle of priority and asterisk wildcard during SDN switch forwarding flow, only has
The stream for having identical destination node is only possible to be aggregated in a list item (feasible constraints for being referred to as stream polymerization), still
This constraint is have ignored in iSTAMP methods, is arbitrarily polymerize;Secondly, in order to find out the preceding k stream of maximum, iSTAMP is utilized
Whole TCAM list items carry out in direct measurement multiple periods the size each flowed, and this can bring the measurement generation that can not ignore again
Valency.
The content of the invention
The purpose of the present invention is a kind of network traffics square based on SDN of research and design the defects of presence for background technology
Battle array measuring method.Using SDN characteristic, to design convergence matrix, (convergence matrix is 0-1 matrixes, and its every a line is corresponding one
TCAM list items, represent a flow measurement rule, and its element (i, j) represents whether stream j forwards through regular i), then utilize matrix
Completing technology estimates traffic matrix, meets that stream polymerization feasible constraints and network measure resource-constrained constrain to reach, reduces
The evaluated error of traffic matrix, the purpose of improving detection Heavy Hitter correct probability.
The scheme of solution of the present invention is:It is postulated that in order to saving TCAM list items, destination node identical stream is all aggregated in one
Rise, convergence matrix is designed by depolymerization and routing rule;Using two-step method come estimated flow matrix, the first step is selection one
The most routing rule of aggregated flow (be referred to as old rule) produces new rule so that the number phase of old regular and new regular aggregated flow
It is poor minimum, then estimate a traffic matrix using matrix completing technology;Second step is the flow to be estimated in the first step
Based on matrix, therefrom select the preceding k stream of maximum to carry out direct measurement, TCAM can be used with SDN switch using asking to characterize to flow
The most authority of the bipartite graph of relation between list item is maximum to be matched to solve the problems, such as that each SDN switch direct measurement adfluxion is closed;
The present invention realizes its purpose with this.Therefore, the present invention use method scheme for:
A kind of network traffics matrix measuring method based on SDN, comprises the following steps:
A. the initial convergence matrix of construction:Assuming that destination node identical stream is all polymerization when initial, i.e., only it need to use one
TCAM list items route the stream with identical destination node;Flow measurement corresponding to each SDN switch TCAM list items is designed to advise
Then, processing procedure is:
Step 1.0:In the TCAM list items of all SDN switches, a most rule of polymerization flow amount is selected (to be referred to as
Old rule) produce new rule, it is old it is regular with new rule in addition to priority, source IP prefix are different, remaining is all identical, newly
The priority of rule is higher than old rule, goes to step 2.0;
Step 2.0:The longer source IP prefix of a length is found for new rule:Using old regular source IP prefix as tree root
Prefix trees in find this prefix so that the numbers of the two regular aggregated flows difference is minimum, goes to step 2.1;
Step 2.1:Caused new rule is deployed in the available TCAM list items of SDN switch, if all SDN switches
TCAM tables it is all full, then go to step 3.0, otherwise go to step 1.0;
Step 3.0:The stream polymerizeing according to each TCAM list items, a convergence matrix is built, because SDN switch quantity has
Limit, part stream do not monitor, so route matrix is incorporated into convergence matrix according to link flow constraint, obtain to the end
Converge matrix;
B. estimated flow matrix:Convergence matrix according to being obtained in step A to the end is adopted come estimated flow matrix, method of estimation
With minimumλ is regular coefficient, and wherein A is convergence matrix, and X is required traffic matrix, and Y is and remittance
Column matrix corresponding to poly- square A, element corresponding with being incorporated to the convergence matrix before route matrix is the convergence of each list item in Y matrixes
Uninterrupted, element corresponding with route matrix are link load;
C. convergence matrix is redesigned:Each SDN switch is redesigned according to the traffic matrix estimated in step B
Flow measurement rule, its processing procedure are corresponding to TCAM list items:
Step 1.0:For each SDN switch, the adfluxion that can be monitored from it selects preceding k (the k values of maximum in closing
Equal to the number that this SDN switch can use TCAM list items) individual stream direct measurement, system of selection is:
Step 1.0.0:Construct virtual bipartite graph:The node of Part I is constructed first, and represent each SDN switch can
With TCAM list items, if the available TCAM list items number of the SDN switch at node j is s, s node j of generation is corresponded to1、
j2、...、js, each node represents an available list item of SDN switch TCAM tables;Then the node of Part II is constructed,
Each stream is represented, if there is L stream, then just has L node, stream i represents i-th of stream, i.e. node i;Finally construct Part I
Link between node and Part II node, if the SDN switch that stream i passes through at node j, node i and j1、j2、...、js
It is respectively connected with, the weight of link is to flow i size, and the addition manner of other links is similar, goes to step 1.0.1;
Step 1.0.1:Ask the maximum matching of most authority of this bipartite graph:It can be asked according to bipartite graph most authority maximum matching algorithm
Go out the maximum matching of most authority of this bipartite graph, then find out each SDN switch from the maximum matching of this most authority and directly to survey
K stream before the maximum of amount;For example node corresponding to 2 available TCAM list items of SDN switch is j at j1、j2, it is calculated
Match as j1- a, j2- b, then 2 streams are a, b before this SDN switch wants maximum measured directly;Go to step 2.0;
Step 2.0:The maximum matching of most authority obtained according to above-mentioned steps reconfigures each SDN switch and directly surveyed
After measuring which stream, it is possible to obtain a new convergence matrix, then link flow constraint is incorporated into convergence matrix, obtain most
New convergence matrix eventually;
D. last traffic matrix is estimated:The final new convergence matrix obtained according to step C is according to step B estimated flows
Matrix, and calculation error.
The present invention is due to when structure converges matrix, being produced for the largest number of rules of each list item aggregated flow
New rule, and ensure that the number difference of two regular aggregated flows is minimum as far as possible, then according to convergence Matrix Estimation stream now
Moment matrix, each SDN switch is finally found according to the traffic matrix estimated and wants adfluxion measured directly to close, it is not necessary to is known
The actual size that road each flows, reduces the cycle of measurement and difficulty that direct measurement each flows, and meets that stream polymerization is feasible
Property constraint, evaluated error is than relatively low.
Brief description of the drawings
Fig. 1 is the network physical topological network schematic diagram for having 5 nodes to form in present invention specific implementation.
Fig. 2 is the virtual bipartite graph for selecting to construct during the preceding k stream of maximum in present invention specific implementation.
Note:WhereinFor the element of actual flow matrix,Estimation obtains
The element of traffic matrix.
Embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.
Accompanying drawing 1 is the network physical topological network schematic diagram being made up of 5 nodes, and its interior joint 1 and 3 is to deploy SDN
The node of interchanger, the link in network be bi-directional link, and the present embodiment estimates traffic matrix, i.e., all nodes pair
Between flow, then with reality traffic matrix be compared.
The information of embodiment is as follows:
Assuming that the available TCAM list items number of two SDN switches (node 1 and node 3) is 7 in Fig. 1, five nodes
IP address be respectively 61.133.128.0/17,202.100.64.0/21,222.88.0.0/15,123.184.0.0/14,
116.192.0.0/16 corresponding IP prefixes are respectively 00111101100001011,110010100110010001000,
110111100101100、01111011101110、0111010011000000。
Document border traffic matrix is:
Implementation steps:
A. the initial convergence matrix of construction:It is minimum hop routing the route in stream injection network, each flowed.When each
SDN switch initialize for stream by list item after, the initial list item of TCAM tables of the SDN switch at node 1 has 5, point
The adfluxion not polymerizeing is combined into { 1->0,2->0,4->0}、{0->1,2->1,3->1,4->1}、{0->2,1->2,4->2}、{1->
3}、{0->4,1->4,2->4 }, the initial list item of TCAM tables of the SDN switch at node 3 has 5, and the adfluxion polymerizeing respectively is closed
For { 3->0}、{3->1}、{3->2}、{0->3,1->3,2->3,4->3}、{3->4}.Ensuing processing procedure is:
Step 1.0:In the TCAM list items of all SDN switches, the largest number of list items of aggregated flow are at node 1
The 2nd list item (or the 4th list item of the SDN switch at node 3) of SDN switch, rule corresponding to this list item is selected (to claim
For old rule) new rule is produced, go to step 2.0;
Step 2.0:Old rule has polymerize 4 streams, and source node is respectively 0,2,3,4, old regular source IP prefix be " " (i.e.
For sky).Prefix " 0111 " is searched out for the prefix trees of tree root with " " by finding, matches the two prefixes (" " and " 0111 ")
The number of stream is respectively 2, i.e., the source IP prefix of new rule is " 0111 ", and the old rule of its priority ratio is high, the number of the stream of polymerization
For 2;Then new rule is deployed in available TCAM list items.Now the two regular (corresponding to two TCAM list items) polymerizations
Adfluxion close be respectively { 3->1,4->1}、{0->1,2->1}.Go to step 2.1;
Step 2.1:If the TCAM list items of all SDN switches all have run out, 3.0 are gone to step, is otherwise gone to step
1.0;
Step 3.0:The stream polymerizeing according to the TCAM list items of all SDN switches, a convergence matrix is built, due to SDN
Interchanger limited amount, part stream are not monitored, route matrix is incorporated into convergence matrix so being constrained according to link flow
In, obtain convergence matrix to the end.
B. estimated flow matrix:Method of estimation is minimumWherein A is convergence matrix, and X is
Required traffic matrix (the traffic matrix that 5 rows 5 arrange by from left to right being write as a row from high to low), Y is the value of measurement.By this
Method obtains estimated flow matrix:
C. convergence matrix is redesigned:The TCAM list items of each SDN switch are redesigned according to the traffic matrix estimated,
Its processing procedure is:
Step 1.0:For each SDN switch, maximum preceding K=7-5=2 is selected in being closed from the adfluxion that can be monitored
(the list item total number of TCAM tables is 7, and the list item for route is 5) individual stream direct measurement, system of selection are:
Step 1.0.0:Construct virtual bipartite graph:Because the available TCAM list items number of each SDN switch is 2, each
The TCAM tables of SDN switch mean that into 2 nodes, see Fig. 2, and its interior joint A1, A2 represent the SDN switch at node 1
Two available TCAM list items, node B1, B2 represent node 3 at SDN switch two available TCAM list items, node 1,
2nd ..., 20 20 streams in network are represented, if some stream passes through node 1, just the node that the stream represents is connected with A1, A2, its
The addition manner of its link is similar.The weight (not marked in figure) of link is expressed as the size of corresponding stream, goes to step 1.0.1.
Step 1.0.1:Obtain the maximum matching of most authority of this bipartite graph:Can be in the hope of maximum of the most authority of this bipartite graph
With for A1-7, A2-14, B1-3, B2-13.It follows that maximum 2 stream of two SDN switch selections is respectively { 1->
3,3->1}、{0->3,3->0}.After so, the adfluxion that 7 list items of the SDN switch at node 1 polymerize respectively is combined into { 1-
>0,2->0,4->0}、{0->1,2->1,3->1,4->1}、{0->2,1->2,4->2}、{1->3}、{0->4,1->4,2->4}、
{1->3}、{3->1 }, the adfluxion that 7 list items of the SDN switch at node 3 polymerize respectively is combined into { 3->0}、{3->1}、{3->
2}、{0->3,1->3,2->3,4->3}、{3->4}、{0->3}、{3->0 }, one can then be obtained newly according to these list items
Convergence matrix, then link flow constraint be incorporated into convergence matrix in, obtain convergence matrix to the end.
D. last traffic matrix is estimated:The traffic matrix that the convergence Matrix Estimation obtained according to step C goes out is
EStimated_TM1, and NMAE errors are calculated as 0.1046.
The foregoing is only a specific embodiment of the invention, any feature disclosed in this specification, except non-specifically
Narration, can alternative features equivalent by other or with similar purpose replaced;Disclosed all features or all sides
Method or during the step of, in addition to mutually exclusive feature and/or step, can be combined in any way.
Claims (1)
1. a kind of network traffics matrix measuring method based on SDN, comprises the following steps:
A. the initial convergence matrix of construction:Destination node identical stream is all polymerization when setting initial, only need to use a TCAM table
It route the stream with identical destination node;Design flow measurement rule corresponding to each SDN switch TCAM list items, processing
Process is:
Step 1.0:In the TCAM list items of all SDN switches, the most rule of a polymerization flow amount is selected as old rule
Then to produce new rule, for old regular and new rule in addition to priority, source IP prefix are different, remaining is all identical, new rule
Priority then is higher than old rule, goes to step 2.0;
Step 2.0:Source IP prefix is found for new rule:Found before this in using old regular source IP prefix as the prefix trees of tree root
Sew so that the number difference of the two regular aggregated flows is minimum, goes to step 2.1;
Step 2.1:Caused new rule is deployed in the available TCAM list items of SDN switch, if all SDN switches
TCAM tables are all full, then go to step 3.0, otherwise go to step 1.0;
Step 3.0:The stream being polymerize according to each TCAM list items, a convergence matrix is built, is constrained further according to link flow road
It is incorporated into by matrix in convergence matrix, obtains convergence matrix to the end;
B. estimated flow matrix:According to the convergence matrix obtained in step A come estimated flow matrix, method of estimation, which uses, to be minimizedWherein λ is regular coefficient, and A is convergence matrix, and X is required traffic matrix, and Y is with converging A pairs of square
The column matrix answered;
C. convergence matrix is redesigned:Each SDN switch TCAM list items are redesigned according to the traffic matrix estimated in step B
Corresponding flow measurement rule, its processing procedure are:
Step 1.0:For each SDN switch, the adfluxion that can be monitored from it selects the preceding k stream of maximum directly to survey in closing
Amount, k value are equal to the number that this SDN switch can use TCAM list items, and system of selection is:
Step 1.0.0:Construct virtual bipartite graph:The node of Part I is constructed first, represents the available of each SDN switch
TCAM list items, if the available TCAM list items number of the SDN switch at node j is s, correspond to s node j of generation1、j2、...、
js, each node represents an available list item of SDN switch TCAM tables;Then the node of Part II is constructed, represents every
Individual stream, if there is L stream, just there is L node, stream i represents i-th of stream, also illustrates that node i;Finally construct Part I node
With the link between Part II node, if stream i by the SDN switch at node j, node i and j1、j2、...、jsRespectively
It is connected, the weight of link is to flow i size;The addition manner of other links is identical, goes to step 1.0.1;
Step 1.0.1:Ask the maximum matching of most authority of this bipartite graph:According to bipartite graph most authority maximum matching algorithm obtain this two
The maximum matching of the most authority of component, then found out from the maximum matching of this most authority each SDN switch want it is measured directly most
K stream before big;Go to step 2.0;
Step 2.0:It is measured directly that the maximum matching of most authority obtained according to above-mentioned steps reconfigures each SDN switch
Stream, obtains a new convergence matrix, and route matrix is incorporated into convergence matrix further according to link flow constraint, obtained final
New convergence matrix;
D. last traffic matrix is estimated:According to the new convergence matrix that step C is obtained according to step B estimated flow matrixes, and
Calculation error.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510260772.3A CN104935476B (en) | 2015-05-20 | 2015-05-20 | A kind of network traffics matrix measuring method based on SDN |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510260772.3A CN104935476B (en) | 2015-05-20 | 2015-05-20 | A kind of network traffics matrix measuring method based on SDN |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104935476A CN104935476A (en) | 2015-09-23 |
CN104935476B true CN104935476B (en) | 2018-04-06 |
Family
ID=54122451
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510260772.3A Expired - Fee Related CN104935476B (en) | 2015-05-20 | 2015-05-20 | A kind of network traffics matrix measuring method based on SDN |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104935476B (en) |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105897518B (en) * | 2016-06-30 | 2018-11-06 | 电子科技大学 | A kind of link performance parameters measurement method of full SDN network |
CN105959167B (en) * | 2016-07-18 | 2019-04-23 | 东北大学 | A kind of global optimization SDN measurement method based on greedy algorithm |
CN106230652B (en) * | 2016-07-19 | 2019-04-23 | 东北大学 | SDN network performance measurement method based on OpenFlow agreement |
CN106850348A (en) * | 2017-01-19 | 2017-06-13 | 中山大学 | A kind of traffic matrix On-line Estimation method of the data center internet based on SDN |
WO2020160780A1 (en) * | 2019-02-08 | 2020-08-13 | Huawei Technologies Co., Ltd. | Devices and method for controlling network routing configurations |
CN110868332B (en) * | 2019-10-25 | 2021-08-13 | 电子科技大学 | SDN-based network-level flow measurement method |
CN114389858B (en) * | 2021-12-24 | 2023-08-25 | 安天科技集团股份有限公司 | Flow processing method and device, electronic equipment and computer readable storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103346922A (en) * | 2013-07-26 | 2013-10-09 | 电子科技大学 | Controller for determining network state based on SDN (Software Defined Networking) and determination method thereof |
CN104158753A (en) * | 2014-06-12 | 2014-11-19 | 南京工程学院 | Dynamic flow dispatch method and system based on software definition network |
CN104333514A (en) * | 2014-11-28 | 2015-02-04 | 北京交通大学 | Network flow control method, device and system |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9407561B2 (en) * | 2013-06-19 | 2016-08-02 | Huawei Technologies Co., Ld. | Systems and methods for traffic engineering in software defined networks |
-
2015
- 2015-05-20 CN CN201510260772.3A patent/CN104935476B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103346922A (en) * | 2013-07-26 | 2013-10-09 | 电子科技大学 | Controller for determining network state based on SDN (Software Defined Networking) and determination method thereof |
CN104158753A (en) * | 2014-06-12 | 2014-11-19 | 南京工程学院 | Dynamic flow dispatch method and system based on software definition network |
CN104333514A (en) * | 2014-11-28 | 2015-02-04 | 北京交通大学 | Network flow control method, device and system |
Also Published As
Publication number | Publication date |
---|---|
CN104935476A (en) | 2015-09-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104935476B (en) | A kind of network traffics matrix measuring method based on SDN | |
CN108600102B (en) | Flexible data transmission system based on intelligent cooperative network | |
Liao et al. | Density cluster based approach for controller placement problem in large-scale software defined networkings | |
US10142186B2 (en) | System and method for designing a network for one or more entities in an enterprise | |
Chen et al. | Multi-controller placement towards SDN based on Louvain heuristic algorithm | |
CN109688056B (en) | Intelligent network control system and method | |
CN109150627A (en) | The construction method mapped based on dynamic resource demand and the virtual network of topology ambiguity | |
CN107332768A (en) | A kind of cross-domain path calculation method of multiple domain controller | |
CN103795596A (en) | Programmable control SDN measuring system and method | |
CN103810260B (en) | Complex network community based on topological property finds method | |
CN110275437B (en) | SDN network flow dominance monitoring node dynamic selection system and method thereof | |
WO2016061957A1 (en) | Method and apparatus for evaluating service traffic in lte network | |
CN106533759B (en) | A kind of link prediction method based on path entropy in multitiered network | |
CN103873364A (en) | Inter-domain multi-path rooting implementation method | |
CN104518899A (en) | Network routing traffic simulation method and device | |
Doshi et al. | Multi-constraint QoS disjoint multipath routing in SDN | |
Wang et al. | Inferring the average as path length of the internet | |
CN113794638A (en) | SDN data center network elephant flow scheduling method based on differential evolution algorithm | |
CN102420797A (en) | Topology mapping method and system | |
CN107222539A (en) | A kind of caching dispositions method that time delay Cost Model is competed based on node | |
CN104363142B (en) | A kind of automation data central site network performance bottleneck analysis method | |
Dai et al. | Mean first-passage times for two biased walks on the weighted rose networks | |
Wang et al. | Quantifying the flattening of internet topology | |
CN109981307A (en) | Method and system for network simulation | |
CN106850348A (en) | A kind of traffic matrix On-line Estimation method of the data center internet based on SDN |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180406 Termination date: 20210520 |
|
CF01 | Termination of patent right due to non-payment of annual fee |