CN106301973B - Network detecting method and the controller for using the method - Google Patents
Network detecting method and the controller for using the method Download PDFInfo
- Publication number
- CN106301973B CN106301973B CN201510245282.6A CN201510245282A CN106301973B CN 106301973 B CN106301973 B CN 106301973B CN 201510245282 A CN201510245282 A CN 201510245282A CN 106301973 B CN106301973 B CN 106301973B
- Authority
- CN
- China
- Prior art keywords
- route
- visit
- network
- network equipment
- visits
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/12—Avoiding congestion; Recovering from congestion
- H04L47/125—Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
Abstract
A kind of network detecting method, multiple nodes in topology network are concatenated to be formed and a plurality of visit side, according to it is described visit side and established by Snake algorithm a plurality of visit route, frequency is visited according to what every inspection frequency level visited on route calculated that each visits route, SDN policy flow (Policy Flow) is generated according to open flows (Openflow) agreement and is implemented to each node, and the pathChirp bandwidth estimation method according to package stream and improvement, all health status for visiting route of real-time dynamic detection and available bandwidth.
Description
Technical field
The present invention relates to a kind of network detecting methods, are based on software defined network (Software in particular to one kind
Defined Network, SDN) network detecting method and using the network detecting method controller.
Background technique
Software defined network (Software-Defined Networking, SDN) is a kind of network architecture of novel state,
Its core technology open flows (Open Flow) is by separating the control layer of the network equipment and data Layer, to realize network
The controller perturbation of flow.
The existing network architecture is during being transferred to software defined network (SDN) framework, with software defined network
(SDN) network equipment based on and legacy network devices will and be deposited for a period of time.
Under traditional network and SDN network and the hybrid network framework deposited, when carrying out network to network or the network equipment
It, need to be from SDN controller (Controller) the actively exchanger into network when diagnosing (Network Diagnostics)
(Switch) inspection requirements is proposed.However, if checking, frequency is too high, will cause the overload of network or the network equipment, for example,
SDN controller need to frequently issue KeepAlive package.If checking, frequency is too low, and SDN controller can not find to ask in real time
Topic.
In addition, the information (for example, bandwidth information) of network state to be obtained, SDN controller need additional package (example
Such as, pathChirp) it is detected.Therefore, the more, then load more to weigh the exchanger device of SDN controller management.
Therefore, it is necessary to a kind of network detecting methods of improvement to obtain network diagnostics information.
Summary of the invention
In view of this, being utilized it is an object of the present invention to provide a kind of network detecting method for being based on software defined network (SDN)
OpenFlow technical standard agreement (for example, ONF-OFPI), and network health information is obtained through specific algorithm.
The object of the invention also provides a kind of controller using the SDN network detection method, and the SDN controller can be with
Running obtains network state information while network diagnosis, compared with existing mode, greatly under the mixed network architecture together
Width reduces package transmission amount and reduces the load of SDN controller.
Embodiment of the present invention provides a kind of network detecting method, is applied to controller, which comprises open up network
Multiple network equipments in Piao are concatenated to be formed and multiple visit side;The network data on side is visited described in quantization to be corresponded to
The multiple healthy scores for visiting side;The healthy score is converted to visit multiple inspection frequency layers on side described in being corresponded to
Grade;It selects to visit the most first network device in side in the network equipment and establishes institute as start node, and according to algorithm
It states a plurality of of network topology and visits route comprising at least first visits route, wherein described first visits route including described
When multiple first network devices and multiple first visited in network equipment are visited;Side is visited according to described first
Inspection frequency level calculate described first visit route visit frequency;It is sent as an envoy to using bandwidth estimation algorithmic derivation for described
First visits the policy flow of route and the policy flow is applied on the first network device;Issue probe packets and in institute
It states first and visits and transmitted on route, the first network device stamps time stamp in the probe packets, and is walked by described first
The last first network device passback probe packets of route are visited to the controller;And it collects via the controller from described
First visits the probe packets of route passback and is analyzed, to obtain the strong of network link that described first visits route
Health information is to carry out network data update.
Preferably, visit route in the network detecting method of embodiment of the present invention, described in foundation to further include: selection is walked
Visit the second most network equipment of side;The second of the inspection frequency level for being less than particular value in second network equipment is selected to walk
It visits when being visited for starting and is moved to third network equipment, wherein visit side addition for described second and visited with forming second
Route;Judge whether the third network equipment includes to visit side;If the third network equipment includes to visit side, then judge
Whether the described second total length for visiting route is less than or equal to the first default value;If the described second total length for visiting route is less than
Equal to first default value, then calculating the third network equipment, its inspection frequency for visiting side is not used in each comprising it
Rate level, and each inspection frequency level that it visits side and third network equipment of being not used is calculated to previous
Difference between the inspection frequency level for visiting side of a network equipment;Judge whether the third network equipment has inspection frequency layer
The difference of grade visits side less than or equal to the second default value;If there is the difference for checking frequency level to be less than or equal to second default
The third of value visits side, then the third is visited side and be added described second and visited route and move to the 4th network equipment;And
It repeats the above steps, until there is no what the difference for checking frequency level was less than or equal to second default value to visit side or described
Until second total length for visiting route is greater than first default value, and exports described second and visit route.
A further embodiment of this invention provides a kind of controller, including conversion module, figure module, arrangement module and point
Analyse module.The conversion module formed to be concatenated multiple network equipments in topology network it is multiple visit side, amount
Visited described in change while network data to visit described in being corresponded to while multiple healthy scores, and the conversion healthy score
To visit multiple inspection frequency levels on side described in being corresponded to;The figure module is visited in the network equipment to select
The most first network device in side establishes a plurality of of the network topology according to algorithm and visits route as start node,
It includes at least first visiting route, wherein described first to visit route include multiple first networks dress in the network equipment
It sets when being visited with multiple first visited in, and described in the inspection frequency level for visiting side according to described first calculates
First visit route visit frequency;The arrangement module using bandwidth estimation algorithmic derivation to send as an envoy to walk for described first
It visits the policy flow of route and the policy flow is applied on the first network device;And the analysis module is to issue
Probe packets and visiting described first transmit on route, and the first network device stamps time stamp in the probe packets,
And the last first network device passback probe packets of route are visited to the controller, and via the control by described first
Device processed is collected the probe packets for visiting route passback from described first and is analyzed, and visits route to obtain described first
Network link health and fitness information to carry out network data update.
Preferably, in the controller of embodiment of the present invention, the second most net of side is visited in the figure module selection
Network device selects to visit less than the second of the inspection frequency level of particular value when visiting for starting in second network equipment
And it is moved to third network equipment, wherein visit side addition for described second and visit route to form second, judge the third
Whether network equipment includes to visit side, if the third network equipment includes to visit side, then judges that described second visits route
Total length whether be less than or equal to the first default value, if described second visit route total length be less than or equal to it is described first default
Value, then calculating the third network equipment, its inspection frequency level for visiting side is not used in each comprising it, and calculates
It is described each be not used inspection frequency level that it visits side and the third network equipment to previous network equipment it
The difference between the inspection frequency level on side is visited, judges whether the third network equipment has and checks that the difference of frequency level is less than
Side is visited equal to the second default value, if the third for having the difference for checking frequency level to be less than or equal to second default value is visited
The third is then visited side and is added described second and visited route and move to the 4th network equipment by side, and, repeat above-mentioned step
Suddenly, until there is no the difference for checking frequency level to be less than or equal to visiting side or described second visiting road for second default value
Until the total length of line is greater than first default value, and exports described second and visit route.
The SDN network detection method of embodiment of the present invention utilizes existing OpenFlow technical standard agreement, and penetrates
Unique calculation process obtains network health information.The SDN network detection method of embodiment of the present invention, which can operate, to be mixed
The network architecture under, obtain network state information together while network diagnosis, compared with existing mode, package be greatly decreased
Transmission quantity and the load for reducing SDN controller.
Below in conjunction with the drawings and specific embodiments, the present invention will be described in detail, but not as a limitation of the invention.
Detailed description of the invention
Fig. 1 is the architecture diagram of the network detection system 100 of embodiment of the present invention.
Fig. 2 is the functional block diagram of the SDN controller 110 of embodiment of the present invention.
Fig. 3 is the step flow chart of the SDN network detection method of embodiment of the present invention.
Fig. 4 is the schematic diagram of the inspection frequency level for visiting side of embodiment of the present invention.
Fig. 5 is the step flow chart for visiting route method for building up of embodiment of the present invention.
Fig. 6 A~6F is that the route method for building up of visiting of embodiment of the present invention establishes the flow chart for visiting route.
Fig. 7 A~7D is to visit route using what the network detecting method of embodiment of the present invention was established.
Fig. 8 is the implementation flow chart of the arrangement module 260 of embodiment of the present invention.
Fig. 9 A is health value and the corresponding schematic diagram for checking frequency level.
Fig. 9 B is the schematic diagram for checking frequency level with the relationship for visiting frequency.
Main element symbol description
Control equipment 10
Network detection system 100
Controller 110
Network equipment 122,124,126
Conversion module 220
Figure module 240
Arrangement module 260
Analysis module 280
Node N10, N12, N14, N16, N18, N20, N22, N24, N26
Start node R
First visits route 710
Second visits route 720
Third visits route 730
4th visits route 740
Exchanger 802,804,806,808,810
The present invention that the following detailed description will be further explained with reference to the above drawings.
Specific embodiment
Fig. 1 is the architecture diagram of the network detection system 100 of embodiment of the present invention, and the network detection system 100 includes
At least one is based on the controller 110 and network equipment 122,124 and 126 of software defined network (SDN).
The controller 110 of embodiment of the present invention establishes network by Snake algorithm using the network detecting method of SDN
The interior network equipment 122,124 and 126 of topology is visited route (Snake Path), then according to inspection frequency level (Check
Frequency Level, CFL) calculate that each visits route visit frequency.In addition, SDN controller 110 issues detection
(Probe) it package and is visited at every and transmits probe packets on route, the detection of route is then visited according to every be collected into
Package analysis each visits the remaining bandwidth and diagnostic message of the network link of route.
In embodiments of the present invention, the node in topology network is expressed as, for example, exchanger, or be other networks
Device, but not limited to this.
Fig. 2 is the functional block diagram of the SDN controller 110 of embodiment of the present invention.SDN controller 110 includes modulus of conversion
Block 220, figure module 240, arrangement module 260 and analysis module 280.
Node (for example, exchanger) in topology network shown in FIG. 1 is concatenated a plurality of to be formed by conversion module 220
Side (Snake Edges) is visited, the network data that each visits side is quantized into healthy score according to multiple factors, and turn
Healthy score is changed to obtain the inspection frequency level that each visits side.
Figure module 240 using Snake algorithm and according to check frequency level by multiple nodes and visit side be connected into it is more
Item is visited route (Snake Path), then the inspection frequency level visited on route according to every visits route to calculate each
Visit frequency.
Arrangement module 260 will visit route and visit frequency translation into SDN strategy according to open flows (Openflow) agreement
It flows (Policy Flow), and implements to each node (for example, exchanger).
Analysis module 280 according to the pathChirp bandwidth estimation method of package stream and improvement, real-time dynamic detection it is all walk
Visit the health status and available bandwidth of route.
Fig. 3 is the step flow chart of the SDN network detection method of embodiment of the present invention.
Node (for example, exchanger) in topology network is concatenated a plurality of to be formed by step S302, conversion module 220
Visit side.
The network data that each visits side is quantized into health point according to multiple factors by step S304, conversion module 220
Number.The factor includes at least critical path (Critical Path), reliability (Reliability), failure rate (Failure
Rate), available bandwidth (Available Bandwidth) and delay (Latency), but not limited to this.
Step S306, conversion module 220 convert above-mentioned healthy score to obtain the inspection frequency level that each visits side
(Check Frequency Level, CFL), as shown in Figure 4.
If it is critical path that a certain item, which visits side, its health value is set as 1 (Health Value=1).If it is described walk
Bian Feiwei critical path is visited, then calculates affiliated inspection frequency level, calculation formula using other factors are as follows:
Health Value=Reliability* α+(Failure Rate) * β+(Available Bandwidth) * γ
+ Latency* δ, wherein alpha+beta+γ+δ=1, the value of α, β, γ and δ are dynamically adapted, and Fig. 9 A indicates health value and corresponding inspection
Frequency level.
Step S308, the selection of figure module 240 are visited the most node (for example, node N10) in side (Snake Edge) and are done
For start node (Root), for example, node N10, and all road is visited according to what Snake algorithm was established in the network topology
Line.
Fig. 5 is the step flow chart for visiting route method for building up of embodiment of the present invention, and the process is by figure module
240 implement.
Step S502, selection visit the most node (for example, node N10) in side as start node.
Step S504 selects to check that frequency level is minimum and visits when visiting to originate and be moved to next node,
For example, selection node N10 visits side (CFL toward node N12a=1), as shown in Figure 6A.At this point, present node is by node N10
The total length for becoming node N12, and visiting route is 1.
Whether step S506 judges present node (for example, node N12) comprising visiting side.
Step S508, if present node (for example, node N12) includes to visit side, then the total of route is currently visited in judgement
Whether length is less than or equal to one first default value (for example, 14) (Esnake≦14)。
Step S510 calculates present node (for example, node if the total length for currently visiting route is less than or equal to 14
N12 it) is not used comprising it each and visits the inspection frequency level on side, and calculates each it is not used and visit side
Inspection frequency level and present node (for example, node N12) to previous node (for example, node N10) the inspection for visiting side
Look into the difference between frequency level.
As shown in Figure 6B, by taking node N12 as an example comprising 2 unused to visit side b and c, wherein CFLb=1,
CFLc=3.Difference when visiting between a of node N12 visited in b and c and node N10 respectively indicates as follows:
|CFLa–CFLb|=| 1-1 |=0;
|CFLa–CFLc|=| 1-3 |=2.
Step S512 is judged whether there is and is checked that the difference of frequency level is less than or equal to visiting for the second default value (for example, 1)
Side (| CFLmax–CFLnext|≦1)。
Step S514, if there is the difference for checking frequency level to visit side less than or equal to 1, for example, Fig. 6 B's visits side b,
Side b addition will be then visited currently to visit route and move to next node.At this point, present node becomes node from node N12
N14, and the total length for currently visiting route is 2.
Then, process returns to step S506, repeats step S506~S514, until not having to check that the difference of frequency level is small
(the E until the total length visited side or currently visit route equal to 1 is greater than 14snake>14or|CFLmax–CFLnext| > 1),
As shown in Fig. 6 C~6E.
It should be noted, if there is the difference of 1 or more inspection frequency level to visit side greater than 1, randomly choose wherein 1
Item visits side.
Step S516, if present node (for example, node N16) is not comprising visiting side, it is new to find to execute backtracking mechanism
Node.As shown in the dotted arrow in Fig. 6 F, node N20 is found after executing backtracking mechanism.At this point, present node is by node
N16 becomes node N20.
Whether step S518 judges present node (for example, node N20) comprising visiting side.
If present node (for example, node N20) includes to visit side, step S504 is returned to, is starting section with present node
Point repeats abovementioned steps to establish another and new visit route.Conversely, if present node (for example, node N20) does not wrap
Containing side is visited, then terminate this method process.
Step S520 indicates that currently visiting own establish of route completes if the total length for currently visiting route is greater than 14, and
It exports and described currently visits route.Then, step S504 is returned to, using present node as start node, repeats abovementioned steps
New route is visited to establish another.
Response of step S512, if not checking, the difference of frequency level visits side less than or equal to 1, exports described current
Visit route (step S520).
By taking the topology network of Fig. 4 as an example, when having executed the method flow of Fig. 5,4 can be established and visit route, including
First visit route 710, second visit route 720, third visits route 730 and the 4th visits route 740, respectively as Fig. 7 A~
Shown in 7D.
Step S310, figure module 240 according to check frequency level calculate every visit route visit frequency.
It visits frequency representation detection each and visits the time interval of route, be based on visiting inspection minimum in route
Frequency level determines.Check that frequency level and the relationship for visiting frequency are as shown in Figure 9 B.
Wherein, time T and time Δ are all adjustable parameters, for example, T and Δ are all preset as 10 seconds.
Compared with the general health diagnosis mechanism under tradition or SDN network, the network detecting method of embodiment of the present invention
Via visit in entire topology network it is a plurality of visit route, can obviously improve network efficiency.
Step S312, arrangement module 260 according to OpenFlow standard agreement, according to established visit route with
PathChirp algorithm is derived policy flow and is implemented on the OpenFlow node (for example, exchanger) of standard.
Arrangement module 260 in step S314, SDN controller 110 issues probe packets and visits route upload at every
It send, visits the exchanger on route and stamp time stamp (Timestamp) in each probe packets, and visit route most by every
The latter node returns the analysis module 280 of probe packets to SDN controller 110.
Analysis module 280 in step S316, SDN controller 110 collects the probe packets that route passback is visited from every
And analyzed, with health and fitness informations such as the remaining bandwidth and network latency, the quality that obtain each network link, and will be healthy
Information feeds back to the conversion module 220 in SDN controller 110, updates the health quantization score of topology network.
Fig. 8 is the implementation flow chart of the arrangement module 260 of embodiment of the present invention.
Arrangement module 260 in SDN controller 110 issues probe packets to the First exchanger 802 visited on route
((1)).When the probe packets reach exchanger 802, exchanger 802 analyze and in the detection according to SDN strategy
Package visits the time stamp of time plus expression package, and the probe packets are transferred to according to the SDN stream formulated and visit route
On exchanger 804 ((2)).When the exchanger 810 on route is visited in probe packets arrival ((3) and (4)), exchanger
810 visit the time stamp of time in the probe packets plus expression package, and the probe packets are then back to SDN controller
110((5))。
The network detecting method based on software defined network (SDN) of embodiment of the present invention utilizes known open flows
(OpenFlow) technical standard agreement (for example, ONF-OFPI), and network health information is obtained through unique calculation process.Institute
It states inventive method the node path of entire topology is conspired to create several and visit route, so that the health and fitness information package of passback tails off, i.e.,
Package transmission amount is greatly decreased, therefore the burden of SDN controller can be reduced.Meanwhile it being calculated through pathChirp estimating and measuring method each
Item visits the available bandwidth of route, can effectively reduce the package of computation bandwidth.
It should be noted that embodiment as described above, does not constitute the restriction to invention protection scope.It is any in this hair
Made modification in bright spirit and principle, equivalent replacement and improvement etc., should be included within the scope of the present invention.
Claims (17)
1. a kind of network detecting method is applied to controller, which is characterized in that the described method includes:
Multiple network equipments in topology network are concatenated to be formed and multiple visit side;
Visited described in quantization while network data to visit described in being corresponded to while multiple healthy scores, wherein according to multiple
The network data that each visits side is quantized into the healthy score (Health Value) by the factor, and the multiple factor is at least
Including reliability (Reliability), failure rate (Failure Rate), available bandwidth (Available Bandwidth) with
Postpone (Latency), the health score=reliability * α+failure rate * β+available bandwidth * γ+delay * δ, wherein alpha+beta+γ+δ
=1;
The healthy score is converted to visit multiple inspection frequency levels on side described in being corresponded to;
It selects to visit the most first network device in side in the network equipment and establishes institute as start node, and according to algorithm
It states a plurality of of network topology and visits route comprising at least first visits route, wherein described first visits route including described
When multiple first network devices and multiple first visited in network equipment are visited;
The inspection frequency level for visiting side according to described first calculate described first visit route visit frequency;
It is sent as an envoy to and the policy flow for visiting route for described first and is applied to the policy flow using bandwidth estimation algorithmic derivation
On the first network device;
It issues probe packets and is visited described first and transmitted on route, the first network device is beaten in the probe packets
Upper time stamp, and the last first network device passback probe packets of route are visited to the controller by described first;And
The probe packets for visiting route passback from described first are collected via the controller and are analyzed, to obtain
The health and fitness information for the network link that first visits route is stated to carry out network data update.
2. network detecting method as described in claim 1, which is characterized in that visit route described in foundation further include:
The second most network equipment of side is visited in selection;
It selects to visit when visiting for starting simultaneously in second network equipment less than the second of the inspection frequency level of particular value
It is moved to third network equipment, wherein visit side addition for described second and visit route to form second;
Judge whether the third network equipment includes to visit side;
If the third network equipment includes to visit side, whether the total length for then judging that described second visits route is less than or equal to
First default value;
If the described second total length for visiting route is less than or equal to first default value, the third network equipment packet is calculated
Containing it is each be not used its inspection frequency level for visiting side, and calculate and described each its inspection for visiting side be not used
Frequency level and the third network equipment are to the difference between the inspection frequency level for visiting side of previous network equipment;
Judge whether the third network equipment has the difference for checking frequency level to visit side less than or equal to the second default value;
If the third for having the difference for checking frequency level to be less than or equal to second default value visits side, the third is visited
Side is added described second and visits route and move to the 4th network equipment;And
The step of repeating the claims 1 and claim 2, it is described until there is no the difference for checking frequency level to be less than or equal to
Second default value visit side or described second visit route total length be greater than first default value until, and export
Described second visits route.
3. network detecting method as claimed in claim 2, which is characterized in that visit route described in foundation further include:
If the third network equipment do not include it is any visit side, execute backtracking mechanism to find the 4th network equipment;
Judge whether the 4th network equipment includes to visit side;
If the 4th network equipment includes to visit side, using the 4th network equipment as initial network device, repeat above-mentioned
Claim 1, claim 2 and new route is visited to establish one the step of claim 3;And
If the 4th network equipment does not include to visit side, terminate this method process.
4. network detecting method as claimed in claim 2, which is characterized in that visit route described in foundation further include:
If the described second total length for visiting route is greater than first default value, exports described second and visit route;And
Using current network device as initial network device, abovementioned steps are repeated to establish another and new visit route.
5. network detecting method as claimed in claim 2, which is characterized in that visit route described in foundation further include:
If the third network equipment does not check that the difference of frequency level visits side less than or equal to second default value,
It exports described second and visits route.
6. network detecting method as claimed in claim 2, which is characterized in that the method also includes: if the third network
Device has the difference of 1 or more inspection frequency level to visit side greater than second default value, then randomly chooses wherein 1
It visits side addition described second and visits route.
7. network detecting method as described in claim 1, which is characterized in that the method also includes: it is built using Snake algorithm
It stands and in the network topology described visits route.
8. network detecting method as described in claim 1, which is characterized in that the bandwidth estimation algorithm is pathChirp calculation
Method.
9. a kind of controller, which is characterized in that the controller includes:
Conversion module, formed to be concatenated multiple network equipments in topology network it is multiple visit side, quantify institute
State visit while network data to visit described in being corresponded to while multiple healthy scores, and the conversion healthy score with
To the multiple inspection frequency levels for visiting side described in correspondence, wherein each is visited to the network data on side according to multiple factors
It is quantized into the healthy score, the multiple factor includes at least reliability, failure rate, available bandwidth and delay, the health
Score=reliability * α+failure rate * β+available bandwidth * γ+delay * δ, wherein+δ=1 alpha+beta+γ;
Figure module, to select to visit the most first network device in side in the network equipment as start node, and
And the multiple of the network topology are established according to algorithm and visit route comprising at least first visits route, wherein described first
It visits route and includes multiple first network devices in the network equipment and while multiple first visited in are visited, with
And the inspection frequency level for according to described first visiting side calculate described first visit route visit frequency;
Arrangement module, to be sent as an envoy to the policy flow for visiting route for described first and by institute using bandwidth estimation algorithmic derivation
Policy flow is stated applied on the first network device;And
Analysis module transmits on route to issue probe packets and visit described first, and the first network device exists
Time stamp is stamped in the probe packets, and visits the last first network device passback probe packets of route to institute by described first
Controller is stated, and collects the probe packets for visiting route passback from described first via the controller and is divided
Analysis, to obtain the health and fitness information for the network link that described first visits route to carry out network data update.
10. controller as claimed in claim 9, which is characterized in that the second most net of side is visited in the figure module selection
Network device selects to visit less than the second of the inspection frequency level of particular value when visiting for starting in second network equipment
And it is moved to third network equipment, wherein visit side addition for described second and visit route to form second, judge the third
Whether network equipment includes to visit side, if the third network equipment includes to visit side, then judges that described second visits route
Total length whether be less than or equal to the first default value, if described second visit route total length be less than or equal to it is described first default
Value, then calculate that the third network equipment includes it is each be not used its inspection frequency level for visiting side, and calculate institute
State it is each be not used it visit while inspection frequency level and the third network equipment to previous network equipment when visiting
Inspection frequency level between difference, judge whether the third network equipment has the difference for checking frequency level to be less than or equal to the
Two default values visit side, if the third for having the difference for checking frequency level to be less than or equal to second default value visits side,
It the third is visited into side is added described second and visit route and move to the 4th network equipment, and, repeat the claims 1
The step of with claim 2, until do not have check frequency level difference be less than or equal to second default value visit side or
It is until the described second total length for visiting route is greater than first default value, and to export described second and visit route.
11. controller as claimed in claim 10, which is characterized in that the figure module does not wrap in the third network equipment
Containing it is any visit side when, execute a backtracking mechanism to find one the 4th network equipment, judge whether the 4th network equipment wraps
Containing side is visited, if the 4th network equipment includes to visit side, using the 4th network equipment as initial network device, repeat
The claims 1, claim 2 and the step of claim 3 with establish one it is new visit route, and, if described the
Four network equipments do not include to visit side, then terminate this process.
12. controller as claimed in claim 10, which is characterized in that the figure module visits the total of route described second
When length is greater than first default value, route is visited in output described second, and, it is filled by initial network of current network device
It sets, repeats abovementioned steps to establish another and new visit route.
13. controller as claimed in claim 10, which is characterized in that the figure module does not have in the third network equipment
Check that the difference of frequency level is less than or equal to when visiting side of second default value, route is visited in output described second.
14. controller as claimed in claim 10, which is characterized in that the figure module has 1 in the third network equipment
The difference of inspection frequency level more than item is greater than when visiting side of second default value, and random selection wherein visits side for 1
It is added described second and visits route.
15. controller as claimed in claim 9, which is characterized in that the figure module establishes the net using Snake algorithm
It is in network topology described to visit route.
16. controller as claimed in claim 9, which is characterized in that the bandwidth estimation algorithm is pathChirp algorithm.
17. controller as claimed in claim 9, which is characterized in that the arrangement module issues probe packets to described first
Each network equipment on route is visited, each network equipment is when receiving the probe packets, Yu Suoshu probe packets
In addition indicating that package visits the time stamp of time, and the next network equipment of probe packets transfer and described first are visited
The probe packets are back to the controller by the last one network equipment on route.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510245282.6A CN106301973B (en) | 2015-05-14 | 2015-05-14 | Network detecting method and the controller for using the method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510245282.6A CN106301973B (en) | 2015-05-14 | 2015-05-14 | Network detecting method and the controller for using the method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106301973A CN106301973A (en) | 2017-01-04 |
CN106301973B true CN106301973B (en) | 2019-07-23 |
Family
ID=57631091
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510245282.6A Expired - Fee Related CN106301973B (en) | 2015-05-14 | 2015-05-14 | Network detecting method and the controller for using the method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106301973B (en) |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6262976B1 (en) * | 1998-09-17 | 2001-07-17 | Ordered Networks, Inc. | System and method for network flow optimization using traffic classes |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8750158B2 (en) * | 2006-08-22 | 2014-06-10 | Centurylink Intellectual Property Llc | System and method for differentiated billing |
US8773992B2 (en) * | 2010-10-11 | 2014-07-08 | At&T Intellectual Property I, L.P. | Methods and apparatus for hierarchical routing in communication networks |
-
2015
- 2015-05-14 CN CN201510245282.6A patent/CN106301973B/en not_active Expired - Fee Related
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6262976B1 (en) * | 1998-09-17 | 2001-07-17 | Ordered Networks, Inc. | System and method for network flow optimization using traffic classes |
Also Published As
Publication number | Publication date |
---|---|
CN106301973A (en) | 2017-01-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105357138B (en) | A kind of adaptive method for reliable transmission based on low priority | |
CN104185298B (en) | Network load dynamic self-adapting parameter regulation means based on priority | |
CN104486163B (en) | A kind of single-end available bandwidth measuring and the terminal using this method | |
CN105024940A (en) | Link adaptation-based heterogeneous network TCP congestion control method | |
RU2477928C1 (en) | Method for evaluating information efficiency of communication system | |
US9054824B2 (en) | Inter-frame gap controller, traffic transmitter, transmission apparatus and inter-frame gap control method | |
TWI566544B (en) | Network detection method and controller using the method | |
CN106302017A (en) | High concurrent low discharge network velocity-measuring system and method | |
CN102224713B (en) | A method for estimation of residual bandwitdth | |
KR101628377B1 (en) | Apparatus and method for performing congestion control in a communication system | |
CN106301973B (en) | Network detecting method and the controller for using the method | |
CN103595552B (en) | Analysis method and analysis system for cluster storage network parallel load | |
CN110505037B (en) | Network interface communication rate matching method, device, equipment and storage medium | |
JP2004364168A (en) | Communication controller, communication control system, and communication control method | |
Hentschel et al. | Predictable communication for mobile systems | |
US7822052B2 (en) | Communication processing apparatus and totalizing system | |
CN108055147B (en) | Method for analyzing service performance of communication data network | |
JP2004364169A (en) | Network node, network system, and communication state adjustment method | |
JP2006279229A (en) | Traffic measurement system and traffic measurement method | |
Razmara et al. | A Hybrid Neural Network Approach for Congestion Control in TCP/IP Networks | |
JP5923914B2 (en) | Network state estimation apparatus and network state estimation program | |
Rés et al. | rt-winf: Real time wireless inference mechanism | |
KR100870182B1 (en) | Traffic estimation system and method for flow per routing pair in the Layer2/3 | |
CN110113794A (en) | A kind of radio network path mass dynamic static evaluation method | |
JP2018148494A (en) | Packet aggregation device and transmission processing program |
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: 20190723 |
|
CF01 | Termination of patent right due to non-payment of annual fee |