CN106453114A - Flow distribution method and apparatus - Google Patents
Flow distribution method and apparatus Download PDFInfo
- Publication number
- CN106453114A CN106453114A CN201610890086.9A CN201610890086A CN106453114A CN 106453114 A CN106453114 A CN 106453114A CN 201610890086 A CN201610890086 A CN 201610890086A CN 106453114 A CN106453114 A CN 106453114A
- Authority
- CN
- China
- Prior art keywords
- flow
- output port
- function
- port
- assignment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/70—Admission control; Resource allocation
- H04L47/80—Actions related to the user profile or the type of traffic
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a flow distribution method and apparatus. The apparatus comprises a mode selection module used for selecting one preset flow distribution mode which has an output port selection function, a flow adjustment function and a session granularity selection function; a parameter input module used for inputting configuration parameters corresponding to the flow distribution function; a port data obtaining module used for obtaining flow data of an input port and an output port in real time and predicting future flow data of the input port according to the obtained flow data; a function calculation module used for, according to the configuration parameters, the flow data of the input port and the output port and the predicted future flow data of the input port, executing the output port selection module, the flow adjustment function and the session granularity selection function; and a distribution module used for distributing flow from the input port to the output port according to a function execution result. According to the scheme disclosed by the invention, the utilization rate of port bandwidth is improved and network experience of users is improved.
Description
Technical field
The present invention relates to network communication field, in particular to a kind of flow allocation method and device.
Background technology
In the network of data center or machine room, such as outer net outlet between network domains typically can arrange assignment of traffic and set
The standby function of undertaking exit/entry assignment of traffic.As a rule, because reliability, content buffering, cost and up-downgoing flow are big
The difference of little and type and the reason such as asymmetric, flowexit may have multiple, such as operator 1 exports, operator 2 exports,
Operator 3 exports.The rent use that exported due to these, amount of bandwidth, up-downgoing uninterrupted, reliability, content caching
Situation is different, generally requires flow distribution device and different flow is targetedly assigned to different outlets.Traditional scheme is
Carry out the selection of flowexit port using technology such as policybased routing, but because the technical scheme model comparision such as traditional measurement route are solid
The granularity that fixed, flow selects is larger and flow of dynamically cannot change dynamic select variable grain degree according to demand with situation
Be assigned to optimal outlet so that this equipment cannot flexibly, be efficiently completed the function of assignment of traffic, the assignment of traffic side backing towards
Method there is also similar problem and do not repeat herein.Especially current, the type of service of network traffics is continuously increased, the network user
The consciousness of network quality and experience is constantly lifted so that the operation of network, management and maintenance side have to propose to flow
The higher requirement of distributing equipment.
Content of the invention
In view of this, it is an object of the invention to provide a kind of flow allocation method and device, using SDN (software defined network
Network) technology realize network traffics dynamic, flexible distribute, the utilization rate that improves port bandwidth is, the experience improving user.
In embodiment of the present invention provide a kind of flow distribution device, including mode selection module, parameter input module,
Port data acquisition module, function computation module, distribute module, wherein mode selection module are used for selecting a default flow
Allocation model, described assignment of traffic pattern has an output port and selects function, a Flow-rate adjustment function and a session
Granularity selects function；Parameter input module is used for inputting the corresponding configuration parameter of described assignment of traffic pattern；Port data obtains
Delivery block is used for real-time acquisition input port and output port data on flows, and according to the described data on flows prediction input obtaining
Following data on flows in port；Function computation module is used for according to described configuration parameter, described input port and output port stream
The future anticipation data on flows described output port of execution of amount data and input port select function, described Flow-rate adjustment function and
Described session granularity selects function；Distribute module is used for carrying out flow from input port to output end according to function implementing result
The distribution of mouth.Heretofore described flow is made up of the session of variable grain degree, selects function by session granularity
The i.e. described flow of session after selection, the no Special Statement of flow described in this patent all can be referred to as the meeting of variable grain degree
Words, are not repeated to state.
Preferably, described assignment of traffic pattern includes port priority pattern, the configuration in described port priority pattern
Parameter includes：The flow bandwidth threshold value of each output port and total bandwidth；The priority value that each output port is endowed.
Preferably, described assignment of traffic pattern includes weighted load pattern, the configuration parameter in described weighted load pattern
Including：The flow bandwidth threshold value of each output port and total bandwidth；The weighted value that each output port is endowed.
Preferably, described assignment of traffic pattern includes user and type of service pattern, described user and type of service pattern
In configuration parameter include：The flow bandwidth threshold value of each output port and total bandwidth；The user class that each output port gives
Other property value；The type of service property value that each output port is endowed.
Preferably, described output port selects function to be used for selecting suitable output port, described flow for described flow
Adjustment function has been assigned to the flow of output port to make it meet the described assignment of traffic pattern of selection for real-time regulation
Require, described session granularity selects function to be used for selecting the flow of suitable particles degree to meet wanting of assignment of traffic with optimum
Ask.
A kind of flow managing method providing in a further embodiment of this invention, the method includes：Selection one is default
Assignment of traffic pattern, described assignment of traffic pattern has an output port and selects function, a Flow-rate adjustment function and one
Session granularity selects function；Input the corresponding configuration parameter of described assignment of traffic pattern；Obtain input port and output in real time
Port flow data；Described data on flows according to obtaining predicts following data on flows of input port；According to described configuration ginseng
Following data on flows of input port that several, described input port and output port traffic statistics, described prediction obtain is held
The described output port of row selects function, described Flow-rate adjustment function and described session granularity to select function；According to function execution
Result carries out distribution from input port to output port for the flow.Heretofore described flow is by the session of variable grain degree
Composition, i.e. the i.e. described flow of session after session granularity selects function to select, described in this patent, flow is no special
Statement all can be referred to as the session of variable grain degree, is not repeated to state.
Preferably, described assignment of traffic pattern includes port priority pattern, the configuration in described port priority pattern
Parameter includes：The flow bandwidth threshold value of each output port and total bandwidth；The priority value that each output port is endowed.
Preferably, described assignment of traffic pattern includes weighted load pattern, the configuration parameter in described weighted load pattern
Including：The flow bandwidth threshold value of each output port and total bandwidth；The weighted value that each output port is endowed.
Preferably, described assignment of traffic pattern includes user and type of service pattern, described user and type of service pattern
In configuration parameter include：The flow bandwidth threshold value of each output port and total bandwidth；The user class that each output port gives
Other property value；The type of service property value that each output port is endowed.
Preferably, described output port selects function to be used for selecting suitable output port, described flow for described flow
Adjustment function has been assigned to the flow of output port to make it meet the described assignment of traffic pattern of selection for real-time regulation
Require, described session granularity selects function to be used for selecting the flow of suitable particles degree to meet wanting of assignment of traffic with optimum
Ask.
Above-mentioned flow allocation method and device, so that the flow in SDN realizes dynamic distribution, improve port band
Wide utilization rate, improves the network experience of user, has saved the cost buying extra bandwidth so that network is more flexible, high
Effect.
Describe the present invention below in conjunction with the drawings and specific embodiments, but not as a limitation of the invention.
Brief description
Fig. 1 is a kind of applied environment figure of flow distribution device of the present invention 10 1 embodiment.
Fig. 2 is a kind of functional block diagram of flow distribution device of the present invention 10 1 embodiment.
Fig. 3 is a kind of functional block diagram of the another embodiment of flow distribution device of the present invention 10.
Fig. 4 is a kind of flow chart of flow allocation method of the present invention one embodiment.
Main element symbol description
Flow distribution device 10
SDN controller 1
SDN equipment 2
Mode selection module 100
Parameter input module 102
Port data acquisition module 104
Function computation module 106
Distribute module 108
Memory 110
Processor 112
Following specific embodiment will further illustrate the present invention in conjunction with above-mentioned accompanying drawing.
Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation description is it is clear that described embodiment a part of embodiment that is the present invention, rather than whole embodiments.Based on this
Embodiment in bright, the every other enforcement that those of ordinary skill in the art are obtained under the premise of not making creative work
Example, broadly falls into the scope of protection of the invention.
Fig. 1 is a kind of applied environment figure of flow distribution device of the present invention 10 1 embodiment.Wherein, flow distribution device
10 are located in SDN controller 1, and SDN controller 1 connection controls the SDN equipment 2 in whole SDN (as SDN exchanges
Machine), wherein SDN equipment 2 has n input port (n be non-zero positive integer) and m output port (m is just whole for non-zero
Number).In the present embodiment, flow distribution device 10 can be incited somebody to action according to the traffic conditions of input port automatically according to pre-conditioned
Flow dynamics distribute to output port.
Fig. 2 is a kind of functional block diagram of flow distribution device of the present invention 10 1 embodiment.Wherein flow distribution device
10 include mode selection module 100, parameter input module 102, port data acquisition module 104, function computation module 106, divide
Join module 108.
Mode selection module 100 is used for selecting a default assignment of traffic pattern, and described assignment of traffic pattern has one
Individual output port selects function, a Flow-rate adjustment function and a session granularity to select function.In the present embodiment, mould
Formula selecting module 100 selects default assignment of traffic pattern, and it actively can be selected it is also possible to be come by other environmental factors by user
The change of triggering, such as network congestion, time change or other conditions.
Parameter input module 102 is used for inputting the corresponding configuration parameter of described assignment of traffic pattern.Wherein, each flow divides
Join pattern and all correspond to a set of configuration parameter, so that corresponding discharge allocation model is mated in the distribution of flow in SDN equipment 2
Requirement.
In the present embodiment, default assignment of traffic pattern can be port priority pattern, as shown in Table-1, described
Configuration parameter in port priority pattern includes：The flow bandwidth threshold value of each output port and total bandwidth, such as output end
Mouth OUT1、OUT2Bandwidth threshold/total bandwidth be respectively 80M/100M, 16M/20M；The priority that each output port is endowed
Value, such as output port OUT1、OUT2Priority be respectively 1,2.
| OUT1 | OUT2 | OUTi | OUTm | |
| Threshold value/total bandwidth | 80M/100M | 16M/20M | ...... | 8M/10M |
| Port priority | 1 | 2 | ...... | n |
Table -1
In the present embodiment, default assignment of traffic pattern can also be weighted load pattern, as shown in table -2, described
Configuration parameter in weighted load pattern includes：The flow bandwidth threshold value of each output port and total bandwidth, such as output port
OUT1、OUT2Bandwidth threshold/total bandwidth be respectively 80M/100M, 16M/20M；The weighted value that each output port is endowed, than
As output port OUT1、OUT2Priority be respectively 0.6,0.1.
| OUT1 | OUT2 | OUTi | OUTm | |
| Threshold value/total bandwidth | 80M/100M | 16M/20M | ...... | 8M/10M |
| Port weighted value | 0.6 | 0.1 | ...... | 0.1 |
Table -2
In the present embodiment, default assignment of traffic pattern can also be user and type of service pattern, described user
And the configuration parameter in type of service pattern includes：The flow bandwidth threshold value of each output port and total bandwidth, such as output end
Mouth OUT1、OUT2Bandwidth threshold/total bandwidth be respectively 80M/100M, 16M/20M；The class of subscriber that each output port gives
Property value, such as output port OUT1、OUT2Class of subscriber attribute manager, research and development group；The business that each output port is endowed
Type Attribute value, such as output port Out1、Out2Type of service http/smtp, http/smtp.
Table -3
Port data acquisition module 104 is used for real-time acquisition flow input mouth and output port traffic statistics, and
Described data on flows according to obtaining predicts following data on flows of input port.
Function computation module 106 is used for according to described configuration parameter, described input port data on flows and output port stream
The described output port of data on flows execution in the input port future that amount data, described prediction obtain selects function, described flow
Adjustment function and session granularity select function.In the present embodiment, the different output of different assignment of traffic pattern correspondences
Port selection function, Flow-rate adjustment function and session granularity select function, specifically with port priority pattern, weighted load mould
Illustrate as a example formula.
1. port priority pattern：
1.1 port selection function F1 (X1, R1) for selecting outlet for new session.Wherein X1For current certain outlet OUTi(0
≤ i≤m) data on flows, X2It is expected flow, R for current sessions1Preferential for the outlet of user setup in port priority pattern
Parameter in level race-card, this function finds, from prepreerence outlet, the output end that first present flow rate is not less than the upper limit successively
Mouth j_F1 (0≤j≤m), returns this output port j_F1, and calculates the poor △ X_ with its flow rate upper limit for this port present flow rate
F1；
1.2 session granularities select function F2 (△ X, X2) for selecting suitable new session to import to the defeated of F1 selection
Exit port j_F1j.Wherein X2It is expected flow for current sessions, by this port present flow rate and the difference △ X of limit flow thereon, select
The new session of suitable granularity, if user does not have particular/special requirement, according to the most suitable session granularity of △ X unrestricted choice.
Wherein, so-called granularity is used to message is classified, and message can be divided into different types and deliver to needs by these classification
Outlet, to realize assignment of traffic；Meanwhile, with granularity, different size of meeting can be extracted as needed to message classification
Words, real assignment of traffic thick, fine granularity is adjustable.Further say, the message set that granularity is classified from certain dimension, classification
Can be the following combination of one or more, can also have more：The ingress physic port of message：Inbound port；Message
Source TCP/UDP port：Source logic port；The destination TCP/UDP port of message：Purpose logic port；
The source ip of message：Source IP address；The destination ip of message：Purpose IP address；The input vlan of message
tag：Entrance vlan；The output vlan tag of message：Outlet vlan；The source MAC of message：Source MAC；Message
destination MAC：Source MAC；The Length of message：Length, unit bit.
1.3 Flow-rate adjustment function F3 (X1, R1) it is for dynamic regulation output port assignment of traffic.Wherein X1For outlet
OUTiCurrent traffic data, R1For the parameter in the outlet priority order table of user setup in port priority pattern, this letter
Number from minimum preferential outlet find the outlet i of first present flow rate non-zero (begin stepping through from the minimum port of priority, excellent
First level finds non-zero flowexit from low to high one by one), if there is than outlet OUTiThe rate of discharge of higher priority does not reach
The upper limit, then by OUTiIn session take out as F2 (△ X, X2) in X2In new session preferentially distribute to and do not reach flow
The outlet of the upper limit.
2. weighted load pattern：
2.1 port selection function F4 (X1, X2, R2) be used for selecting minimum variance to export, wherein X1For rate of discharge data, X2
It is expected data on flows, R for current sessions2The assignment of traffic ratio table arranging for each outlet for user in weighted load pattern；
Outlet i (0≤i≤m) for each allocation proportion non-zero, calculates：
Outlet i present flow rate proportion ri(ri=X1i/X1all, wherein X1allIt is outlet total flow) to setting ratio Ri
Deviation square：e0i=[(ri-Ri)/Ri]2(explanation：Each port is calculated)；
Outlet i present flow rate adds that current sessions are expected flow s (s=X2/(X1all+X2)) afterwards to setting ratio Ri's
Deviation square:e1i=[(ri+s-Ri)/Ri]2(explanation：Each port is calculated)；
Calculate △ e for each porti=e1i-e0iValue；
If e1i-e0iThe flow of the outlet i of value minimum (can be negative) is not less than the preset flow upper limit (ri+s-Ri<0),
Returning outlet i (will this session X2Assignment of traffic on i port)；
e1i-e0i=[s-2 (Ri-ri)]s/Ri, riMore convergence Ri, value bigger, the bigger e1 of drag of this port of calculatingi-e0i
For being R when bearingi-ri<S/2, that is, during the not enough half to be allocated of current residual ratio, can exceed setting value after now adding s,
If now adding X2I port thresholding will be exceeded, so as the description in function F5 can pass through e1i-e0iIt is suitable to select
X2(new session of certain granularity is expected flow) is iterated calculating and imports effect to obtain optimal flow；
2.2 session granularities select function F5 (X2, △ ei), by △ ei=e1i-e0iSelect suitable session granularity
Flow X2It is assigned to corresponding port, so F4 and F5 will be calculated through multiple adjustment, so that F4 middle outlet i selects
Assignment of traffic is more nearly the traffic requirement under this pattern afterwards.
2.3 Flow-rate adjustment function F6 (X1, R2) for adjusting outlet distribution, X1For rate of discharge data, R2For weighted load
In pattern, user is the assignment of traffic ratio of each outlet setting, finds present flow rate proportion most beyond setting ratio
The session exporting in i is taken out call function F4 and is re-started distribution by outlet i.
Distribute module 108 is used for carrying out assignment of traffic according to function implementing result.In the present embodiment, calculate in function
The assignment of traffic of each port after module 106 has executed function calculating process, can be adjusted according to its result of calculation in real time, enter
And make the flow of certain output port to make it meet the requirement of the described assignment of traffic pattern of selection.
Fig. 3 is a kind of functional block diagram of the another embodiment of flow distribution device of the present invention 10.Wherein, assignment of traffic dress
Put 10 include mode selection module 100, parameter input module 102, port data acquisition module 104, function computation module 106,
Distribute module 108, memory 110 and processor 112, wherein mode selection module 100, parameter input module 102, port data
Acquisition module 104, function computation module 106, distribute module 108 are program code, in storage presented in functional module
Device 110, is processed by processor 112 to realize corresponding function.
Fig. 4 is a kind of flow chart of flow allocation method of the present invention one embodiment.
In step S400, mode selection module 100 is used for selecting a default assignment of traffic pattern, described assignment of traffic
Pattern has an output port and selects function, a Flow-rate adjustment function and a session granularity to select function.In this reality
Apply in mode, mode selection module selects default assignment of traffic pattern actively can select to trigger it is also possible to by it by user
His environmental factor triggering, such as network congestion.
In step S402, parameter input module 102 is used for inputting the corresponding configuration parameter of described assignment of traffic pattern.Its
In, each assignment of traffic pattern all correspond to a set of configuration parameter, so that in SDN equipment 2 distribution of flow is as far as possible
Join the requirement of response traffic allocation model.
In the present embodiment, default assignment of traffic pattern can be port priority pattern, as shown in table -1 above,
Configuration parameter in described port priority pattern includes：The flow bandwidth threshold value of each output port and total bandwidth, such as defeated
Exit port OUT1、OUT2Threshold value/total bandwidth be respectively 80M/100M, 16M/20M；The priority that each output port is endowed
Value, the such as priority of output port OUT1, OUT2 are respectively 1,2.
In the present embodiment, default assignment of traffic pattern can be weighted load pattern, as shown in table -2, described plus
Configuration parameter in power load model includes：The flow bandwidth threshold value of each output port and total bandwidth, such as output port
OUT1、OUt2Threshold value/total bandwidth be respectively 80M/100M, 16M/20M；The weighted value that each output port is endowed, such as defeated
Exit port OUT1、OUT2Priority be respectively 0.6,0.1.
In the present embodiment, default assignment of traffic pattern can be user and type of service pattern, described user and
Configuration parameter in type of service pattern includes：The flow bandwidth threshold value of each output port and total bandwidth, such as output port
OUT1、OUT2Threshold value/total bandwidth be respectively 80M/100M, 16M/20M；The class of subscriber property value that each output port gives,
Such as output port OUT1、OUT2Class of subscriber attribute manager, research and development group；The type of service that each output port is endowed belongs to
Property value, such as output port OUT1、OUT2Type of service http/smtp, http/smtp.
In step S404, port data acquisition module 104 is used for real-time acquisition flow input mouth and output port flow
Statistics, and following data on flows of this input port is predicted according to the described data on flows obtaining.
In step S406, function computation module 106 be used for according to described configuration parameter, described input port data on flows and
The described output port of data on flows execution in the input port future that output port data on flows, described prediction obtain selects letter
Several, described Flow-rate adjustment function and session granularity select function.In the present embodiment, different assignment of traffic patterns corresponds to
Different output port selects function, Flow-rate adjustment function and session granularity to select function, specifically with port priority pattern,
Illustrate as a example weighted load pattern.
1. port priority pattern：
1.1 port selection function F1 (X1, R1) for selecting outlet for new session.Wherein X1For current certain outlet OUTi(m
≤ i≤m) data on flows, X2It is expected flow, R for current sessions1Preferential for the outlet of user setup in port priority pattern
Parameter in level race-card, this function finds, from prepreerence outlet, the output end that first present flow rate is not less than the upper limit successively
Mouth j_F1 (0≤j≤m), returns this output port j_F1, and calculates the poor △ X_ with its flow rate upper limit for this port present flow rate
F1；
1.2 session granularities select function F2 (△ X, X2) be used for selecting suitable new session to import to F1 selection
Output port j_F1.Wherein X2It is expected flow for current sessions, by this port present flow rate and the difference △ X of limit flow thereon,
Select the new session of suitable granularity, if user does not have particular/special requirement, according to the most suitable session of △ X unrestricted choice
The description of granularity, wherein granularity is as described above.
1.3 Flow-rate adjustment function F3 (X1, R1) it is for dynamic regulation output port assignment of traffic.Wherein X1For outlet
OUTiCurrent traffic data, R1For the parameter in the outlet priority order table of user setup in port priority pattern, this letter
Number from minimum preferential outlet find the outlet i of first present flow rate non-zero (begin stepping through from the minimum port of priority, excellent
First level finds non-zero flowexit from low to high one by one), if there is than outlet OUTiThe rate of discharge of higher priority does not reach
The upper limit, then by OUTiIn session take out as F2 (△ X, X2) in X2In new session preferentially distribute to and do not reach flow
The outlet of the upper limit.
2. weighted load pattern：
2.1 function F4 (X1, X2, R2) be used for selecting minimum variance to export, wherein X1For rate of discharge data, X2For current
Session is expected data on flows, R2The assignment of traffic ratio table arranging for each outlet for user in weighted load pattern；For every
The outlet i (0≤i≤m) of individual allocation proportion non-zero, calculates：
Outlet i present flow rate proportion ri(ri=X1i/X1all, wherein X1allIt is outlet total flow) to setting ratio Ri
Deviation square：e0i=[(ri-Ri)/Ri]2(explanation：Each port is calculated)；
Outlet i present flow rate adds that current sessions are expected flow s (s=X2/(X1all+X2)) afterwards to setting ratio Ri's
Deviation square:e1i=[(ri+s-Ri)/Ri]2(explanation：Each port is calculated)；
Calculate e1 for each porti-e0iValue；
If e1i-e0iThe flow of the outlet i of value minimum (can be negative) is not less than the preset flow upper limit (ri+s-Ri<0),
Returning outlet i (will this session X2Assignment of traffic on i port)；
e1i-e0i=[s-2 (Ri-ri)]s/RiriMore convergence Ri, value bigger, the bigger e1 of drag of this port of calculatingi-e0iFor
It is R when negativei-ri<S/2, that is, during the not enough half to be allocated of current residual ratio, can exceed setting value, such as after now adding s
Fruit now adds X2I port thresholding will be exceeded, so as the description in function F5 can pass through e1i-e0iSelect suitable X2
(new session of certain granularity is expected flow) is iterated calculating and imports effect to obtain optimal flow；
2.2 session granularities select function F5 (X2, △ ei), by △ ei=e1i-e0iSelect suitable session granularity
Flow X2It is assigned to corresponding port, so F4 and F5 will be calculated through multiple adjustment, so that F4 middle outlet i selects
Assignment of traffic is more nearly the traffic requirement under this pattern afterwards.
2.3 Flow-rate adjustment function F6 (X1, R2) for adjusting outlet distribution, X1For rate of discharge data, R2Negative for weighting
In load pattern, user is the assignment of traffic ratio of each outlet setting, finds present flow rate proportion most beyond setting ratio
Outlet i by export i in session take out call function F4 re-start distribution.
In step S408, distribute module 108 is used for carrying out assignment of traffic according to function implementing result.In present embodiment
In, after function computation module 106 has executed function calculating process, each port can be adjusted in real time according to its result of calculation
Assignment of traffic, and then make the flow of certain output port to make it meet the requirement of the described assignment of traffic pattern of selection.
In addition, described in the invention be certain direction flow allocation method and device, another one direction (backs towards)
The distribution of data traffic can also equally operate, suitably increase in the case of the optional attributes such as trip path with the present invention above
Description be similar to, explanation is not repeated.
By inventive flow distributor 10 traffic distribution method so that the flow in SDN is realized dynamically
Distribution, improve the utilization rate of port bandwidth, improve user network experience, saved buy extra bandwidth cost,
Make network more flexibly, efficiently.
It should be noted that embodiment as described above, do not constitute the restriction to invention protection domain.Any at this
The modification made in bright spirit and principle, equivalent and improvement etc., should be included in protection scope of the present invention.
Claims (10)
1. a kind of flow allocation method is it is characterised in that include：
Select a default assignment of traffic pattern, described assignment of traffic pattern have output port select function, one
Flow-rate adjustment function and a session granularity select function；
Input the corresponding configuration parameter of described assignment of traffic pattern；
Obtain input port data on flows and output port data on flows in real time；
Following stream of input port is predicted according to the described input port data on flows obtaining and described output port data on flows
Amount data；
According to described configuration parameter, described input port data on flows and output port data on flows, described input port future
The data on flows described output port of execution select function, described Flow-rate adjustment function and described session granularity to select function；
And
Carry out assignment of traffic according to function implementing result.
2. flow allocation method as claimed in claim 1 is it is characterised in that described assignment of traffic pattern includes port priority
Pattern, the configuration parameter in described port priority pattern includes：
The flow bandwidth threshold value of each output port and total bandwidth；
The priority value that each output port is endowed.
3. flow allocation method as claimed in claim 1 is it is characterised in that described assignment of traffic pattern includes weighted load mould
Formula, the configuration parameter in described weighted load pattern includes：
The flow bandwidth threshold value of each output port and total bandwidth；
The weighted value that each output port is endowed.
4. flow allocation method as claimed in claim 1 is it is characterised in that described assignment of traffic pattern includes user and business
Configuration parameter in type-scheme, described user and type of service pattern includes：
The flow bandwidth threshold value of each output port and total bandwidth；
The class of subscriber property value that each output port gives；
The type of service property value that each output port is endowed.
5. flow allocation method as claimed in claim 1 is it is characterised in that described output port selects function to be used for as described
Flow selects a suitable output port, and described Flow-rate adjustment function is used for the flow of certain output port of real-time regulation to make
It meets the requirement of the described assignment of traffic pattern of selection, and described session granularity selects function to be used for selecting suitable granularity
Session export to suitable output port with reach best suit the corresponding assignment of traffic of described assignment of traffic pattern require.
6. a kind of flow distribution device is it is characterised in that include：
Mode selection module, for selecting a default assignment of traffic pattern, described assignment of traffic pattern has an output
Port selection function, a Flow-rate adjustment function and a session granularity select function；
Parameter input module, for inputting the corresponding configuration parameter of described assignment of traffic pattern；
Port data acquisition module, obtains input port data on flows and output port data on flows for real-time, and according to obtaining
The described input port data on flows taking and described output port data on flows predict following data on flows of input port；
Function computation module, for according to described configuration parameter, described input port data on flows and output port data on flows,
Following described output port of data on flows execution of described input port selects function, described Flow-rate adjustment function and described session
Granularity selects function；And
Distribute module, for carrying out assignment of traffic according to function implementing result.
7. flow distribution device as claimed in claim 6 is it is characterised in that described assignment of traffic pattern includes port priority
Pattern, the configuration parameter in described port priority pattern includes：
The flow bandwidth threshold value of each output port and total bandwidth；
The priority value that each output port is endowed.
8. flow distribution device as claimed in claim 6 is it is characterised in that described assignment of traffic pattern includes weighted load mould
Formula, the configuration parameter in described weighted load pattern includes：
The flow bandwidth threshold value of each output port and total bandwidth；
The weighted value that each output port is endowed.
9. flow distribution device as claimed in claim 6 is it is characterised in that described assignment of traffic pattern includes user and business
Configuration parameter in type-scheme, described user and type of service pattern includes：
The flow bandwidth threshold value of each output port and total bandwidth；
The class of subscriber property value that each output port gives；
The type of service property value that each output port is endowed.
10. flow distribution device as claimed in claim 6 is it is characterised in that described output port selects function to be used for as institute
State flow select a suitable output port, described Flow-rate adjustment function for certain output port of real-time regulation flow with
It is made to meet the requirement of the described assignment of traffic pattern of selection, described session granularity selects function to be used for selecting suitable particle
The session of degree exports and best suits the requirement of described assignment of traffic pattern corresponding assignment of traffic to suitable output port to reach.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610890086.9A CN106453114B (en) | 2016-10-11 | 2016-10-11 | Flow distribution method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610890086.9A CN106453114B (en) | 2016-10-11 | 2016-10-11 | Flow distribution method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106453114A true CN106453114A (en) | 2017-02-22 |
| CN106453114B CN106453114B (en) | 2020-03-17 |
Family
ID=58173649
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610890086.9A Active CN106453114B (en) | 2016-10-11 | 2016-10-11 | Flow distribution method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106453114B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106911581A (en) * | 2017-02-28 | 2017-06-30 | 联想(北京)有限公司 | A kind of transmission method, device and terminal |
| CN107689901A (en) * | 2017-11-13 | 2018-02-13 | 锐捷网络股份有限公司 | A kind of method and device of monitoring business message flow |
| WO2020140311A1 (en) * | 2019-01-04 | 2020-07-09 | 网宿科技股份有限公司 | Traffic scheduling method and system |
| CN112486871A (en) * | 2020-11-25 | 2021-03-12 | 海光信息技术股份有限公司 | Routing method and system for on-chip bus |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1913488A (en) * | 2004-12-29 | 2007-02-14 | 阿尔卡特公司 | Predictive congestion management in a data communications switch using traffic and system statistics |
| CN105162698A (en) * | 2015-10-10 | 2015-12-16 | 浪潮(北京)电子信息产业有限公司 | Method and device for cloud server to adjust SDN network paths based on memory model |
| CN105721577A (en) * | 2016-02-15 | 2016-06-29 | 安徽大学 | Software defined network-oriented server load balancing method |
-
2016
- 2016-10-11 CN CN201610890086.9A patent/CN106453114B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1913488A (en) * | 2004-12-29 | 2007-02-14 | 阿尔卡特公司 | Predictive congestion management in a data communications switch using traffic and system statistics |
| CN105162698A (en) * | 2015-10-10 | 2015-12-16 | 浪潮(北京)电子信息产业有限公司 | Method and device for cloud server to adjust SDN network paths based on memory model |
| CN105721577A (en) * | 2016-02-15 | 2016-06-29 | 安徽大学 | Software defined network-oriented server load balancing method |
Non-Patent Citations (2)
| Title |
|---|
| MA WEI ZHE等: "SDN-based Solution of Path Restoration for Smart Grids", 《4TH INTERNATIONAL CONFERENCE ON COMPUTER, MECHATRONICS, CONTROL AND ELECTRONIC ENGINEERING (ICCMCEE 2015)》 * |
| YASUNORI NAKAHODO等: "IMPLEMENTATION OF SMART-OSPF IN HYBRID SOFTWARE-DEFINED NETWORK", 《PROCEEDINGS OF 2014 4TH IEEE INTERNATIONAL CONFERENCE ON NETWORK INFRASTRUCTURE AND DIGITAL CONTENT》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106911581A (en) * | 2017-02-28 | 2017-06-30 | 联想(北京)有限公司 | A kind of transmission method, device and terminal |
| CN106911581B (en) * | 2017-02-28 | 2020-06-23 | 联想(北京)有限公司 | Transmission method, device and terminal |
| CN107689901A (en) * | 2017-11-13 | 2018-02-13 | 锐捷网络股份有限公司 | A kind of method and device of monitoring business message flow |
| CN107689901B (en) * | 2017-11-13 | 2020-12-18 | 锐捷网络股份有限公司 | Method and device for monitoring traffic message flow |
| WO2020140311A1 (en) * | 2019-01-04 | 2020-07-09 | 网宿科技股份有限公司 | Traffic scheduling method and system |
| CN112486871A (en) * | 2020-11-25 | 2021-03-12 | 海光信息技术股份有限公司 | Routing method and system for on-chip bus |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106453114B (en) | 2020-03-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Joint optimization of radio and virtual machine resources with uncertain user demands in mobile cloud computing | |
| CN106453114A (en) | Flow distribution method and apparatus | |
| EP1851921B1 (en) | Admission control and routing in a telecommunications network based on the consideration of all physical and logical links | |
| CN101115013B (en) | Method of providing resource admission control | |
| CN101938403B (en) | Assurance method of multi-user and multi-service quality of service and service access control point | |
| JP4705542B2 (en) | Best effort bandwidth allocation method and apparatus | |
| CN1557110A (en) | Data transmission in a packet-oriented communication network | |
| CN110048965A (en) | Network switch bandwidth management method and network system | |
| CN102239668A (en) | Method to assign traffic priority or bandwidth for application at the end users-device | |
| Wójcik et al. | Flow oriented approaches to QoS assurance | |
| CN106059954A (en) | Network bandwidth regulation method and device | |
| CN104618270B (en) | The method and apparatus of intelligent bandwidth distribution | |
| CN105071992B (en) | Substation's adapted electric industry business communication bandwidth prediction technique | |
| US11509566B2 (en) | Shaping outgoing traffic of network packets in a network management system | |
| CN106059942A (en) | Flow control method based on load prediction in SDN network | |
| CN1964326A (en) | A method to monitor flow and flow monitoring equipment | |
| CN107689992A (en) | A kind of high performance firewall cluster implementation method | |
| CN104735673A (en) | Bound link configuration processing method and configuration method and device | |
| CN107579963A (en) | A kind of high performance firewall cluster | |
| CN1234663A (en) | Frame-relay network whose frame-relay node having controlled over-subscribed band-width relay line | |
| CN102904828B (en) | A kind of load-balancing method and device | |
| CN110177056A (en) | A kind of self-reacting band width control method | |
| CN103023819A (en) | Coarse-grained dynamic bandwidth allocation method | |
| CN109474523B (en) | Networking method and system based on SDN | |
| CN104902570B (en) | A kind of dynamics of channels configuration method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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 |