CN109150740B - A kind of distributed network flow control method and device - Google Patents
A kind of distributed network flow control method and device Download PDFInfo
- Publication number
- CN109150740B CN109150740B CN201810863592.8A CN201810863592A CN109150740B CN 109150740 B CN109150740 B CN 109150740B CN 201810863592 A CN201810863592 A CN 201810863592A CN 109150740 B CN109150740 B CN 109150740B
- Authority
- CN
- China
- Prior art keywords
- flow
- control node
- flow control
- node
- summarize
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
- H04L43/0894—Packet rate
-
- 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
- H04L43/16—Threshold monitoring
-
- 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
-
- 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
- H04L47/20—Traffic policing
-
- 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
- H04L47/29—Flow control; Congestion control using a combination of thresholds
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Environmental & Geological Engineering (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention relates to Internet technical field, a kind of distributed network flow control method and device are provided.Method includes that each flow control node counts flow to be controlled, and statistics flow is sent to and summarizes control node；Summarize control node and sum to obtain by each node flow received and summarize flow, and by it is described summarize flow and synchronize be sent to each flow control node；Whether when each flow control node generates new flow to be controlled, analyzing newly generated flow plus the sum of described reference flow is more than flow control threshold value；And default behaviour control is carried out to newly generated flow based on the analysis results.Method provided by the invention, whether each node-flow magnitude of real-time synchronization is balanced without the concern for the flow between each node, whether have fluctuation, can fast and accurately control whole flow.Flow the adjustment even increase of node between node in distributed network are deleted all generate whole control effect and be significantly affected.
Description
[technical field]
The present invention relates to Internet technical fields, more particularly to a kind of distributed network flow control method and device.
[background technique]
With the explosive growth of each class of electronic devices such as PC, mobile phone, plate, the network size of internet also expands rapidly
Greatly, the behavior and network application of user are also more and more diversified.Many non-key service applications, which are robbed, on network accounts for ours
Bandwidth, so that many normal key businesses and application are affected.When user stays at home normal browsing webpage, but occur
When the case where network qualities such as network Caton reduce, the experience of user can have a greatly reduced quality.In order to promote user experience, Logistics networks
The unimpeded and regular traffic of physical channel, then there has been flow control.
Currently used flow control scene has the flow control for single user, and the scene is mainly in user's couple in router
On, or controlled on BRAS/BAS.But it if to carry out flow control by agreement to whole network, generally requires
The arrangement of fluidic device is carried out in Metropolitan Area Network (MAN) or the exit for saving net.
In order to avoid disaster, the influence to service such as failure, the access of Metropolitan Area Network (MAN) to the backbone network of operator will use more
A outlet, or even can be accessed in different computer rooms, flow generally presses stream and carries out load balancing between each access point.Another party
Face, as it currently stands, Metropolitan Area Network (MAN) or save the access bandwidth of net to backbone network and can arrive several hundred or even thousands of G, but also
In continuing to increase.
Using the mode of traditional single node flow control, on the one hand physically it is difficult that all flows is accomplished all to converge to
In one equipment, it is also difficult on the other hand carrying out flow control according to agreement and user using flow of the single device to several hundred thousands of G
With what is accomplished.
Currently used way is to divide equally global threshold according to number of devices, but the distribution of existing net flow is general
It is non-uniform, and often fluctuate.On the whole, larger using this method progress flow control error, or even will affect normal industry
Business.
[summary of the invention]
The technical problem to be solved by the present invention is to when flow control object is more, need synchronous mass data, to node
Between network bandwidth requirement it is higher.
The further technical problems to be solved of the present invention are each node mass flow discrepancy weighing apparatuses in distributed network flow control
It is still accurate to flow control, stable and in the case where there is fluctuation.
The present invention adopts the following technical scheme:
In a first aspect, including summarizing in distributed network the present invention provides a kind of distributed network flow control method
Control node and multiple flow control nodes connected to it, flow control node are serially connected in the network to be controlled, controlling party
Method includes:
Each flow control node counts flow to be controlled, and statistics flow is sent to and summarizes control node；
Summarize control node and sum to obtain by each node flow received and summarize flow, and summarizes the synchronous hair of flow for described
Give each flow control node；Wherein, when being synchronized to each flow control node and summarizing flow, each flow control node is by the remittance
Total flow subtracts the flow itself reported and obtains reference flow；
When each flow control node generates new flow to be controlled, analyze newly generated flow plus the reference flow it
It whether is more than flow control threshold value；
If analyzing result is to let pass to newly generated flow, less than flow control threshold value otherwise, to newly generated flow
Carry out default behaviour control.
Preferably, when each flow control node generates new flow to be controlled, newly generated flow is analyzed plus the ginseng
Examine whether the sum of flow is more than flow control threshold value, further includes:
The analysis of flow control node is determined from reporting statistics flow meeting the stream in preset time to after summarizing node
The correspondence flow that is monitored of amount control node is increased speed more than corresponding to flow incrementss in pre-set velocity value and/or preset time
More than the preset percentage of the remaining flow rate zone that can let pass, then to summarizing, control node transmission is updated to summarize flow at once.
Preferably, the object that each flow control node counts flow to be controlled is the stream of the first data object
Amount then summarizes the statistics flow that control node is also used to return according to each flow control node, corresponding first data of confirmation
The flow that summarizes of object has reached or close to its flow control threshold value, then the control method further include:
Summarize control node and determines distribution shape of the flow of corresponding first data object on each flow control node
State；
One or more flow control by the flow occupancy of first data object less than the first preset threshold
It is default more than or equal to first to be transferred to other flow occupancies for the traffic service task that first data object is corresponded in node
The flow control node of threshold value.
Preferably, the flow occupancy by first data object is less than one of the first preset threshold or more
The traffic service task that first data object is corresponded in a flow control node, be transferred to other flow occupancies be greater than etc.
In the flow control node of the first preset threshold, further includes:
The flow control node being transferred into, after obtaining the traffic service task for being transferred to and corresponding to first data object,
The flow of the correspondence counted first data object, which needs to meet, is more than or equal to the first preset threshold, and default less than second
Threshold value.
Preferably, the control node that summarizes is in the traffic service task transfer for completing corresponding first data object
Afterwards, the control method further include:
The flow for summarizing control node setting and being currently owned by the traffic service task of corresponding first data object
Control node is the exclusive node of the traffic service task of first data object；
The control node that summarizes removes corresponding exclusive section in flow control node when reaffirming that summarizing flow is zero
The mark content of point.
Preferably, it is described summarize each node flow that control node will receive and sum to obtain summarize flow, and by the remittance
Total flow synchronizes when being sent to each flow control node, also by the flat of flow to be controlled in each flow control node being calculated
Equal speed synchronizes together is sent to flow control node；
When then each flow control node generates new flow to be controlled, newly generated flow is analyzed plus the reference stream
It measures, and whether the sum of pre-configuration flow caused by the average speed plus the flow to be controlled is more than flow control threshold
Value；If analyzing result is to let pass less than flow control threshold value to newly generated flow, otherwise, newly generated flow is carried out
Default behaviour control.
Preferably, the identification information in the statistics flow also comprising accessed data object, then it is described to summarize control
Node is also used to summarize the accessed number of each data object；It is more than third predetermined threshold value, accessed time for accessed number
It is more than any one condition feelings in the 4th preset threshold that number is accessed number in default ranking section, in preset time range
Condition meets, then summarizes control node for the traffic interface of corresponding data object and be assigned to one or more flow control node,
And generate each flow control node that assignment information is sent under its administration；So that each flow control node can be according to the finger
Send whether validation of information forwards access request to one or more flow control node being assigned.
Preferably, common discharge control node detects first in the access access interface as the first intelligent terminal
The data access of intelligent terminal is consistent with the data object in the assignment information, then first intelligent terminal is directed to institute
The access request for stating data object is transmitted to described one or more flow control node being assigned.
Preferably, the data object for summarizing control node and being also used to count appointment for each flow control node
Quantity, and dividing for each appointment is completed according to the processing capacity of the flow control threshold value of each data object and each flow control node
Match.
Second aspect, the present invention also provides a kind of distributed network flow control devices, for realizing first aspect institute
The distributed network flow control method stated, described device include:
At least one processor；And the memory being connect at least one described processor communication；Wherein, described to deposit
Reservoir is stored with the instruction that can be executed by least one described processor, and described instruction is arranged to carry out first aspect institute by program
The distributed network flow control method stated.
The third aspect, the present invention also provides a kind of nonvolatile computer storage media, the computer storage medium
Computer executable instructions are stored with, which is executed by one or more processors, for completing first
Distributed network flow control method described in aspect.
Method provided by the invention, each node-flow magnitude of real-time synchronization, without the concern for the flow between each node whether
Whether equilibrium has fluctuation, can fast and accurately control whole flow.Flow tune between node in distributed network
Whole or even node increase is deleted all generate whole control effect and be significantly affected.
In addition, being made in preferred implementation scheme of the invention by introducing the concept of each flow control node average speed
Each flow control node receive summarize that control node sends over when summarizing flow, remittance can received next time
In the time cycle of total flow, has to the flow control situation of entire environment and more reliably estimate and judge, thus into one
Step improves the precision (avoiding that total flow occurs to be more than flow control threshold value) of flow control.
In addition to this, also by introducing the forwarding function of traffic service in preferred implementation scheme of the invention, so that whole
The flow of a network reports fine-grained management smaller with the occupied bandwidth of flow control, also, realizing flow control node.
[Detailed description of the invention]
In order to illustrate the technical solution of the embodiments of the present invention more clearly, will make below to required in the embodiment of the present invention
Attached drawing is briefly described.It should be evident that drawings described below is only some embodiments of the present invention, for
For those of ordinary skill in the art, without creative efforts, it can also be obtained according to these attached drawings other
Attached drawing.
Fig. 1 is a kind of distributed network flow control method flow diagram provided in an embodiment of the present invention；
Fig. 2 is a kind of distributed network flow control method signaling structure figure provided in an embodiment of the present invention；
Fig. 3 is the letter of flow control node side in a kind of distributed network flow control method provided in an embodiment of the present invention
Enable flow chart；
Fig. 4 is a kind of topological structure figure of distributed network provided in an embodiment of the present invention；
Fig. 5 is the letter for summarizing control node side in a kind of distributed network flow control method provided in an embodiment of the present invention
Enable flow chart；
Fig. 6 is a kind of distributed network flow control method flow diagram provided in an embodiment of the present invention；
Fig. 7 is a kind of structural schematic diagram of distributed network flow control device provided in an embodiment of the present invention.
[specific embodiment]
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
In the description of the present invention, term "inner", "outside", " longitudinal direction ", " transverse direction ", "upper", "lower", "top", "bottom" etc. refer to
The orientation or positional relationship shown be based on the orientation or positional relationship shown in the drawings, be merely for convenience of description the present invention rather than
It is required that the present invention must be constructed and operated in a specific orientation, therefore it is not construed as limitation of the present invention.
In addition, as long as technical characteristic involved in the various embodiments of the present invention described below is each other not
Constituting conflict can be combined with each other.
Embodiment 1:
The embodiment of the present invention 1 provides a kind of distributed network flow control method, includes summarizing control in distributed network
Node processed and multiple flow control nodes connected to it, flow control node are serially connected in the network to be controlled.In the present invention
In embodiment, the control node that summarizes is node in logic, when actual deployment, can be multiplexed with flow control node same
Physical equipment exists simultaneously on same physical equipment and summarizes control node and flow control node.As shown in Figure 1-Figure 3,
Control method includes:
In step 201, each flow control node counts flow to be controlled, and statistics flow is sent to remittance
Master control node.
Wherein, the flow to be controlled can be flow, a certain media data generated for a certain application and generate
The flow etc. that flow, a certain service type generate.The calculating for summarizing flow is completed for the ease of summarizing control node, preferably
It is that be sent in the statistics flow for summarize control node include time tag, the time tag is used to summarize control section
Point carries out fining flow and summarizes offer foundation, also, can also be the average speed of realization flow in subsequent expansion scheme
Calculating provide calculation basis.
In step 202, summarize control node and sum to obtain by each node flow received and summarize flow, and by the remittance
Total flow, which synchronizes, is sent to each flow control node；Wherein, when being synchronized to each flow control node and summarizing flow, each flow control
Node by it is described summarize flow and subtract the flow itself reported obtain reference flow.
As shown in Fig. 2, being to carry out signaling by taking flow control node 1, flow control node 2 and flow control node 3 as an example
The effect diagram that process is shown.
In step 203, when each flow control node generates new flow to be controlled, newly generated flow is analyzed plus institute
State whether the sum of reference flow is more than flow control threshold value.
As shown in figure 3, not only needing to summarize control node side for flow control node and obtaining and summarize flow
Be converted to reference flow Tc, it is also necessary to which flow Tx newly generated for this node carries out the calculating of total amount flow, to be step
Analysis provides judgment basis in 204.
In step 204, otherwise, right if analysis result is to let pass less than flow control threshold value to newly generated flow
Newly generated flow carries out default behaviour control.
Wherein, the default behaviour control, specifically includes: delay is sent or packet loss.
Method provided in an embodiment of the present invention, each node-flow magnitude of real-time synchronization, without the concern for the stream between each node
Whether amount is balanced, whether has fluctuation, can fast and accurately control whole flow.Between node in distributed network
Flow the adjustment even increase of node are deleted all generate whole control effect and be significantly affected.Also, implement in the present invention
In example, summarizes flow by summarize that control node is calculated in real time, be issued to the mode of each flow control node, and combine
It informs flow threshold corresponding to flow control node, realizes the self-service flow control in flow control node side, improve
Response speed and reduce the control complexity for summarizing control node.
Above step is executed always using newly generating flow as trigger condition on summarizing control node and flow control node
(including periodic and/or condition trigger-type, wherein the period is usually several milliseconds, and maximum is no more than 100 milliseconds)；Summarize
Control node can have multiple, and summarizing between control node can be with synchrodata, to achieve the purpose that reduce synchronization times；Such as figure
Shown in 4, multiple control nodes that summarize can carry out cascade and company, series winding etc. according to network topology.
The embodiment of the present invention provide that a kind of simplification summarizes control node computing resource occupancy by step 201-204 one
Kind control method of flow, i.e., wherein summarize control node and summarize to each flow control node feeding back flow, and by each flow control
Statistics flow that node is uploaded according to itself obtains the reference flow；This mode can save the meter for summarizing control node
Resource is calculated, avoids summarizing control node being that the realization of each flow control node calculates reference flow, certainly, this mode mentions
The high calculation amount of flow control node, also, also need to store flow control threshold value on flow control node, still, compare
For the computation complexity for calculating the reference flow for needing to be sent to each flow control node by summarizing control node, compare
The complexity for relatively transferring to each flow control node itself to realize, the resource occupation of the latter can be ignored.
By above-mentioned analysis, the embodiment of the invention also provides a kind of alternative schemes, belong to the same invention structure of the present invention
The implementation of think of, as shown in figure 5, in the alternative scheme, from summarizing that control node side shows for each flow
Statistics flow T1, T2 and T3 that control node reports, summarize control node after calculating total flow, further complete to implement
When summarizing flow to the transmission of each flow control node of step 202, summarizes flow for this and replaces with the ginseng calculated in advance in example 1
Flow is examined (as shown in figure 5, this is received for flow control node 1, flow control node 2 and flow control node 3
Reference flow shows as Ta-T1, Ta-T2 and Ta-T3 respectively) so that each flow control node is no longer needed in respective sheet
Ground calculates a wheel reference flow.Certainly, the optimized integration of the above-mentioned alternative scheme for embodiment 1 is to summarize control node
The far super common flow control node of calculated performance premised on.
During the embodiment of the present invention is realized, there is also the control modes that one kind more refines, and can guarantee to summarize
The correlative flow that control node can more timely get controlled data object has reached the flow threshold of setting.Specifically
, specifically it is unfolded to realize in the step 203 of the embodiment of the present invention, including following sub- content:
The analysis of flow control node determine from report statistics flow to summarizing after node (in this, as the time of subsequent judgement
Node), it increases speed meeting the correspondence flow that the flow control node is monitored in preset time more than pre-set velocity value
And/or it (is the equal of the stream that flow incrementss are corresponded in preset time more than the preset percentage of the remaining flow rate zone that can let pass
The burst that flow has occurred in amount control node increases, also, has exceeded and summarize control node for the pre-set velocity value of its setting
And/or residue can let pass flow rate zone), then updated summarize flow to summarizing control node and send at once.Wherein, when presetting
Between can be it is preset summarize the flow report cycle time, can also be and weighted on summarizing flow report cycle time basis
To time value (e.g. summarize the flow report cycle time 1/2 or 1/3 weighting after time value).Implement in the present invention
During example is realized, residue can let pass flow rate zone it is smaller when, can reduce synchronizing cycle, improve response speed；Equally exist
Residue can let pass flow rate zone it is larger when, can increase synchronizing cycle the interval of transmission (increase reported data), reduce same
Step data amount.At the same time it can also the synchronizing cycle by dynamic adjustment under normal circumstances, the occupancy of Lai Gaishan channel width, example
As: it, can be appropriate when flow control node flow growth rate caused by the data object for confirming itself monitoring is smaller
Increase synchronizing cycle；And confirm itself monitoring data object caused by flow growth rate it is larger when, can fit
When reduction synchronizing cycle (reducing the interval of reported data transmission)；The mode of the dynamic adjustment is suitable for entire flow control
The control process of node processed can be used as one of the assembled scheme of realization of the embodiment of the present invention.
Wherein, pre-set velocity value and/or residue can let pass flow rate zone can be summarize control node according to statistics flow,
The average speed and flow control threshold value of flow comprehensively consider solution and obtain, such as: simplest mode is exactly according to (flow control
Threshold value-processed summarizes flow) quantity=residue of/flow control node can let pass flow rate zone；In more preferably calculation, then
Pair the size for the statistics flow that can be further fed back according to each flow control node, gives corresponding weight proportion, i.e.,
In the higher flow control node of numerical value of the statistics flow reported itself, summarizing control node can let pass for its distribution residue
When flow rate zone, proportion space can be correspondingly improved.And for the pre-set velocity value, it can above-mentioned residue can be calculated
When clearance flow rate zone, can solving for also just synchronizing comes, and specifically can be the time by occupancy needed for feedback cycle, counts
Calculating and completing can the let pass speed of flow rate zone permission of the residue is the pre-set velocity value.
Wherein, preset percentage can be set to stepped according to the number of real surplus amount, such as: real surplus amount
Then for 5G, corresponding percentage value can be set in the section 50-60%；And when real surplus amount be 1G then, corresponding percentage
Value can be set in the section 20%-30%, more ladders can be arranged according to the demand of fineness in the actual operation process
Interval value, details are not described herein.Wherein, the setting of the pre-set velocity value, for than the setting of the percentage according to
According to increasingly complex, pre-set velocity value should at least consider two aspects in simple terms, and first aspect is the speedup journey of preset time
Degree；Second aspect is the velocity amplitude actually reached；Above-mentioned second aspect, that is, so-called pre-set velocity value, setting means can join
Be admitted to the percent value stated, realized using similar step-wise manner, i.e., surplus get over multiple preset velocity amplitude can be set it is larger,
And the surplus the few, pre-set velocity value should be arranged smaller；And above-mentioned first aspect is it may be considered that be that the staged is default
The access situation of explosive growth in the short time is further contemplated, so that being directed in the supplement reference factor of velocity amplitude
The judgement of pre-set velocity value reports update to summarize flow, has higher break-up value for control node for summarizing.
Flow control used in the embodiment of the present invention, which can be, is fine to concrete application, specific media data, specific clothes
Service type etc. is referred to as data object.Next, will be by the first data object as description object, providing one kind can tie
Close the way to manage between the preferred flow control node that the embodiment of the present invention is realized.Wherein, each flow control node is treated
The object that the flow of control is counted is the flow of the first data object, the applicable scene of corresponding way to manage are as follows: summarize
Control node is also used to the statistics flow returned according to each flow control node, confirms summarizing for corresponding first data object
Flow has reached or close to its flow control threshold value, as shown in fig. 6, the then control method further include:
In step 301, summarizing control node determines the flow of corresponding first data object in each flow control section
Distribution on point.
Wherein, the distribution refers specifically to flow relevant to the first data object, on each flow control node
Partition density, such as: using framework shown in Fig. 2 as description object, the relevant flow of opposite first data object is flowing
Measuring the specific gravity in control node 1 is 30%, and the specific gravity on flow control node 2 is 45%, the ratio on flow control node 3
Weight is 25%.In a practical situation, because of the distribution of flow control node and the diversity of way to manage, first data pair
As the partition density of relevant flow can be more increasingly complex than the example above, moreover, what some flow control nodes even obtained
Proportion is 0.
In step 302, one by the flow occupancy of first data object less than the first preset threshold or
The traffic service task that first data object is corresponded in multiple flow control nodes is transferred to other flow occupancies and is greater than
Equal to the flow control node of the first preset threshold.
Wherein, for the selection for the flow control node being transferred into, in addition to above-mentioned consideration occupancy is more than or equal to first in advance
If outside the flow control node of threshold value, in conjunction with specific implementation environment, usually also needing its occupancy simultaneously less than the second default threshold
Value ensure that the flow control node being transferred into has enough accommodation spaces.Wherein, first preset threshold and second
Preset threshold can be set according to the actual situation, can be obtained monitoring flow with flow control node each in actual conditions
As a result it is associated, in order to reach the normal operating conditions for guaranteeing the flow control node for being unlikely to excessively to influence to be transferred into, again
It can reduce as far as possible by the traffic monitoring of first data object, and be reported to bring channel strip when summarizing control node
Wide occupancy.When reporting sufficiently high frequency of flow of statistics, in the correlative flow of the first data object, in each flow control
When distribution situation in node processed is excessively dispersed, the improvement project that the embodiment of the present invention that compares is proposed, can extra band come compared with
The occupancy of big channel width.
In conjunction with the embodiment of the present invention, in order to further embody the flow control node being transferred into for by first data pair
As the special relationship of the correlative flow of generation, it is preferred that the control node that summarizes completes corresponding first data object
Traffic service task transfer after, the control method further include:
The flow for summarizing control node setting and being currently owned by the traffic service task of corresponding first data object
Control node is the exclusive node of the traffic service task of first data object.Exclusive node herein is meant that, subsequent
The related access for corresponding to first data object all completed via the exclusive node.In order to guarantee whole network
The flexibility of control and anti-robustness, the above-mentioned imparting for relating to exclusive node permission, reasonable situation are to give to design accordingly
Permission reclaim mechanism, specific: the control node that summarizes removes flow control section when reaffirming that summarizing flow is zero
The mark content of corresponding exclusive node in point.
Embodiment 2:
A kind of distributed network flow controlling party and related expanding scheme are proposed in the embodiment of the present invention 1, based on implementing
On the basis of example 1, the embodiment of the present invention 2 then proposes the improvement side that can be realized in combination with example 1 from Precise control angle emphatically
Case.Specifically, it is described summarize each node flow that control node will receive and sum to obtain summarize flow, and summarize flow for described
It synchronizes when being sent to each flow control node, also by the average speed of flow to be controlled in each flow control node being calculated
It synchronizes together and is sent to flow control node；Wherein, the average speed is by summarizing control node according to the carrying got
Having time label, and summarize flow from each flow control node.
When then each flow control node generates new flow to be controlled, newly generated flow is analyzed plus the reference stream
It measures, and whether the sum of pre-configuration flow caused by the average speed plus the flow to be controlled is more than flow control threshold
Value；If analyzing result is to let pass less than flow control threshold value to newly generated flow, otherwise, newly generated flow is carried out
Default behaviour control.In embodiment 1 and the embodiment of the present invention 2, described flow control threshold value and specific data object
Between be usually one-to-one relationship, in the actual implementation process, summarize control node can with each flow control node it
Between safeguard multiple data objects simultaneously.
Further, in conjunction with various embodiments of the present invention, there is also a kind of preferred expansion schemes, wherein the statistic fluid
Identification information in amount also comprising accessed data object, then it is described to summarize control node and be also used to summarize each data object quilt
The number of access；For meeting any one following condition, then summarize control node for the flow of corresponding data object
Interface is assigned to one or more flow control node, and generates each flow control section that assignment information is sent under its administration
Point；Can be confirmed whether to forward access request to one be assigned according to the assignment information so as to each flow control node or
The multiple flow control nodes of person.Situation includes:
Situation one, accessed number are more than third predetermined threshold value.Wherein, third predetermined threshold value is for characterizing the data pair
The state that is concerned is needed as the number not accessed has reached, at this point, summarizing control node can be according to the class of data object
Type and characteristic select the flow control node of characteristics match therewith to provide data flow service, to improve the fine of whole network
Change controlled level.
Situation two, accessed number are in default ranking section.Wherein, for situation two-phase comparable situation one, it is more suitable for
In having the quantity for the flow control node for supporting above-mentioned specific assignment task to be limited, at this point, can only be according to the side of ranking
Formula, selection are wherein accessed amount in pre-set interval, execute the processing mode of above-mentioned appointment task.
It is more than the 4th preset threshold that number is accessed in situation three, preset time range.The situation three and situation one have
Have biggish similitude, still, herein the case where three be more growth rate for accessed data object, that is, determine
Growth rate is more than the 4th preset threshold, then regards as needing to be assigned to specific flow control node to be monitored.
Wherein, for situation three, the flow control node being assigned can be the equipment with more high analyte performance, and corresponding
Preset time range in be accessed the data object that number is more than the 4th preset threshold and may be considered the number attacked of suspection
According to object, the design parameter value of the 4th preset threshold will be far more than other parameter values for normally doing popularity judgement at this time.
Further, after in order to illustrate above-mentioned setting, the case where being encountered in specific execution flowchart process, such as: it is common
Flow control node (be the flow control node for distinguishing and being assigned and rise nickname) is as the first intelligent terminal
When accessing access interface, detect that the data access of the first intelligent terminal is consistent with the data object in the assignment information, then
The access request for the data object of first intelligent terminal is transmitted to one or more being assigned
Flow control node.
In embodiments of the present invention, it is with the following functions that control node setting can also be summarized for described in: statistics is for each
The quantity of the data object of the appointment of flow control node, and according to the flow control threshold value and each flow control of each data object
The processing capacity of node completes the distribution of each appointment.
Embodiment 3:
Referring to Fig. 7, being the configuration diagram of the distributed network flow control device of the embodiment of the present invention.The present embodiment
Distributed network flow control device include one or more processors 21 and memory 22.Wherein, at one in Fig. 7
For reason device 21.
Processor 21 can be connected with memory 22 by bus or other modes, to be connected as by bus in Fig. 7
Example.
Memory 22 is used as a kind of distributed network flow control method and device non-volatile computer readable storage medium
Matter can be used for storing non-volatile software program, non-volatile computer executable program, such as the distributed network in embodiment 1
Network flow control methods and corresponding program instruction (for example, Fig. 1 and method flow shown in fig. 6).Processor 21 passes through fortune
The capable non-volatile software program being stored in memory 22, instruction, thereby executing each of distributed network flow control device
The distributed network flow control method of embodiment 1 is realized in kind functional application and data processing.
Memory 22 may include high-speed random access memory, can also include nonvolatile memory, for example, at least
One disk memory, flush memory device or other non-volatile solid state memory parts.In some embodiments, memory 22
Optional includes the memory remotely located relative to processor 21, these remote memories can pass through network connection to processor
21.The example of above-mentioned network includes but is not limited to internet, intranet, local area network, mobile radio communication and combinations thereof.
Described program instruction is stored in the memory 22, when being executed by one or more of processors 21,
The distributed network flow control method in above-described embodiment 1 is executed, for example, executing Fig. 1 described above and shown in fig. 6 each
A step.
It is worth noting that the contents such as information exchange, implementation procedure between unit in above-mentioned apparatus, due to this hair
Bright processing method embodiment 1 and embodiment 2 is based on same design, and particular content can be found in chatting in embodiment of the present invention method
It states, details are not described herein again.
Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of embodiment is can to lead to
Program is crossed to instruct relevant hardware and complete, which can be stored in a computer readable storage medium, storage medium
It may include: read-only memory (ROM, Read Only Memory), random access memory (RAM, Random Access
Memory), disk or CD etc..
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (9)
1. a kind of distributed network flow control method, which is characterized in that in distributed network include summarize control node and with
Its multiple flow control node connected, flow control node are serially connected in the network to be controlled, and control method includes:
Each flow control node counts flow to be controlled, and statistics flow is sent to and summarizes control node；
Summarize control node and sum to obtain by each node flow received and summarize flow, and by it is described summarize flow and synchronize be sent to
Each flow control node；Wherein, when being synchronized to each flow control node and summarizing flow, each flow control node summarizes stream for described
Amount subtracts the flow itself reported and obtains reference flow；
When each flow control node generates new flow to be controlled, analyzing newly generated flow plus the sum of described reference flow is
No is more than flow control threshold value；
If analyzing result is to let pass less than flow control threshold value to newly generated flow, otherwise, newly generated flow is carried out
Default behaviour control；
It is described summarize each node flow that control node will receive and sum to obtain summarize flow, and summarize the synchronous hair of flow for described
It is also that the average speed of flow to be controlled in each flow control node being calculated is same together when giving each flow control node
Step is sent to flow control node；
When then each flow control node generates new flow to be controlled, analyzes newly generated flow and add the reference flow, with
And whether the sum of pre-configuration flow caused by the average speed plus the flow to be controlled is more than flow control threshold value；If point
Analysing result is then to let pass to newly generated flow less than flow control threshold value, otherwise, carries out default behavior to newly generated flow
Control.
2. distributed network flow control method according to claim 1, which is characterized in that each flow control node generates
When new flow to be controlled, whether be more than flow control threshold value, also if analyzing newly generated flow plus the sum of described reference flow
Include:
The analysis of flow control node is determined from reporting statistics flow meeting the flow control in preset time to after summarizing node
The correspondence flow that node processed is monitored is increased speed more than corresponding to flow incrementss in pre-set velocity value and/or preset time
Residue can let pass the preset percentage of flow rate zone, then updated summarize flow to summarizing control node and send at once.
3. distributed network flow control method according to claim 1, which is characterized in that each flow control node is treated
The object that the flow of control is counted is the flow of the first data object, then summarizes control node and be also used to according to each flow control
The flow that summarizes of the statistics flow that node processed returns, corresponding first data object of confirmation has reached or close to its flow
Control threshold, the then control method further include:
Summarize control node and determines distribution of the flow of corresponding first data object on each flow control node；
One or more flow control node by the flow occupancy of first data object less than the first preset threshold
The traffic service task of middle correspondence first data object is transferred to other flow occupancies more than or equal to the first preset threshold
Flow control node.
4. distributed network flow control method according to claim 3, which is characterized in that described by first data
The flow occupancy of object is less than corresponding to first data in one or more flow control node of the first preset threshold
The traffic service task of object is transferred to the flow control node that other flow occupancies are more than or equal to the first preset threshold, also
Include:
The flow control node being transferred into, after obtaining the traffic service task for being transferred to and corresponding to first data object, statistics
The flow of first data object of correspondence out, which needs to meet, is more than or equal to the first preset threshold, and less than the second default threshold
Value.
5. distributed network flow control method according to claim 3, which is characterized in that the control node that summarizes exists
After the traffic service task transfer for completing corresponding first data object, the control method further include:
The flow control for summarizing control node setting and being currently owned by the traffic service task of corresponding first data object
Node is the exclusive node of the traffic service task of first data object；
The control node that summarizes removes corresponding exclusive node in flow control node when reaffirming that summarizing flow is zero
Identify content.
6. -5 any distributed network flow control method according to claim 1, which is characterized in that the statistics flow
In identification information also comprising accessed data object, then it is described to summarize control node to be also used to summarize each data object interviewed
The number asked；It is more than third predetermined threshold value, accessed number in default ranking section, preset time model for accessed number
Enclosing interior accessed number is more than that any one condition in the 4th preset threshold meets, then summarizes control node for corresponding number
It is assigned to one or more flow control node according to the traffic interface of object, and generates assignment information and is sent under its administration
Each flow control node；So that each flow control node can be confirmed whether to forward access request to quilt according to the assignment information
One or more flow control node assigned.
7. distributed network flow control method according to claim 6, which is characterized in that common discharge control node exists
When access access interface as the first intelligent terminal, detect in data access and the assignment information of the first intelligent terminal
Data object it is consistent, then the access request for the data object of first intelligent terminal is transmitted to described referred to
One or more flow control node of group.
8. distributed network flow control method according to claim 1, which is characterized in that described to summarize control node also
For counting the quantity of the data object of the appointment for each flow control node, and according to the flow control threshold of each data object
Value and the processing capacity of each flow control node complete the distribution of each appointment.
9. a kind of distributed network flow control device, which is characterized in that described device includes:
At least one processor；And the memory being connect at least one described processor communication；Wherein, the memory
It is stored with the instruction that can be executed by least one described processor, described instruction is arranged to carry out claim 1-8 by program and appoints
Distributed network flow control method described in one.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810863592.8A CN109150740B (en) | 2018-08-01 | 2018-08-01 | A kind of distributed network flow control method and device |
| CN201910595290.1A CN110224944B (en) | 2018-08-01 | 2018-08-01 | Distributed network flow control method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810863592.8A CN109150740B (en) | 2018-08-01 | 2018-08-01 | A kind of distributed network flow control method and device |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910595290.1A Division CN110224944B (en) | 2018-08-01 | 2018-08-01 | Distributed network flow control method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109150740A CN109150740A (en) | 2019-01-04 |
| CN109150740B true CN109150740B (en) | 2019-09-10 |
Family
ID=64799422
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910595290.1A Active CN110224944B (en) | 2018-08-01 | 2018-08-01 | Distributed network flow control method and device |
| CN201810863592.8A Active CN109150740B (en) | 2018-08-01 | 2018-08-01 | A kind of distributed network flow control method and device |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910595290.1A Active CN110224944B (en) | 2018-08-01 | 2018-08-01 | Distributed network flow control method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN110224944B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110034956A (en) * | 2019-03-27 | 2019-07-19 | 广州供电局有限公司 | Network Data Control method, apparatus, computer equipment and storage medium |
| CN118509315A (en) * | 2020-04-28 | 2024-08-16 | 腾讯科技（深圳）有限公司 | Network resource parameter configuration method, device and computer readable storage medium |
| CN111770150B (en) * | 2020-06-24 | 2022-10-18 | 中国工商银行股份有限公司 | Access flow control method and device and electronic equipment |
| CN113300966B (en) * | 2020-07-27 | 2024-05-28 | 阿里巴巴集团控股有限公司 | Flow control method, device, system and electronic equipment |
| CN112003790B (en) * | 2020-08-26 | 2021-06-01 | 上海松鼠课堂人工智能科技有限公司 | Distribution method of network traffic used by intelligent school |
| CN112615794B (en) * | 2020-12-08 | 2022-07-29 | 四川迅游网络科技股份有限公司 | Intelligent acceleration system and method for service flow characteristics |
| CN114710447B (en) * | 2020-12-16 | 2024-05-03 | 百寻信息科技(上海)有限公司 | Balanced advertisement flow control method and system |
| CN113328906B (en) * | 2021-04-22 | 2023-01-06 | 成都欧珀通信科技有限公司 | Flow real-time monitoring method and device, storage medium and electronic equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101286937A (en) * | 2008-05-16 | 2008-10-15 | 华为技术有限公司 | Network flow control method, device and system |
| CN102970240A (en) * | 2012-11-01 | 2013-03-13 | 杭州华三通信技术有限公司 | Flow balancing method and device in software package builder (SPB) network |
| CN106921584A (en) * | 2017-03-31 | 2017-07-04 | 武汉绿色网络信息服务有限责任公司 | A kind of distributed network flow control method |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2017130840A (en) * | 2016-01-21 | 2017-07-27 | 富士通株式会社 | Network switch, network controller, and network system |
| CN107222426B (en) * | 2016-03-21 | 2021-07-20 | 阿里巴巴集团控股有限公司 | Flow control method, device and system |
| CN106100999B (en) * | 2016-08-28 | 2019-05-24 | 北京瑞和云图科技有限公司 | Image network flow control methods in a kind of virtualized network environment |
-
2018
- 2018-08-01 CN CN201910595290.1A patent/CN110224944B/en active Active
- 2018-08-01 CN CN201810863592.8A patent/CN109150740B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101286937A (en) * | 2008-05-16 | 2008-10-15 | 华为技术有限公司 | Network flow control method, device and system |
| CN102970240A (en) * | 2012-11-01 | 2013-03-13 | 杭州华三通信技术有限公司 | Flow balancing method and device in software package builder (SPB) network |
| CN106921584A (en) * | 2017-03-31 | 2017-07-04 | 武汉绿色网络信息服务有限责任公司 | A kind of distributed network flow control method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109150740A (en) | 2019-01-04 |
| CN110224944B (en) | 2021-10-08 |
| CN110224944A (en) | 2019-09-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109150740B (en) | A kind of distributed network flow control method and device | |
| Tizghadam et al. | Betweenness centrality and resistance distance in communication networks | |
| Dias et al. | Online traffic-aware virtual machine placement in data center networks | |
| CN110177054B (en) | Port queue scheduling method, device, network controller and storage medium | |
| CN107005480A (en) | The system and method cooperated for SDT and NFV and SDN | |
| SA517390145B1 (en) | Systems, Methods, and Computer Medium to Provide Adaptive Priority Scheduling of Communications Over A Network | |
| CN105229973A (en) | The formation of network and the visualization device of action | |
| CN103248521B (en) | Method, device and the communication system of a kind of business game rule configuration | |
| CN108809707A (en) | A kind of TSN dispatching methods towards real-time application demand | |
| CN106921584A (en) | A kind of distributed network flow control method | |
| CN104202264A (en) | Carrying resource allocation method for clouded data center network, device and system | |
| CN105007287A (en) | public cloud resource dynamic configuration method and system | |
| CN106301952A (en) | A kind of SDN datum plane link backup method and device | |
| CN105812217A (en) | Virtual network division method and multi-controller agent device | |
| CN104253767A (en) | Virtual shard network implementing method and switch | |
| CN103997432A (en) | Measuring system and method for supporting analysis of OpenFlow application performance | |
| CN112153153B (en) | Coordinated distributed intra-network resource scheduling method and system and storage medium | |
| CN109660624A (en) | Planing method, server and the storage medium of content distributing network resource | |
| CN113746750A (en) | Method, network node, system and storage medium for forwarding data message | |
| Liu et al. | Mapping distributed application SLA to network QoS parameters | |
| CN109617806A (en) | A kind of data traffic dispatching method and device | |
| CN104683252B (en) | A kind of gateway applied to gaming network is connected into method and system | |
| CN102752804B (en) | Communication resource allocation systems | |
| CN102448120A (en) | Multipath load sharing method and device | |
| Buzhin et al. | Evaluation of Telecommunication Equipment Delays in Software-Defined Networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |