CN102868671A - Method, equipment and system for controlling network congestion - Google Patents

Method, equipment and system for controlling network congestion Download PDF

Info

Publication number
CN102868671A
CN102868671A CN2011101914957A CN201110191495A CN102868671A CN 102868671 A CN102868671 A CN 102868671A CN 2011101914957 A CN2011101914957 A CN 2011101914957A CN 201110191495 A CN201110191495 A CN 201110191495A CN 102868671 A CN102868671 A CN 102868671A
Authority
CN
China
Prior art keywords
data flow
network
network node
congested
input reference
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
Application number
CN2011101914957A
Other languages
Chinese (zh)
Other versions
CN102868671B (en
Inventor
张登银
陈齐标
程春玲
李正
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Nanjing Post and Telecommunication University
Nanjing University of Posts and Telecommunications
Original Assignee
Huawei Technologies Co Ltd
Nanjing Post and Telecommunication University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd, Nanjing Post and Telecommunication University filed Critical Huawei Technologies Co Ltd
Priority to CN201110191495.7A priority Critical patent/CN102868671B/en
Priority claimed from CN201110191495.7A external-priority patent/CN102868671B/en
Priority to PCT/CN2012/078375 priority patent/WO2013007180A1/en
Publication of CN102868671A publication Critical patent/CN102868671A/en
Application granted granted Critical
Publication of CN102868671B publication Critical patent/CN102868671B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
    • H04L47/2425—Traffic characterised by specific attributes, e.g. priority or QoS for supporting services specification, e.g. SLA
    • 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
    • H04L47/11—Identifying congestion

Abstract

The invention provides a method, equipment and a system for controlling network congestion. The method comprises that a congestion judge parameter and an input reference speed corresponding to data flow are obtained by an entrance network node, the congestion judge parameter is determined according to accumulated matters, and the input reference speed is the most excellent solution which guarantees the non-congestion of a network; and occurrence of the network congestion is judged according to the congestion judge parameter by the entrance network node, and if the network congestion occurs, according to SLA (service level arrangement) information corresponding to the input reference speed and the data flow, the data flow is subjected to network congestion control. By the adoption of the method, the equipment and the system of the embodiment, the situation of congestion on an overall chain can be effectively relieved, aspiration of a user can be reflected, and the network utility is high.

Description

Method for controlling network congestion, equipment and system
Technical field
The present invention relates to the network communications technology, relate in particular to a kind of method for controlling network congestion, equipment and system.
Background technology
Network congestion is comparison distinct issues in the current network, utilizes congestion control to guarantee that network service quality (Quality of Service, QoS) is very important.Network congestion can show as packet packetization delay increase, packet loss increase, upper layer application hydraulic performance decline etc.
Existing Research of Congestion Control Techniques, can be divided into two classes in the position of network layer according to congestion control, one class is based on transmission control protocol (Transport Control Protocol, the TCP) congestion control of source, carries out in main frame and network edge device; The another kind of IP congestion control that is based on network is carried out in the network equipment.Network congestion control also can be called admits control, and corresponding admission control algorithm comprises: based on the admission control algorithm of parameter, based on the admission control algorithm of measuring, based on the admission control algorithm of bandwidth broker with based on the admission control algorithm of strategy.
Existing Research of Congestion Control Techniques can only reflect the congestion situation of individual node in the network, can not effectively alleviate the congestion situation of whole piece link.Existing admission control algorithm can not embody user intention, and network utilization is not high.
Summary of the invention
The embodiment of the invention provides a kind of method for controlling network congestion, equipment and system, effectively alleviates the congestion situation of whole piece link, and can embody user intention and network utilization height.
The embodiment of the invention provides a kind of method for controlling network congestion, comprising:
The entrance network node obtains congested judgement parameter corresponding to data flow and input reference rate, and described congested judgement parameter determines that according to accumulation described input reference rate is for guaranteeing the not congested extremely excellent solution of network;
The entrance network node determines whether to occur network congestion according to described congested judgement parameter, if network congestion occurs, according to described input reference rate and SLA information corresponding to described data flow, described data flow is carried out network congestion control.
The embodiment of the invention provides a kind of network congestion control appliance, comprising:
Acquisition module is used for obtaining congested judgement parameter corresponding to data flow and input reference rate, and described congested judgement parameter determines that according to accumulation described input reference rate is for guaranteeing the not congested extremely excellent solution of network;
Control module is used for determining whether to occur network congestion according to described congested judgement parameter, if network congestion occurs, according to described input reference rate and SLA information corresponding to described data flow, described data flow is carried out network congestion control.
The embodiment of the invention provides a kind of network congestion control system, comprising:
Network agent, be used for obtaining the input rate of entrance network node, and the output speed of obtaining the outlet network node, determine accumulation according to described input rate and output speed, determine congested judgement parameter according to described accumulation, and determine to guarantee that the not congested extremely excellent solution of network is as the input reference rate;
The entrance network node is used for determining whether to occur network congestion according to described congested judgement parameter, if network congestion occurs, according to described input reference rate and SLA information corresponding to described data flow, described data flow is carried out network congestion control.
By above-mentioned technical method as can be known, the embodiment of the invention is by judging that according to congested parameter judges whether network is congested, and congested judgement parameter is determined according to accumulation, accumulation can embody the congestion situation of whole network, therefore, the embodiment of the invention can reflect the congestion situation of whole network and not merely be the congestion situation of a node, thereby effectively alleviate the congestion condition of whole link.By employing SLA when network congestion is controlled, and SLA is the embodiment of user intention, so the embodiment of the invention can reflect user intention; By adopting the input reference rate when network congestion is controlled, the extremely excellent solution of this input reference rate when guaranteeing that whole network is not congested can be increased network utilization on the not congested basis of network to greatest extent guaranteeing.
Description of drawings
In order to be illustrated more clearly in the technical scheme in the embodiment of the invention, the accompanying drawing of required use was done a simply introduction during the below will describe embodiment, apparently, accompanying drawing in the following describes is some embodiments of the present invention, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain according to these accompanying drawings other accompanying drawing.
Fig. 1 is the method flow schematic diagram of first embodiment of the invention;
Fig. 2 is the structural representation of the system of second embodiment of the invention;
Fig. 3 is method flow schematic diagram corresponding to second embodiment of the invention;
Fig. 4 is the structural representation of the system of third embodiment of the invention;
Fig. 5 is the handling process schematic diagram of congested judge module in the embodiment of the invention;
Fig. 6 is the structural representation of network topology in the territory in the embodiment of the invention;
Fig. 7 is the method flow schematic diagram of dynamic accommodation control in the embodiment of the invention;
Fig. 8 is the method flow schematic diagram of third embodiment of the invention;
Fig. 9 is the device structure schematic diagram of fourth embodiment of the invention;
Figure 10 is the system configuration schematic diagram of fifth embodiment of the invention.
Embodiment
For the purpose, technical scheme and the advantage that make the embodiment of the invention clearer, below in conjunction with the accompanying drawing in the embodiment of the invention, technical scheme in the embodiment of the invention is clearly and completely described, obviously, described embodiment is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills belong to the scope of protection of the invention not making the every other embodiment that obtains under the creative work prerequisite.
Fig. 1 is the method flow schematic diagram of first embodiment of the invention, comprising:
Step 11: the entrance network node obtains congested judgement parameter corresponding to data flow and input reference rate, and described congested judgement parameter determines that according to accumulation described input reference rate is for guaranteeing the not congested extremely excellent solution of network;
Wherein, the entrance network node can obtain above-mentioned congested judgement parameter and input parameter speed from network agent.At this moment, network agent can obtain the input rate of described entrance network node, and the output speed of obtaining the outlet network node, determines described accumulation according to described input rate and output speed.And then determine congested judgement parameter according to accumulation, and determine to show the function of network congestion, ask for afterwards the extremely excellent solution of this function as the input reference rate.The formula of calculated product polymers, congested judgement parameter and input reference rate can be referring to following embodiment.
Step 12: the entrance network node determines whether to occur network congestion according to described congested judgement parameter, if network congestion occurs, according to described input reference rate and SLA information corresponding to described data flow, described data flow is carried out network congestion control.
For example, the static state degradation be need to carry out if SLA information corresponding to described data flow shows, then described data flow threshold value corresponding to described static degradation and the smaller value in the described input reference rate are adjusted in the output speed of entrance network node; Perhaps, if SLA information corresponding to described data flow show and need to dynamically demote, then described data flow is adjusted into described input reference rate in the output speed of entrance network node; Perhaps, need to stop current business if SLA information corresponding to described data flow shows, then stop the link of described data flow.
Present embodiment is by judging that according to congested parameter judges whether network is congested, and congested judgement parameter is determined according to accumulation, accumulation can embody the congestion situation of whole network, therefore, can reflect the congestion situation of whole network and not merely be the congestion situation of a node, thereby effectively alleviate the congestion condition of whole link.By employing SLA when network congestion is controlled, and SLA is the embodiment of user intention, therefore can reflect user intention; By adopting the input reference rate when network congestion is controlled, the extremely excellent solution of this input reference rate when guaranteeing that whole network is not congested can be increased network utilization on the not congested basis of network to greatest extent guaranteeing.
Fig. 2 is the structural representation of the system of second embodiment of the invention, and Fig. 3 is method flow schematic diagram corresponding to second embodiment of the invention.
Referring to Fig. 2, comprise source 21, entrance network node 22, outlet network node 23, destination 24 and network agent 25.
Wherein, source 21 is used for proposing service request, sends data flow to entrance network node 22.Source 21 can be passed through existing FAST Transmission Control Protocol, carries out the source congestion control according to congestion window size.
To be data flow enter gateway or the network edge node of network from source to entrance network node 22, and the control of complete paired data stream realizes that resource distributes.Wherein, the control of data flow can comprise: based on the static current limliting of service-level agreement (Service-Level Agreement, SLA), based on the dynamic accommodation control of SLA and input reference rate.Particularly, entrance network node 22 is after receiving data flow, if determine not congested according to congested judgement parameter, after perhaps not having congested judgement parameter, namely under normal circumstances, the SLA information of signing according to user and operator, for the IP address of carrying in each data flow or with the user name of IP address dynamic binding, carry out the current limliting of front end, in order to avoid unique user sends excessive flow; In the rear end of entrance network node, the data flow of having classified is carried out traffic shaping, in order to avoid the flow of certain class business takies too much bandwidth.When determining that according to congested judgement parameter generation is congested, then carry out dynamic accommodation control, to carry out congestion control.Wherein, dynamic accommodation control can comprise: static state is demoted, is dynamically demoted or stops current business.If static the degradation then adjusted to the flow velocity of data flow the static threshold corresponding to SLA information of prior signing; If dynamically demote, then adjust the flow velocity of data flow according to the input reference rate; If stop current business, then stop the link of this professional data flow.Further, if after adopting static degradation, flow velocity can dynamically be demoted again greater than the input reference rate.In addition, entrance network node 22 also sends to network agent 25 with the input rate of self.
Outlet network node 23 is data flow enters destination 24 from network gateway or network edge nodes.In addition, outlet network node 23 also sends to network agent 25 with the output speed of self.
Network agent 25 is used for obtaining the input rate of entrance network node 22 and the egress rate of outlet network node 23, and calculates accordingly congested judgement parameter and input reference rate, afterwards congested judgement parameter and input reference rate is sent to the entrance network node.
In addition, data flow can enter network by entrance network node 22 by source 21, and entrance network node 22 sends to outlet network node 23 with this data flow by network afterwards, and then sends to destination 24 by outlet network node 23.
Referring to Fig. 3, the method for present embodiment can comprise:
Step 31: source sends data flow to network boundary entrance network node.
Wherein, can carry out control based on congestion window in source.
Step 32: the entrance network node is by the network intermediate equipment and export the target node data flow to destination.
For example, under normal circumstances, carry out static current limliting based on SLA information, according to SLA information corresponding to data flow, this SLA information comprises corresponding flow velocity, afterwards according to this flow velocity data streams.
Step 33: network agent obtains the input rate of entrance network node and the output speed of outlet network node.
Can be that entrance network node active Gather and input speed and outlet network node initiatively gather output speed, send respectively afterwards each network agent; Also can be that network agent is from entrance network node and outlet network node difference active obtaining input rate and output speed.
Step 34: network agent calculates congested judgement parameter and input reference rate;
Wherein, network agent can be according to the input rate of every data flow and congested judgement parameter and the input reference rate of every data flow of output speed calculating.Concrete computational process can be referring to embodiment illustrated in fig. 5.
Step 35: network agent sends to the entrance network node with congested judgement parameter and input reference rate.
Wherein, can be that network agent initiatively sends to the entrance network node with congested judgement parameter and input reference rate, also can be that network agent sends to the entrance network node with congested judgement parameter and input reference rate after the request that receives the entrance network node.
Step 36: the entrance network node carries out dynamic accommodation control.
Wherein, if the entrance network node determine to occur can to carry out dynamic accommodation control after congested according to congested judgement parameter.For example, the SLA information corresponding according to data flow is determined static degradation, dynamically demotes or is stopped current business, carries out afterwards corresponding the processing.Concrete processing procedure can be referring to embodiment illustrated in fig. 7.
Step 37: the entrance network node sends the data flow after the dynamic accommodation control.By admitting control, data flow reduces, congestion relief.
Fig. 2,3 has provided embodiment of the invention signal on the whole, and in the specific implementation, the said equipment can specifically comprise the module shown in the following embodiment.
Fig. 4 is the structural representation of the system of third embodiment of the invention, and in the present embodiment, source 41 can comprise congestion window estimation block 411 and congestion window control module 412; Congestion window estimation block 411 is measured or is calculated for the parameter of determining congestion window size, calculates the size of congestion window with this.Congestion window control module 412 is controlled package forward speed according to the size of the congestion window that this congestion window estimation block calculates, so that package forward speed is no more than the size of congestion window.
Entrance network node 42 comprises static current limliting module 421, dynamic accommodation control module 422, the first flow velocity acquisition module 423.Static current limliting module 421 is carried out static current limliting to data stream under normal circumstances according to the SLA information of obtaining from the SLA database; Dynamic accommodation control module 422 is carried out dynamic accommodation control after generation is congested.The first flow velocity acquisition module 423 is used for the input rate of entrance network node is sent to network agent.In addition, entrance network node 42 can also comprise metering data library module 424 and accounting module 425, in order to client's consumption service is carried out charging.
Outlet network node 43 can comprise the second flow velocity acquisition module 431, is used for sending output speed to network agent.
Network agent 44 can comprise SLA database 441 and congested judge module 442; SLA database 441 is used for SLA information corresponding to save data stream, and congested judge module 442 is used for obtaining congested judgement parameter and input reference rate.
Congested judge module is positioned on the network agent, is used for calculating congested judgement parameter and input reference rate.Simultaneously, the congested judgement parameter that calculates and input reference rate are offered the dynamic accommodation control module.After congested judge module need to be collected flow rate information from the first flow velocity acquisition module and the second flow velocity acquisition module, calculate the accumulation of each data flow, congested judgement parameter and input reference rate.The concrete handling process of congested judge module can be referring to figure below.
In addition, when data stream transmitting, this data flow is to send to the outlet network node by the entrance network node, therefore, has annexation between entrance network node and the outlet network node as shown in Figure 2.But, mainly be for the calculating of each parameter in the present embodiment and control accordingly that therefore, the annexation between entrance network node and the outlet network node does not embody.But, be understandable that, if when between entrance network node and the outlet network node transfer of data being arranged, have annexation between entrance network node and the outlet network node.
Fig. 5 is the handling process schematic diagram of congested judge module in the embodiment of the invention, comprising:
Step 501: obtain the input rate of entrance network node and the output speed of outlet network node.
Wherein, λ iThe input rate of expression data flow i, u iThe output speed of expression data flow i.
Step 502: according to input rate and output speed, the calculated product polymers.
Wherein, accumulation is for a link, in all nodes of data flow process, and the set of the packet of waiting in line.
Take network system shown in Figure 6 as example, accumulation can calculate in the following way:
Fig. 6 is the structural representation of network topology in the territory in the embodiment of the invention, wherein, and λ IjExpression data flow i is at the input rate at bottleneck j place, u IjExpression data flow i is in the output speed at bottleneck j place.N is the sequence number of sense cycle, and τ is the data sampling time interval.
Data flow i represents through input and output speed difference such as the formula (1) at bottleneck j place:
q ij(n)＝λ ij(n)-u ij(n) (1)
Accumulation such as formula (2) calculate:
a i ( n ) = Σ j q ij ( n ) × τ - - - ( 2 )
The recurrence formula of accumulation such as formula (3) expression:
a i(n)＝a i(n-1)+(λ i(n)-u i(n))×τ (3)
That is, for data flow i, when cycle n, accumulation is: a i(n)=a i(n-1)+(λ i(n)-u i(n)) * τ.
Step 503: whether judge accumulation greater than threshold value, if, execution in step 504, otherwise, execution in step 505.
Wherein, this threshold value b can set, and for example sets according to physical characteristic or artificial experience.
Objectively, when accumulation during greater than threshold value, show the generation network congestion, otherwise network congestion does not occur.In the embodiment of the invention, network agent is determined congested judgement parameter according to accumulation, and then the entrance network node can congestedly judge parameter judges whether network is congested according to this, to control accordingly.
Step 504: congested judgement parameter is set to r, and execution in step 507 afterwards.
Wherein, r can be the value of arbitrary setting, and this r can be 2 or 3 usually.
Step 505: judge congestedly judge whether parameter is 0, if, execution in step 508, otherwise, execution in step 506.
Step 506: congested judgement parameter is subtracted 1.
The computing formula that can obtain congested judgement parameter by above-mentioned steps is as follows:
Wherein, K i(n) the expression congested judgement parameter of data flow i when cycle n, its computing formula can for:
Figure BDA0000074669160000091
Wherein, n is the sequence number of sense cycle, K i(n) the congested judgement parameter of data flow i in n sense cycle of expression, r and b are predefined value, a i(n) accumulation of data flow i in n sense cycle of expression, its computing formula is:
a i(n)＝a i(n-1)+(λ i(n)-u i(n))×τ
λ i(n) input rate of data flow i in n sense cycle of expression, u i(n) output speed of expression data flow i in n sense cycle, τ is the data sampling time interval.
Threshold value b represents that each stream under the balanced attitude attempts the queue length that will keep in each buffering area of network in the formula.The value of b can be defaulted as the value of determining according to the equipment physical characteristic, and for example, the memory size of physical equipment is 512K, and then b can be defaulted as 512K.Certainly, the value of b also can artificially be set by the network manager, and the network manager of this moment can select less than the value of the physical characteristic value as this b according to practical experience.
Work as a i(n)＞K namely occurs when congested in b i(n) be the number of times r that links up with regard to assignment.R is illustrated in and does not occur in the continuous r period tau to assert that just current network conditions is not for congested when congested.The value of r can be the value of arbitrary setting, usually can be taken as 2 or 3.
In one-period τ, if a i(n)≤and during b, K i(n) value subtracts 1.Only at continuous r in the cycle, all do not occur congested, K i(n) just can become 0.
Step 507: judge congestedly judge whether parameter is 0, if, execution in step 508, otherwise execution in step 511.
Step 508: whether judge actual input rate less than or equal to the input reference rate, if, execution in step 509, otherwise execution in step 510.
Wherein, actual input rate refers to the input rate λ of entrance network node i, input reference rate ω i(n) initial value is arranged, can adjust afterwards, its computing formula can be as follows.
Step 509: keep the input reference rate constant, afterwards repeated execution of steps 501 and subsequent step thereof.
Step 510: will input reference rate and increase, afterwards repeated execution of steps 501 and subsequent step thereof.
Step 511: will input reference rate and reduce, afterwards repeated execution of steps 501 and subsequent step thereof.
That is, input reference rate ω i(n) computing formula is:
Figure BDA0000074669160000101
Wherein, n is the sequence number of sense cycle, ω i(n) be the input reference rate of data flow i in n the sense cycle, v iBe the value corresponding with the SLA of data flow i, λ i(n) input rate of data flow i in n sense cycle of expression, u i(n) output speed of data flow i in n sense cycle of expression, K i(n) the congested judgement parameter of data flow i in n sense cycle of expression, α and β are set point.
Particularly, n is the sequence number of sense cycle, n=0, and 1,2 ...; v iThe normal level of determining in the SLA information of signing for user and operator, i.e. corresponding flow velocity during static current limliting.
In the above-mentioned formula, K i(n-1)=0 the expression the last cycle network condition for not congested, λ i(n-1)＞ω i(n-1) input rate in last cycle of expression is greater than reference input speed.In conjunction with both of these case, when namely the input rate in last cycle was greater than reference input speed, network was not congested.Show that reference input speed is less than normal this moment, can suitably increase, and added value is α, and α is the quiescent value that system sets in advance, the rate value that expression can suitably increase.
In the above-mentioned formula, K i(n-1)＞0 the network condition in last cycle of expression is congested, and namely the reference input speed in last cycle is bigger than normal, need to reduce.Value after reducing depends on λ i(n-1), β * u i(n-1) and ω i(n-1).ω i(n-1) be the reference input speed in last cycle, ω i(n) should be less than ω i(n-1) ability alleviate congestion; λ i(n-1) be the actual input rate in last cycle, ω i(n) should be less than λ i(n-1), could control flow from entrance, avoid network congestion; u i(n-1) be the output speed in last cycle, ω i(n) should be less than u i(n-1), ability is alleviated in the territory rapidly congested; β is the quiescent value that system sets in advance, and β ∈ (0,1) passes through ω i(n)=β * u i(n-1) can realize ω i(n) less than u i(n-1).ω i(n) only have when getting above-mentioned min{ λ i(n-1), β * u i(n-1), ω i(n-1) } could alleviating network congestion.
Above-mentioned this input rate control method can be revised the input rate of a upper sense cycle according to current network condition, so that the input rate variation is comparatively level and smooth, the current network congestion situation is more pressed close in speed control.
After congested judge module calculates above-mentioned congested judgement parameter and input reference rate, these two parameters can be sent to dynamic accommodation control module in the entrance network node to admit control.
The dynamic accommodation control module is positioned on the entrance network node, according to the service inquiry SLA database of user request, obtains SLA information and congested processing mode that this user should business, judges whether to admit the user to ask in conjunction with the current network resources situation, how to admit; And the business of admitting carried out scheduling of resource.Particular content can comprise:
Can in advance the user be stored in the SLA database the form of congested feedback operation with rule.User and operator arrange in the SLA that signs: when generation is congested, to each professional processing mode, can be the static current business of demoting, dynamically demote, stop.That is, above-mentioned congested processing mode can comprise: static state is demoted, is dynamically demoted or stops current business.If the current network resources situation shows the generation network congestion, then can inquire about the SLA database and obtain SLA information and congested processing mode, suppose that SLA information comprises video traffic, then when current business is video traffic, if corresponding congested processing mode is static degradation, a static threshold of determining when then flow restriction is to the prior SLA of signing; If corresponding congested processing mode is dynamically to demote, then take the input reference rate as threshold value, dynamic speed limit; If corresponding congested processing mode then stops the link of this business data flow (such as video traffic) for stopping current business; Behind the static state degradation, the discovery flow velocity still greater than the input reference rate, then can dynamically be demoted again.In addition, if the congested processing mode of obtaining then can be understood as the user's request of having admitted for static degradation and dynamic degradation, if the congested processing mode of obtaining is for stopping current business, then can be understood as and do not admit the user to ask.Further, also can determine concrete admittance mode by congested processing mode, for example be static degradation or dynamic degradation.And, after determining the static state degradation or dynamically demoting, then can carry out respective handling according to above-mentioned static state degradation or dynamic degraded mode, to finish the business of admitting is carried out scheduling of resource.
The congested judge module of network agent provides congested judgement parameter for the dynamic accommodation control module of entrance network node, and it is congested to be used for judging whether current business occurs; Also provide the input reference rate, when carrying out dynamic degradation, take this speed as threshold value; The SLA database provides customer sla information, on this basis, carries out accordingly operation, can be the static business of demoting, dynamically demote, stop.
Fig. 7 is the method flow schematic diagram of dynamic accommodation control in the embodiment of the invention, comprising:
Step 701: reception congestion is judged parameter and input reference rate.
Step 702: judge whether to occur congested, if execution in step 703, otherwise repeated execution of steps 701.
For example, when congested judgement parameter K i(n) show that greater than 0 o'clock generation is congested, work as K i(n)=0 o'clock show do not occur congested.
Step 703: inquiry SLA information.
For example, SLA information corresponding to data query stream from the SLA database.
Step 704: need to judge whether degradation, if execution in step 705, otherwise execution in step 710.
Step 705: judge whether it is static degradation, if execution in step 706, otherwise execution in step 708.
Wherein, comprise processing mode in the SLA information, processing mode can comprise static degradation, dynamically demotes or stop current business.
Step 706: carry out the static state degradation.
Wherein, can comprise threshold value corresponding when static state is demoted in the SLA information, when static state is demoted, data flow be adjusted to this threshold value in the output speed of entrance network node.
Step 707: judge whether actual flow velocity surpasses the input reference rate, if, execution in step 708, otherwise execution in step 709.
Step 708: dynamically demote.
Be about to data flow and be adjusted into the input reference rate in the output speed of entrance network node.
Step 709: carry out resource and distribute.
Step 710: stop current business.
In conjunction with above-mentioned processing mode, can provide following application example: the P of operator and user A sign the SLA agreement, and the SLA information of user A is stored in the SLA database.During user A login, the authentication module of entry gateway C carries out authentication to user A.User A by authentication asks VOD service to the streaming media server D in the DiffServ territory.Network agent G gathers the flow rate information of entry gateway and outlet gateway, thereby calculate accumulation, congested judgement parameter and input reference rate, wherein, whether network agent G congestedly can occur congestedly to determine according to current business when judging parameter calculating.For example, as mentioned above, occur congestedly when accumulation then shows during greater than threshold value, can be set to r this moment with congested judgement parameter.Under normal circumstances, entry gateway C carries out static current limliting, flow velocity collection and charging processing to the VOD service data flow; When generation was congested, entry gateway C carries out dynamic accommodation control, flow velocity collection and charging to the VOD service data flow to be processed.
Referring to Fig. 8, Fig. 8 is the method flow schematic diagram of third embodiment of the invention, and present embodiment comprises:
Step 801: user A sends the VOD service data flow to entry gateway C.
Step 802: the static current limliting module of entry gateway C is obtained the SLA information of user A.
For example, extract the information (suppose ID number that user A registers in operator be 0001) of user A from data flow, it is 0001 user that inquiry SLA database obtains ID, through validated user and the current account of registering 200 yuan of remaining sums to be arranged, this user's SLA information is: open VOD service, the VOD service bandwidth is 2M/s, adopt dynamically degradation when being intended to congested generation together, qos parameter requires to be average delay 0.5s, average packet loss ratio 0.5%, make a call to 6 foldings at 22 in evening to 7 of next day, this user's total bandwidth is 10M/s to the maximum.
Step 803: the static current limliting module of entry gateway C is carried out static current limliting, so that the total flow of user A is 10M/s to the maximum.
Step 804: network agent G calculates the accumulation of VOD service data flow, congested judgement parameter and input reference rate.
For example, the input rate that collects entry gateway C is 2.5M/s; The accumulation that calculates is 10M, determines that current generation is congested.The congested judgement parameter that calculates is 5, and the input reference rate is reduced to 2M/s.
Step 805: entry gateway C judges parameter and input reference rate from network agent G reception congestion.
For example, congested judgement parameter is 5, and the input reference rate is 2M/s, can determine VOD service according to congested judgement parameter greater than 0 and be in congestion state.
Step 806: the dynamic accommodation control module of entry gateway C is obtained SLA information.
For example, to the SLA information of SLA data base querying user A, SLA information shows that user A has signed the clause of agreeing dynamic degradation.
Step 807: the dynamic accommodation control module of entry gateway C is dynamically demoted.
For example, recording actual flow velocity is 2.2M/s, and current input reference rate is 2M/s, then flow velocity is adjusted into 2M/s from 2.2M/s.
Step 808: the first flow velocity acquisition module of entry gateway C writes billing database with the information that gathers.
For example, the information of collection comprises flow velocity, qos parameter, and wherein, flow velocity is that 2.2M/s, qos parameter comprise: average delay 1s and average packet loss ratio 1%.Congested judgement parameter is 3, inputs reference rate 2M/s, carries out data and processes, and writes billing database;
Step 809: the accounting module of entry gateway C carries out charging.
For example, the SLA information that accounting module takes out user A from the SLA database, taking out real time rate from billing database is that 2.2M/s and time are 7200s, calculate cost of use being 5 yuan is 0.5 yuan with access fee; Calculate 5.5 yuan of real-time total costs.
In addition, can arrive 30 days and clear, make a call to 6 foldings at 22 in evening to 7 of next day, not satisfy user's SLA demand fully, 20 yuan of compensation users.
Afterwards, calculating this total cost of 30 days is 145 yuan.
In the embodiment of the invention, by take accumulation as the network congestion signal, replace traditionally take packet loss as congestion signal, only need the borderline node in monitoring territory, reduced the monitoring complexity, reduced the control signal amount, save bandwidth, can impel each node in the network to keep a metastable queue length, reduced the delay of network measure, realize simply, need the amount of information of management little.Mutually coordinate by source, entrance network node, outlet network node, network agent, interrelated, acting in conjunction reaches and jointly controls congested purpose.By adopt the dynamic accommodation control based on SLA and input reference rate at the entrance network node, the user signs SLA, determine when generation is congested, the processing method of selecting current business is static degradation, dynamically demote or stop current business, the user is write the SLA database to congested feedback operation with the form of SLA rule, both embody user's wish, reduced again the complexity of operation.By selecting the input reference rate, this input reference rate is in the current network situation, the utmost point figure of merit of input rate, and the flow rate threshold with the input reference rate during as dynamic the degradation under the prerequisite of avoid congestion, is increased network utilization to greatest extent.
Fig. 9 is the device structure schematic diagram of fourth embodiment of the invention, and this equipment can be the entrance network node, for example, and entry gateway or entrance network edge node.This equipment comprises acquisition module 91 and control module 92; Acquisition module 91 is used for obtaining congested judgement parameter corresponding to data flow and input reference rate, and described congested judgement parameter determines that according to accumulation described input reference rate is for guaranteeing the not congested extremely excellent solution of network; Control module 92 is used for determining whether to occur network congestion according to described congested judgement parameter, if network congestion occurs, according to described input reference rate and SLA information corresponding to described data flow, described data flow is carried out network congestion control.
Among the embodiment, described control module specifically is used for: need to carry out the static state degradation if SLA information corresponding to described data flow shows, then described data flow is adjusted into threshold value corresponding to described static degradation and the smaller value in the described input reference rate in the output speed of entrance network node; Perhaps, if SLA information corresponding to described data flow show and need to dynamically demote, then described data flow is adjusted into described input reference rate in the output speed of entrance network node; Perhaps, need to stop current business if SLA information corresponding to described data flow shows, then stop the link of described data flow.At this moment, control module can be specially above-mentioned dynamic accommodation control module.
Among the embodiment, described control module also is used for: if determine not occur network congestion according to described congested judgement parameter, then described data flow is adjusted into static current limit threshold in SLA information corresponding to described data flow in the output speed of entrance network node.At this moment, this control module can be specially above-mentioned static state and admit control module.
Among the embodiment, described acquisition module is used for receiving described congested judgement parameter and the input reference rate that network agent initiatively sends; Perhaps, send request to described network agent, and receive described congested judgement parameter and input reference rate that described network agent sends after receiving described request.
Present embodiment is by judging that according to congested parameter judges whether network is congested, and congested judgement parameter is determined according to accumulation, accumulation can embody the congestion situation of whole network, therefore, can reflect the congestion situation of whole network and not merely be the congestion situation of a node, thereby effectively alleviate the congestion condition of whole link.By employing SLA when network congestion is controlled, and SLA is the embodiment of user intention, therefore can reflect user intention; By adopting the input reference rate when network congestion is controlled, the extremely excellent solution of this input reference rate when guaranteeing that whole network is not congested can be increased network utilization on the not congested basis of network to greatest extent guaranteeing.
Figure 10 is the system configuration schematic diagram of fifth embodiment of the invention, comprises network agent 101 and entrance network node 102; Network agent 101 is used for obtaining the input rate of entrance network node, and the output speed of obtaining the outlet network node, determine accumulation according to described input rate and output speed, determine congested judgement parameter according to described accumulation, and determine to guarantee that the not congested extremely excellent solution of network is as the input reference rate; Entrance network node 102 is used for determining whether to occur network congestion according to described congested judgement parameter, if network congestion occurs, according to described input reference rate and SLA information corresponding to described data flow, described data flow is carried out network congestion control.
Can be that described network agent adopts following formula to determine network congestion judgement parameter:
Figure BDA0000074669160000171
Wherein, n is the sequence number of sense cycle, K i(n) the congested judgement parameter of data flow i in n sense cycle of expression, r and b are predefined value, a i(n) accumulation of data flow i in n sense cycle of expression, its computing formula is:
a i(n)＝a i(n-1)+(λ i(n)-u i(n))×τ
λ i(n) input rate of data flow i in n sense cycle of expression, u i(n) output speed of data flow i in n sense cycle of expression, τ is the data sampling time interval.
Can be that described network agent adopts following formula to determine the input reference rate:
Figure BDA0000074669160000172
Wherein, n is the sequence number of sense cycle, ω i(n) be the input reference rate of data flow i in n the sense cycle, v iBe the value corresponding with the SLA of data flow i, λ i(n) input rate of data flow i in n sense cycle of expression, u i(n) output speed of data flow i in n sense cycle of expression, K i(n) the congested judgement parameter of data flow i in n sense cycle of expression, α and β are set point.
The module of the congested judgement parameter of above-mentioned computing network and input reference rate can be specially above-mentioned congested judge module.
Described entrance network node can specifically be used for: if described congested judgement parameter greater than 0, then determines to occur network congestion, otherwise determine not occur network congestion.
Described network agent can specifically be used for: receive described input rate and output speed that described entrance network node and outlet network node initiatively send respectively; Perhaps, send request to described entrance network node and outlet network node respectively, and receive described entrance network node and export described input rate and the output speed that network node sends respectively after receiving described request.
Present embodiment is by judging that according to congested parameter judges whether network is congested, and congested judgement parameter is determined according to accumulation, accumulation can embody the congestion situation of whole network, therefore, can reflect the congestion situation of whole network and not merely be the congestion situation of a node, thereby effectively alleviate the congestion condition of whole link.By employing SLA when network congestion is controlled, and SLA is the embodiment of user intention, therefore can reflect user intention; By adopting the input reference rate when network congestion is controlled, the extremely excellent solution of this input reference rate when guaranteeing that whole network is not congested can be increased network utilization on the not congested basis of network to greatest extent guaranteeing.Be understandable that, mutually reference of the correlated characteristic in said method and the equipment, above-mentioned equipment and system can be specially equipment and the system that carries out said method.In addition, " first " in above-described embodiment, " second " etc. are for each embodiment of differentiation, and do not represent the quality of each embodiment.
The equipment that the embodiment of the invention discloses and system are for realizing method that the embodiment of the invention discloses, and the content that its concrete implementation procedure can reference method embodiment part does not repeat them here.
One of ordinary skill in the art will appreciate that: all or part of step that realizes said method embodiment can be finished by the relevant hardware of program command, aforesaid program can be stored in the computer read/write memory medium, this program is carried out the step that comprises said method embodiment when carrying out; And aforesaid storage medium comprises: the various media that can be program code stored such as ROM, RAM, magnetic disc or CD.
It should be noted that at last: above embodiment only in order to technical scheme of the present invention to be described, is not intended to limit; Although with reference to previous embodiment the present invention is had been described in detail, those of ordinary skill in the art is to be understood that: it still can be made amendment to the technical scheme that aforementioned each embodiment puts down in writing, and perhaps part technical characterictic wherein is equal to replacement; And these modifications or replacement do not make the essence of appropriate technical solution break away from the spirit and scope of various embodiments of the present invention technical scheme.

Claims (16)

1. a method for controlling network congestion is characterized in that, comprising:
The entrance network node obtains congested judgement parameter corresponding to data flow and input reference rate, and described congested judgement parameter determines that according to accumulation described input reference rate is for guaranteeing the not congested extremely excellent solution of network;
The entrance network node determines whether to occur network congestion according to described congested judgement parameter, if network congestion occurs, according to described input reference rate and SLA information corresponding to described data flow, described data flow is carried out network congestion control.
2. method according to claim 1 is characterized in that, also comprises:
Network agent obtains the input rate of described entrance network node, and the output speed of obtaining the outlet network node, determines described accumulation according to described input rate and output speed.
3. method according to claim 2 is characterized in that, the computing formula of described congested judgement parameter is:
Figure FDA0000074669150000011
Wherein, n is the sequence number of sense cycle, K i(n) the congested judgement parameter of data flow i in n sense cycle of expression, r and b are predefined value, a i(n) accumulation of data flow i in n sense cycle of expression, its computing formula is:
a i(n)＝a i(n-1)+(λ i(n)-u i(n))×τ
λ i(n) input rate of data flow i in n sense cycle of expression, u i(n) output speed of data flow i in n sense cycle of expression, τ is the data sampling time interval.
4. method according to claim 1 is characterized in that, the computing formula of described input reference rate is:
Figure FDA0000074669150000012
Wherein, n is the sequence number of sense cycle, ω i(n) be the input reference rate of data flow i in n the sense cycle, v iBe the value corresponding with the SLA of data flow i, λ i(n) input rate of data flow i in n sense cycle of expression, u i(n) output speed of data flow i in n sense cycle of expression, K i(n) the congested judgement parameter of data flow i in n sense cycle of expression, α and β are set point.
5. method according to claim 1 is characterized in that, and is described according to described input reference rate and SLA information corresponding to described data flow, and described data flow is carried out network congestion control, comprising:
The static state degradation be need to carry out if SLA information corresponding to described data flow shows, then described data flow threshold value corresponding to described static degradation and the smaller value in the described input reference rate are adjusted in the output speed of entrance network node; Perhaps,
If SLA information corresponding to described data flow shows and need to dynamically demote, then described data flow is adjusted into described input reference rate in the output speed of entrance network node; Perhaps,
Need to stop current business if SLA information corresponding to described data flow shows, then stop the link of described data flow.
6. method according to claim 3 is characterized in that, describedly determines whether to occur network congestion according to described congested judgement parameter, comprising:
If described congested judgement parameter greater than 0, then determines to occur network congestion.
7. method according to claim 2 is characterized in that, described network agent obtains the input rate of institute's entrance network node, and the output speed of obtaining the outlet network node, comprising:
Network agent receives described input rate and the output speed of described entrance network node and outlet network node difference active transmission;
Perhaps,
Network agent sends request to described entrance network node and outlet network node respectively, and receives described entrance network node and export described input rate and the output speed that network node sends respectively after receiving described request.
8. method according to claim 1 is characterized in that, described entrance network node obtains congested judgement parameter corresponding to data flow and input reference rate, comprising:
Described entrance network node receives described congested judgement parameter and the input reference rate that network agent initiatively sends;
Perhaps,
Described entrance network node sends request to described network agent, and receives described congested judgement parameter and input reference rate that described network agent sends after receiving described request.
9. a network congestion control appliance is characterized in that, comprising:
Acquisition module is used for obtaining congested judgement parameter corresponding to data flow and input reference rate, and described congested judgement parameter determines that according to accumulation described input reference rate is for guaranteeing the not congested extremely excellent solution of network;
Control module is used for determining whether to occur network congestion according to described congested judgement parameter, if network congestion occurs, according to described input reference rate and SLA information corresponding to described data flow, described data flow is carried out network congestion control.
10. equipment according to claim 9 is characterized in that, described control module specifically is used for:
The static state degradation be need to carry out if SLA information corresponding to described data flow shows, then described data flow threshold value corresponding to described static degradation and the smaller value in the described input reference rate are adjusted in the output speed of entrance network node; Perhaps,
If SLA information corresponding to described data flow shows and need to dynamically demote, then described data flow is adjusted into described input reference rate in the output speed of entrance network node; Perhaps,
Need to stop current business if SLA information corresponding to described data flow shows, then stop the link of described data flow.
11. equipment according to claim 9 is characterized in that, described acquisition module specifically is used for:
Receive described congested judgement parameter and input reference rate that network agent initiatively sends; Perhaps,
Send request to described network agent, and receive described congested judgement parameter and input reference rate that described network agent sends after receiving described request.
12. a network congestion control system is characterized in that, comprising:
Network agent, be used for obtaining the input rate of entrance network node, and the output speed of obtaining the outlet network node, determine accumulation according to described input rate and output speed, determine congested judgement parameter according to described accumulation, and determine to guarantee that the not congested extremely excellent solution of network is as the input reference rate;
The entrance network node is used for determining whether to occur network congestion according to described congested judgement parameter, if network congestion occurs, according to described input reference rate and SLA information corresponding to described data flow, described data flow is carried out network congestion control.
13. system according to claim 12 is characterized in that, described network agent adopts following formula to determine network congestion judgement parameter:
Figure FDA0000074669150000041
Wherein, n is the sequence number of sense cycle, K i(n) the congested judgement parameter of data flow i in n sense cycle of expression, r and b are predefined value, a i(n) accumulation of data flow i in n sense cycle of expression, its computing formula is:
a i(n)＝a i(n-1)+(λ i(n)-u i(n))×τ
λ i(n) input rate of data flow i in n sense cycle of expression, u i(n) output speed of data flow i in n sense cycle of expression, τ is the data sampling time interval.
14. system according to claim 12 is characterized in that, described network agent adopts following formula to determine the input reference rate:
Figure FDA0000074669150000042
Wherein, n is the sequence number of sense cycle, ω i(n) be the input reference rate of data flow i in n the sense cycle, v iBe the value corresponding with the SLA of data flow i, λ i(n) input rate of data flow i in n sense cycle of expression, u i(n) output speed of data flow i in n sense cycle of expression, K i(n) the congested judgement parameter of data flow i in n sense cycle of expression, α and β are set point.
15. system according to claim 13 is characterized in that, described entrance network node specifically is used for: if described congested judgement parameter greater than 0, then determines to occur network congestion.
16. system according to claim 12 is characterized in that, described network agent specifically is used for:
Receive described input rate and output speed that described entrance network node and outlet network node initiatively send respectively;
Perhaps,
Send request to described entrance network node and outlet network node respectively, and receive described entrance network node and export described input rate and the output speed that network node sends respectively after receiving described request.
CN201110191495.7A 2011-07-08 2011-07-08 Method for controlling network congestion, equipment and system Expired - Fee Related CN102868671B (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201110191495.7A CN102868671B (en) 2011-07-08 Method for controlling network congestion, equipment and system
PCT/CN2012/078375 WO2013007180A1 (en) 2011-07-08 2012-07-09 Method, device and system for network congestion control

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201110191495.7A CN102868671B (en) 2011-07-08 Method for controlling network congestion, equipment and system

Publications (2)

Publication Number Publication Date
CN102868671A true CN102868671A (en) 2013-01-09
CN102868671B CN102868671B (en) 2016-11-30

Family

ID=

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106102094A (en) * 2016-07-06 2016-11-09 广州海格通信集团股份有限公司 A kind of TCP jamming control method based on SDN
CN106789704A (en) * 2017-01-13 2017-05-31 北京交通大学 The precise quantification method of network-on-chip current limiting factor
CN107637032A (en) * 2015-03-03 2018-01-26 欧庞戈网络有限公司 System and method for adjusting data flow
CN108064058A (en) * 2016-11-07 2018-05-22 中兴通讯股份有限公司 Jamming control method and device, base station
CN108093027A (en) * 2017-11-22 2018-05-29 余仪琳 A kind of traffic congestion based on Internet of Things solves method
WO2019029318A1 (en) * 2017-08-11 2019-02-14 华为技术有限公司 Network congestion notification method, proxy node and computer device
CN110955529A (en) * 2020-02-13 2020-04-03 北京一流科技有限公司 Memory resource static deployment system and method
CN111886875A (en) * 2018-03-27 2020-11-03 英国电讯有限公司 Congestion response for timely media delivery

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2000028706A1 (en) * 1998-11-10 2000-05-18 Odyssia Systems, Inc. Method and apparatus to minimize congestion in a packet switched network
CN101166140A (en) * 2006-10-18 2008-04-23 中国科学院自动化研究所 A network congestion control system and method for Internet

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2000028706A1 (en) * 1998-11-10 2000-05-18 Odyssia Systems, Inc. Method and apparatus to minimize congestion in a packet switched network
CN101166140A (en) * 2006-10-18 2008-04-23 中国科学院自动化研究所 A network congestion control system and method for Internet

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10834002B2 (en) 2015-03-03 2020-11-10 Opanga Networks, Inc. Systems and methods for pacing data flows
CN107637032A (en) * 2015-03-03 2018-01-26 欧庞戈网络有限公司 System and method for adjusting data flow
US11546268B2 (en) 2015-03-03 2023-01-03 Opanga Networks, Inc. Systems and methods for pacing data flows
CN106102094A (en) * 2016-07-06 2016-11-09 广州海格通信集团股份有限公司 A kind of TCP jamming control method based on SDN
CN106102094B (en) * 2016-07-06 2019-09-06 广州海格通信集团股份有限公司 A kind of TCP jamming control method based on SDN network
US11751096B2 (en) 2016-11-07 2023-09-05 Zte Corporation Congestion control method and device, and base station
CN108064058A (en) * 2016-11-07 2018-05-22 中兴通讯股份有限公司 Jamming control method and device, base station
CN108064058B (en) * 2016-11-07 2022-11-01 中兴通讯股份有限公司 Congestion control method and device and base station
CN106789704B (en) * 2017-01-13 2019-06-14 北京交通大学 The precise quantification method of network-on-chip current limiting factor
CN106789704A (en) * 2017-01-13 2017-05-31 北京交通大学 The precise quantification method of network-on-chip current limiting factor
US11374870B2 (en) 2017-08-11 2022-06-28 Huawei Technologies Co., Ltd. Network congestion notification method, agent node, and computer device
WO2019029318A1 (en) * 2017-08-11 2019-02-14 华为技术有限公司 Network congestion notification method, proxy node and computer device
CN108093027A (en) * 2017-11-22 2018-05-29 余仪琳 A kind of traffic congestion based on Internet of Things solves method
CN111886875A (en) * 2018-03-27 2020-11-03 英国电讯有限公司 Congestion response for timely media delivery
CN110955529A (en) * 2020-02-13 2020-04-03 北京一流科技有限公司 Memory resource static deployment system and method

Also Published As

Publication number Publication date
WO2013007180A1 (en) 2013-01-17

Similar Documents

Publication Publication Date Title
US10091675B2 (en) System and method for estimating an effective bandwidth
CN112954385B (en) Self-adaptive shunt decision method based on control theory and data driving
CN102611605B (en) Scheduling method, device and system of data exchange network
US20080013449A1 (en) Mmpp analysis of network traffic using a transition window
CN101674242B (en) Service message sending control method and device
CN109831522A (en) A kind of vehicle connection cloud and mist system dynamic resource Optimal Management System and method based on SMDP
Kbah et al. Resource allocation in cellular systems for applications with random parameters
CN112217725B (en) Delay optimization method based on edge calculation
CN114039918B (en) Information age optimization method and device, computer equipment and storage medium
US11411865B2 (en) Network resource scheduling method, apparatus, electronic device and storage medium
CN110784399A (en) Multi-path optimization combined transmission method and system for MPTCP (Multi-path Transmission control protocol) network
CN107483355B (en) Data center-oriented online scene low-bandwidth overhead traffic scheduling scheme
Zhang et al. A dynamic resource overbooking mechanism in fog computing
CN109688065B (en) Parameter processing method and device and storage medium
Lenzini et al. Delay bounds for FIFO aggregates: a case study
Liu et al. Perturbation analysis for stochastic fluid queueing systems
CN108513318B (en) User service queuing optimization method based on edge calculation
CN108616377A (en) Business chain virtual machine control method and system
CN113747450A (en) Service deployment method and device in mobile network and electronic equipment
Sun et al. Bandwidth estimation for aggregate traffic under delay QoS constraint based on supermartingale theory
CN110677352A (en) Single-switch single-controller transmission control method in software defined network
CN110012507A (en) A kind of car networking resource allocation methods that user experience is preferential and system
CN102739508A (en) Method and system for data transmission of high-speed network
CN114024913B (en) Network performance optimization method, device, equipment and storage medium
US20220232584A1 (en) Method and apparatus for bandwidth allocation

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20161130

Termination date: 20180708

CF01 Termination of patent right due to non-payment of annual fee