CN102271092A - Congestion control system and method based on node load quantization level - Google Patents
Congestion control system and method based on node load quantization level Download PDFInfo
- Publication number
- CN102271092A CN102271092A CN2011102678491A CN201110267849A CN102271092A CN 102271092 A CN102271092 A CN 102271092A CN 2011102678491 A CN2011102678491 A CN 2011102678491A CN 201110267849 A CN201110267849 A CN 201110267849A CN 102271092 A CN102271092 A CN 102271092A
- Authority
- CN
- China
- Prior art keywords
- node
- packet
- congestion
- congestion cause
- network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a congestion control system and method based on a node load quantization level, and the system and method provided by the invention are mainly used for solving the defect that the network throughput improved by virtue of the conventional congestion control method is smaller when a network is under a high load. The system comprises a data processing unit, a node load state unit and a control measure unit; the data processing unit calculates a queue length mark and a data packet processing number mark and submits two types of mark information to the node load state unit; the node load state unit determines a congestion reason quantity mark according to the queue length mark and the data packet processing number mark, and submits the mark information to the control measure unit; and the control measure unit judges whether to abandon a data packet or not according to the congestion reason quantity mark, sets regulation parameters for processing capability and an effective path, and feeds packet abandoning information to the data processing unit. According to the invention, under the high load in the network, on the premise that a time delay requirement is met, the network throughput is increased as much as possible, and the system and method provided by the invention can be used for an internet and an ATM (asynchronous transfer mode) network.
Description
Technical field
The invention belongs to data communication technology field, relate to Route Selection and jamming control method in the packet switching network, be applicable to internet and atm network etc.
Background technology
Along with the rise and the development of Complex Networks Theory, studies show that in a large number large-scale communication network such as the Internet and World Wide Web (WWW) etc. all has worldlet effect and no characteristics of scale, be typical complex network.Network congestion is a very general phenomenon.In case it is congested that network takes place, will cause the message transmission time significantly to increase and overall performance of network sharply descends, how about Control Network disposal ability congested and that improve network packet is the problem that people are concerned about very much.
People are that instrument has carried out research and obtained some being familiar with substantially to congested correlation properties with the Complex Networks Theory.Congested and improve for the network transmission efficiency for control, many researchs mainly concentrate on two aspects: revise infrastructure network or develop better routing policy.Because the modification cost of network infrastructure is higher, thus the latter more people pay close attention to.
Routing strategy is mainly used to select transmission path for the packet of being received, though routing policy does not cause congested generation from essence, the quality of routing policy design but has appreciable impact to network performance.Particularly when network enters congestion state, good routing policy can utilize local congestion's information of network, by being distributed to, packet stops congested aggravation on other link or the node, thereby improve the performance of network, the routing policy of difference then can further aggravate network congestion, even causes periods of network disruption.
General routing policy such as BFS, random walk etc. focus on the transmission of finishing packet and promptly divide the group selection path, do not consider the situation such as congested of network.In addition, the raising of network node disposal ability generally is subject to two factors: the buffer memory capacity of node and disposal ability.Yet many method supposition nodes in the past have unlimited capacity or identical disposal ability, and this system with reality does not conform to.
Based on above shortcoming, on the basis of the routing policy of shortest path first, people such as Yan G have proposed " active path (efficient path) ", it in essence degree of being to use calculate shortest path as the weight of node.In order to strengthen the ability to communicate of network, the disposal ability of proposition node i such as Liang Zhao is C
i=C+int[β k
i], 0＜β＜1 is a Control Parameter, the disposal ability that improves the maincenter node will increase the throughput of whole network.These congestion avoidance algorithms substantially all are to start with from producing certain congested single reason, do not consider that it may be which kind of reason causes network congestion and takes to improve accordingly measure at different reasons that node is analyzed the network and the understanding of s own situation, improve less in the prerequisite lower network throughput that satisfies delay requirement when making the high load capacity situation.
Summary of the invention
The objective of the invention is to overcome the defective of above-mentioned prior art, a kind of congestion control system and control method of adjusting the router behavior based on node state is provided, with descending in advance at the nodal cache finite capacity, can guarantee the time delay minimum when making network load low, and satisfy increase network throughput big as far as possible under the prerequisite of delay requirement during the high load capacity situation.
For achieving the above object, control system of the present invention comprises:
Data processing unit, be used to extract the present node length of buffer queue, calculate that queue length indicates and number sign that node at the appointed time can handle packet, and submit queue length beacon information and processing data packets number beacon information to the node state unit;
The node load state cell is used for determining the node load quantification gradation according to the queue length beacon information, and statistics congestion cause quantity indicates, and submits this congestion cause quantity beacon information to the control measure unit;
The control measure unit is used for indicating according to node congestion cause quantity, and whether the judgment data bag abandons, and determines the number of Data transmission bag, selects the path of required time delay minimum for the energy data packets for transmission.
Described data processing unit comprises: queue length indicates module and processing data packets number module, this queue length indicates module and indicates according to queue length computing node queue length, and the queue length beacon information is submitted to the node load state cell, this processing data packets number module, the maximum allowable delay computing node that can also in network, exist according to packet can handle packet number indicate, and processing data packets number beacon information is submitted to the node load state cell.
Described node load state cell comprises: node load quantification gradation module and congestion cause quantity module, this node load quantification gradation module, queue length is indicated and the contrast of processing data packets number sign, determine the node load quantification gradation, this congestion cause quantity module, cause congested reason and add up the quantity of reason according to the analysis of node load quantification gradation, generate congestion cause quantity beacon information.
Described control measure unit comprises: the active queue management module, disposal ability selects module and routing policy to select module, this active queue management module according to node congestion cause quantity indicate abandon in the buffer memory part consumption of network resources minimal data bag and initiatively the packet loss feedback information give a data processing unit and a last hop node, this disposal ability selects module to indicate the number of calculating transfer data packets according to node congestion cause quantity, determine those packets to transmit, this routing policy selects module to be denoted as the path that the packet that can transmit is selected required time delay minimum according to node congestion cause quantity.
For achieving the above object, control method of the present invention comprises the steps:
(1) at the initial time of each time slot t, from the buffer memory of node i, reads queue length
With the packet sum that receives in the last time slot
Be calculated as follows queue length and indicate L
i
(2) maximum allowable delay of packet: τ among the computing node i
i=τ
Max-(H
i+ N
i) * T, wherein N
iBe the shortest jumping figure of last packet in the buffer queue from node i arrival destination node, H
iBe the node jumping figure of this packet from source node arrival node i process, T is the normalization time of single-hop in the data packet delivery process, τ
MaxThe maximum delay that allows for network is the life cycle of packet;
(3) computing node i is at τ
iThe maximum number M of energy handle packet in time
MaxWith minimum number M
Min: according to node processing ability C
i=C+int (β k
i) calculate
Wherein, C is a constant, and β is the adjusting parameter of transmittability, k
iDegree for node i;
(4) with L
iRespectively with M
Max, M
MinCompare, determine the node load quantification gradation: if m
Max〉=L
i＞M
Max, then the node load quantification gradation is A; If M
Min＜L
i≤ M
Max, then the node load quantification gradation is B; If 0＜L
i≤ M
Min, then the node load quantification gradation is C, wherein,
Be the buffer memory capacity of all nodes, N is the network node sum;
(5) indicate M according to node load quantification gradation statistics congestion cause quantity: if the node load quantification gradation is A, then congestion cause quantity is denoted as M=3, if the node load quantification gradation is B, then node congestion cause quantity is denoted as M=2, if the node load quantification gradation is C, then node congestion cause quantity is denoted as M=3, otherwise the quantity of congestion cause is denoted as M=0;
(6) indicate the adjustment nodes ' behavior according to node congestion cause quantity:
If 6a) node congestion cause quantity indicates M=0, then the node processing ability is selected basic value, and routing policy is selected shortest path, i.e. set handling capacity adjustment parameter beta=0, and source node s is through the active path of n via node arrival destination node d
Adjusting parameter alpha=0;
If 6b) node congestion cause quantity indicates M=1, then the node processing ability is selected basic value, and routing policy is selected active path, i.e. set handling capacity adjustment parameter beta=0, and active path is regulated parameter alpha＞0;
If 6c) node congestion cause quantity indicates M=2, then the node processing ability is selected higher value, and routing policy is selected active path, i.e. set handling capacity adjustment parameter 0＜β＜1, and active path is regulated parameter alpha＞0;
If 6d) node congestion cause quantity indicates M=3, then read the shortest jumping figure n of all packet b to its destination node
bArrive the node that this node skipped with it and count h
b, calculate
Press η
bDescending order is to the data packet number, abandons numbering in front
Individual packet also is changed to 2 with M, changes 6c);
(7) according to the value computing node disposal ability C of β and α
i=C+int (β k
i), and be data packets for transmission to select
The path that value is minimum.
The present invention is because according to the node load quantification gradation, analysis causes the reason of network congestion and determines that congestion cause quantity indicates, be indicated in the network light hours according to this congestion cause quantity, the node routing policy is selected shortest path, disposal ability is selected basic value, thereby guaranteed service time delay minimum; Simultaneously when the middle and high load of network, node initiatively abandons a part minimum packet of consumption of network resources, routing policy is selected the shortest active path of time delay, and disposal ability is selected higher value, thus can be under the prerequisite of assurance delay requirement big as far as possible increase network throughput.
Description of drawings
Fig. 1 is a congestion control system structured flowchart of the present invention;
Fig. 2 is the data processing unit structured flowchart among the present invention;
Fig. 3 is the node load state cell structured flowchart among the present invention;
Fig. 4 is the control measure cellular construction block diagram among the present invention;
Fig. 5 is the general flow chart of jamming control method of the present invention;
Fig. 6 is provided with the sub-process figure that regulates the parameter adjustment nodes ' behavior in the inventive method;
Fig. 7 is the sub-process figure of active packet loss strategy in the inventive method;
Fig. 8 is the analogous diagram that active path is regulated parameter best value in the inventive method;
Fig. 9 is the analogous diagram that disposal ability is regulated parameter best value in the inventive method.
Embodiment
Below in conjunction with accompanying drawing content of the present invention is described in detail:
With reference to Fig. 1, congestion control system structure of the present invention comprises: data processing unit 1, node load state cell 2 and control measure unit 3.Wherein, undertaken alternately by transmitting beacon information between data processing unit 1, node load state cell 2 and the control measure unit 3, beacon information comprises 3 kinds of queue length beacon information, processing data packets number beacon information and congestion cause quantity beacon information.
Described data processing unit 1, every time slot extracts the queue length of present node buffer memory the zero hour, calculates that queue length indicates and number sign that node at the appointed time can handle packet.The concrete structure figure of this unit as shown in Figure 2, it comprises that queue length indicates module 11 and processing data packets number module 12.Wherein, queue length indicates module 11, indicates according to length of buffer queue computing node queue length, and this queue length beacon information is submitted to node load state cell 2; Processing data packets number module 12, the maximum allowable delay computing node that can also in network, exist according to packet can handle packet number indicate, this processing data packets number beacon information is submitted to node load state cell 2.
Described node load state cell 2, indicate and processing data packets number sign according to length of buffer queue, determine the node load quantification gradation, analysis causes the reason of network congestion, statistics congestion cause quantity, generate congestion cause quantity and indicate, and submit this congestion cause quantity beacon information to control measure unit 3.The concrete structure figure of this unit as shown in Figure 3, it comprises node load quantification gradation module 21 and congestion cause quantity module 22, wherein, node load quantification gradation module 21, determine the node load quantification gradation according to queue length beacon information and processing data packets number information, the node load quantification gradation is passed to congestion cause quantity module 22; Congestion cause quantity module 22 causes congested multiple reason and adds up the quantity of reason according to the quantification gradation analysis of node load, generates congestion cause quantity and indicates, and this congestion cause quantity beacon information is submitted to control measure unit 3.
Described control measure unit 3, indicating the judgment data bag according to congestion cause quantity should abandon, transmits or wait for, the packet information that abandons is fed back to a data processing unit 1 and a last hop node, and select the path of required time delay minimum for the packet that can transmit.The concrete structure figure of this unit as shown in Figure 4, it comprises that active queue management module 31, disposal ability select module 32 and routing policy to select module 33.Wherein, active queue management module 31, indicate according to congestion cause quantity and to abandon in the buffer memory part consumption of network resources minimal data bag, indicate to upgrade congestion cause quantity, and the congestion cause quantity beacon information after will upgrading is submitted to disposal ability selection module 32 and routing policy is selected module 33, and initiatively the packet loss feedback information is given a data processing unit 1 and a last hop node simultaneously; Disposal ability is selected module 32, indicates set handling capacity adjustment parameter according to congestion cause quantity, the computing node disposal ability, and press the still wait of regular determination data bag transmission of serving earlier first; Routing policy is selected module 33, then indicates according to congestion cause quantity active path adjusting parameter is set, and selects the transmission path of required time delay minimum for the packet that can transmit.
With reference to Fig. 5, the present invention is used for the jamming control method that packet is transmitted between network node, comprise the steps:
Step 1:, from the buffer memory of node router or switch i, read the packet sum that receives in the last time slot by data processing unit at the initial time of each time slot t
The packet number that initiatively abandons with node i
Upgrade queue length
Be calculated as follows node queue's length indication information:
Step 2: according to the close characteristic of any two node beelines of complex network, the information of getting last packet in the buffer queue, the maximum allowable delay that all packets can also exist in network among the computing node i: τ
i=τ
Max-(H
i+ N
i) * T, wherein, N
iBe the shortest path jumping figure of last packet in the buffer queue from node i arrival destination node, H
iBe the node jumping figure of this packet from source node arrival node i process, T is the normalization time of single-hop in the data packet delivery process, τ
MaxThe maximum delay that allows for network is the life cycle of packet.
Step 3: the maximum allowable delay τ that can also in network, exist according to packet
iWith node processing ability C
i=C+int (β k
i) computing node i is at τ
iThe maximum number M of energy handle packet in time
MaxWith minimum number M
Min:
Wherein, C is a constant, k
iBe the degree of node i, β is the adjusting parameter of disposal ability, and there is the optimum value that makes network throughput maximum in β, can contrast the pairing network throughput of different adjustment parameter beta by emulation and determine.
Step 4: queue length is indicated L
iRespectively with the maximum M of processing data packets number
MaxAnd minimum M
MinCompare, determine the node load quantification gradation:
If m
Max〉=L
i＞M
Max, then the node load quantification gradation is A; If M
Min＜Li≤M
Max, then the node load quantification gradation is B; If 0＜L
i≤ M
Min, then the node load quantification gradation is C, wherein,
Be the buffer memory capacity of all nodes, N is the network node sum.
Step 5: according to the node load quantification gradation, analysis causes congested multiple reason, determines that the congestion cause quantity of node indicates M:
If the node load quantification gradation is A, think that then node congestion cause quantity indicates M=3, promptly cause the main cause of network congestion to have three kinds: the one, the bad packet queuing delay that causes of routing policy increases, the 2nd, by the too little packet queuing delay increase that causes of disposal ability of node, the 3rd, be dropped by the excessive packet that causes exceeding buffer memory capacity of the load of fan-in network;
If the node load quantification gradation is B, think that then node congestion cause quantity indicates M=2, promptly cause congested main cause to have two kinds: the one, the bad packet queuing propagation delay time that causes of routing policy increases, and the 2nd, the too little packet queuing delay that causes of the disposal ability of node increases;
If the node load quantification gradation is C, think that then node congestion cause quantity indicates M=1, promptly congestion cause is that the bad packet queuing delay that causes of routing policy increases;
Otherwise, think that node congestion cause quantity indicates M=0, promptly node does not produce the congested free state that is in.
Step 6: indicate set handling capacity adjustment parameter and active path adjusting parameter according to congestion cause quantity, adjust nodes ' behavior:
With reference to Fig. 6, the concrete steps that this step is adjusted nodes ' behavior are as follows:
Step 6.1: if congestion cause quantity indicates M=0, then the node processing ability is selected basic value, and routing policy is selected shortest path, i.e. set handling capacity adjustment parameter beta=0, active path
Regulate parameter alpha=0, wherein,
Expression source node s is through the active path length of n via node arrival destination node d, and α=0 an o'clock active path is equivalent to shortest path;
Step 6.2: if congestion cause quantity indicates M=1, then the node processing ability is selected basic value, routing policy is selected active path, it is set handling capacity adjustment parameter beta=0, active path is regulated parameter alpha＞0, wherein, there is the optimum value that makes network throughput maximum in α, can contrast the pairing network throughput of different adjustment parameter alpha by emulation and determine;
Step 6.3: if congestion cause quantity indicates M=2, then the node processing ability is selected higher value, routing policy is selected active path, it is set handling capacity adjustment parameter 0＜β＜1, active path is regulated parameter alpha＞0, wherein, β and α are respectively the optimum value of the above step 3 and step 6.3;
Step 6.4:, press initiatively packet loss strategy packet discard if congestion cause quantity indicates M=3:
With reference to Fig. 7, the concrete steps of this step packet loss are as follows:
Step 6.4a: from nodal cache, read the shortest path jumping figure n of all packets to its destination node
bArrive the node that this node skipped with it and count h
b, calculating, consumption of network resources indicates
Step 6.4b: all packets in the buffer memory are numbered: if the η of these packets
bBe worth unequally, then press η
bBe worth descending order to the data packet number, if the η of these packets
bEquate, then work as η
bPressed n at 〉=1 o'clock
bDescending order is worked as η to the data packet number
bPressed n at＜1 o'clock
bAscending order is to the data packet number;
Step 6.4c: from buffer memory, abandon numbering the preceding
Individual packet, and packet loss information is submitted to a data processing unit and a last hop node;
Step 6.4d: upgrading M is 2, changes step 6.3.
Step 7: according to the value computing node disposal ability C of β and α
i=C+int (β k
i), by the C of the rule selection buffer queue front of serving earlier first
iIndividual data packet transmission is the C that can transmit
iIndividual packet is selected
The transmission path that value is minimum is transmitted packet and renewal
Effect of the present invention can illustrate by following emulation:
1, simulation parameter setting
Emulation is pressed BA network model construction algorithm and is generated network, gets m
0=3, m=3, network node sum N=100, the basic transmittability C=2 of node/time slot, τ life cycle of packet
Max=32 time slots.
2, emulation content
When the best value of parameter alpha was regulated in emulation 1, emulation active path, β was taken as certain value β=0.05, calculates C
Max=max{C
i}=4/time slot, nodal cache capacity m
Max=128, difference statistics network arrival rate R=43, the network throughput of 45,47,49/time slot, simulation result such as Fig. 8.
When the best value of parameter beta was regulated in emulation 2, emulation active path, α was definite value α=0.5, statistics network arrival rate R=43 respectively, the network throughput of 45,47,49,51/time slot, simulation result such as Fig. 9.
3, simulation result and analysis:
From Fig. 8 as seen, α is that the network throughput of 0.5 o'clock different R all reaches maximum, α＞0.5 o'clock R=47, and 49 network throughput reduces, and therefore thinks that the optimum value of α should be 0.5.
From Fig. 9 as seen, it is maximum that β is that the network throughput of 0.1 o'clock different R all reaches, and the network throughput of β＞0.1 o'clock different R is basic identical with β=0.1 o'clock all, therefore thinks that the optimum value of β should be 0.1.
Claims (9)
1. congestion control system based on the node load quantification gradation comprises:
Data processing unit (1), be used to extract the present node length of buffer queue, calculate that queue length indicates and number sign that node at the appointed time can handle packet, and submit queue length beacon information and processing data packets number beacon information to node state unit (2);
Node load state cell (2) is used for determining the node load quantification gradation according to the queue length beacon information, and statistics congestion cause quantity indicates, and submits this congestion cause quantity beacon information to control measure unit (3);
Control measure unit (3) is used for indicating according to node congestion cause quantity, and whether the judgment data bag abandons, and determines the number of Data transmission bag, selects the path of required time delay minimum for the energy data packets for transmission.
2. control system according to claim 1, wherein data processing unit (1) comprising: queue length indicates module (11) and processing data packets number module (12), this queue length indicates module (11) and indicates according to queue length computing node queue length, and the queue length beacon information is submitted to node load state cell (2), this processing data packets number module (12), the maximum allowable delay computing node that can also in network, exist according to packet can handle packet number indicate, and processing data packets number beacon information is submitted to node load state cell (2).
3. control system according to claim 1, wherein node load state cell (2) comprising: node load quantification gradation module (21) and congestion cause quantity module (22), this node load quantification gradation module (21), queue length is indicated and the contrast of processing data packets number sign, determine the node load quantification gradation, this congestion cause quantity module (22), cause congested reason and add up the quantity of reason according to the analysis of node load quantification gradation, generate congestion cause quantity beacon information.
4. control system according to claim 1, wherein control measure unit (3) comprising: active queue management module (31), disposal ability selects module (32) and routing policy to select module (33), this active queue management module (31) according to congestion cause quantity indicate abandon in the buffer memory part consumption of network resources minimal data bag and initiatively the packet loss feedback information give a data processing unit (1) and a last hop node, this disposal ability selects module (32) to indicate the number of calculating transfer data packets according to congestion cause quantity, determine those packets to transmit, this routing policy selects module (33) to be denoted as the path that packet that this time slot can transmit is selected required time delay minimum according to congestion cause quantity.
5. the jamming control method based on the node load quantification gradation comprises the steps:
(1) at the initial time of each time slot t, from the buffer memory of node i, reads queue length
With the packet sum that receives in the last time slot
Be calculated as follows queue length and indicate L
i
(2) maximum allowable delay of packet: τ among the computing node i
i=τ
Max-(H
i+ N
i) * T, wherein N
iBe the shortest jumping figure of last packet in the buffer queue from node i arrival destination node, H
iBe the node jumping figure of this packet from source node arrival node i process, T is the normalization time of single-hop in the data packet delivery process, τ
MaxThe maximum delay that allows for network is the life cycle of packet;
(3) computing node i is at τ
iThe maximum number M of energy handle packet in time
MaxWith minimum number M
Min: according to node processing ability C
i=C+int (β k
i) calculate
Wherein, C is a constant, and β is the adjusting parameter of disposal ability, k
iDegree for node i;
(4) with L
iRespectively with M
Max, M
MinCompare, determine the node load quantification gradation: if m
Max〉=L
i＞M
Max, then the node load quantification gradation is A; If m
Min＜L
i≤ M
Max, then the node load quantification gradation is B; If 0＜L
i≤ M
Min, then the node load quantification gradation is C, wherein,
Be the buffer memory capacity of all nodes, N is the network node sum;
(5) indicate M according to node load quantification gradation statistics congestion cause quantity: if the node load quantification gradation is A, then congestion cause quantity is denoted as M=3, if the node load quantification gradation is B, then node congestion cause quantity is denoted as M=2, if the node load quantification gradation is C, then node congestion cause quantity is denoted as M=3, otherwise congestion cause quantity is denoted as M=0;
(6) indicate the adjustment nodes ' behavior according to node congestion cause quantity:
If 6a) node congestion cause quantity indicates M=0, then the node processing ability is selected basic value, and routing policy is selected shortest path, i.e. set handling capacity adjustment parameter beta=0, and source node s is through the active path of n via node arrival destination node d
Adjusting parameter alpha=0;
If 6b) node congestion cause quantity indicates M=1, then the node processing ability is selected basic value, and routing policy is selected active path, i.e. set handling capacity adjustment parameter beta=0, and active path is regulated parameter alpha＞0;
If 6c) node congestion cause quantity indicates M=2, then the node processing ability is selected higher value, and routing policy is selected active path, i.e. set handling capacity adjustment parameter 0＜β＜1, and active path is regulated parameter alpha＞0;
If 6d) node congestion cause quantity indicates M=3, then read the shortest jumping figure n of all packets to its destination node
bArrive the node that this node skipped with it and count h
b, calculate
Press η
bDescending order is to the data packet number, abandons numbering in front
Individual packet also is changed to 2 with M, changes 6c);
(7) according to the value computing node disposal ability C of β and α
i=C+int (β k
i), and be data packets for transmission to select
The path that value is minimum.
6. control method according to claim 5, congestion cause quantity in the wherein said step (5) indicates M=3, expression causes the main cause of network congestion to have three kinds: the one, increase by the data packet transmission time delay that routing policy caused, the 2nd, wait for that by the too little packet that causes of disposal ability of node time delay increases, the 3rd, be dropped by the excessive packet that causes exceeding buffer memory capacity of the load of fan-in network.
7. control method according to claim 5, congestion cause quantity in the wherein said step (5) indicates M=2, expression causes the main cause of network congestion to have two kinds: the one, increase by the data packet transmission time delay that routing policy caused, and the 2nd, the too little packet that causes of the disposal ability of node waits for that time delay increases.
8. control method according to claim 5, the congestion cause quantity in the wherein said step (5) indicates M=1, and the expression congestion cause is that the data packet transmission time delay that routing policy causes increases.
9. control method according to claim 5, the congestion cause quantity in the wherein said step (5) indicates M=0, and the expression node does not produce the congested free state that is in.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110267849.1A CN102271092B (en) | 2011-09-09 | 2011-09-09 | Congestion control system and method based on node load quantization level |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110267849.1A CN102271092B (en) | 2011-09-09 | 2011-09-09 | Congestion control system and method based on node load quantization level |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102271092A true CN102271092A (en) | 2011-12-07 |
| CN102271092B CN102271092B (en) | 2014-03-12 |
Family
ID=45053259
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110267849.1A Active CN102271092B (en) | 2011-09-09 | 2011-09-09 | Congestion control system and method based on node load quantization level |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102271092B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102821001A (en) * | 2012-09-17 | 2012-12-12 | 吉林大学 | Method for realizing fuzzy neuron active queue management method in IPCOP |
| CN104092623A (en) * | 2013-04-01 | 2014-10-08 | 株式会社日立制作所 | Method and device used for carrying out overload protection on network node or server |
| CN106713035A (en) * | 2016-12-23 | 2017-05-24 | 西安电子科技大学 | Method for positioning congestion link based on group testing |
| CN108345621A (en) * | 2017-01-24 | 2018-07-31 | 腾讯科技（深圳）有限公司 | A kind of data account checking method, system and data system |
| CN113099488A (en) * | 2019-12-23 | 2021-07-09 | ***通信集团陕西有限公司 | Method, device, computing equipment and computer storage medium for solving network congestion |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1964310A (en) * | 2005-11-09 | 2007-05-16 | 中兴通讯股份有限公司 | Packet discarding system and method of IP network |
| CN101286930A (en) * | 2008-05-30 | 2008-10-15 | 华南理工大学 | Self-adapting routing method for congestion of multi-hop wireless self-organizing network |
| US20100027425A1 (en) * | 2008-07-30 | 2010-02-04 | Fimax Technology Limited | Fair weighted network congestion avoidance |
| CN101656674A (en) * | 2009-09-23 | 2010-02-24 | 中国人民解放军信息工程大学 | Congestion control method and network nodes |
-
2011
- 2011-09-09 CN CN201110267849.1A patent/CN102271092B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1964310A (en) * | 2005-11-09 | 2007-05-16 | 中兴通讯股份有限公司 | Packet discarding system and method of IP network |
| CN101286930A (en) * | 2008-05-30 | 2008-10-15 | 华南理工大学 | Self-adapting routing method for congestion of multi-hop wireless self-organizing network |
| US20100027425A1 (en) * | 2008-07-30 | 2010-02-04 | Fimax Technology Limited | Fair weighted network congestion avoidance |
| CN101656674A (en) * | 2009-09-23 | 2010-02-24 | 中国人民解放军信息工程大学 | Congestion control method and network nodes |
Non-Patent Citations (5)
| Title |
|---|
| GANG YAN 等: "Efficient routing on complex networks", 《AMERICAN PHYSICAL SOCIETY》, vol. 73, no. 4, 7 April 2006 (2006-04-07) * |
| LIANG ZHAO 等: "Onset of traffic congestion in complex networks", 《AMERICAN PHYSICAL SOCIETY》, vol. 71, no. 2, 24 February 2005 (2005-02-24) * |
| 刘宁 等: "拥塞控制相关研究", 《中国科技论文在线》, 12 August 2010 (2010-08-12) * |
| 吴春明 等: "几种主动式队列管理算法的比较研究", 《电子学报》, vol. 32, no. 3, 31 March 2004 (2004-03-31) * |
| 江昊 等: "动态权重RED网关", 《电子与信息学报》, vol. 26, no. 7, 31 July 2004 (2004-07-31) * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102821001A (en) * | 2012-09-17 | 2012-12-12 | 吉林大学 | Method for realizing fuzzy neuron active queue management method in IPCOP |
| CN104092623A (en) * | 2013-04-01 | 2014-10-08 | 株式会社日立制作所 | Method and device used for carrying out overload protection on network node or server |
| CN106713035A (en) * | 2016-12-23 | 2017-05-24 | 西安电子科技大学 | Method for positioning congestion link based on group testing |
| CN108345621A (en) * | 2017-01-24 | 2018-07-31 | 腾讯科技（深圳）有限公司 | A kind of data account checking method, system and data system |
| CN108345621B (en) * | 2017-01-24 | 2022-03-08 | 腾讯科技（深圳）有限公司 | Data reconciliation method, system and data system |
| CN113099488A (en) * | 2019-12-23 | 2021-07-09 | ***通信集团陕西有限公司 | Method, device, computing equipment and computer storage medium for solving network congestion |
| CN113099488B (en) * | 2019-12-23 | 2024-04-09 | ***通信集团陕西有限公司 | Method, device, computing equipment and computer storage medium for solving network congestion |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102271092B (en) | 2014-03-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102075444B (en) | Network system and method for guaranteeing multi-type service quality | |
| CN105471764B (en) | A kind of method of end-to-end QoS guarantee in SDN network | |
| CN107579922B (en) | Network load balancing device and method | |
| EP3537656B1 (en) | Network control method, apparatus and system, storage medium | |
| US10129043B2 (en) | Apparatus and method for network flow scheduling | |
| CN103329492B (en) | The system and method for property performance period early drop in buffer storage on the chip of network element | |
| CN105827540B (en) | A kind of data center network transport layer data stream transmission method priority-based | |
| CN102415059B (en) | Bus control device | |
| CN107370676A (en) | Fusion QoS and load balancing demand a kind of route selection method | |
| CN102271092B (en) | Congestion control system and method based on node load quantization level | |
| CN106411733A (en) | SDN dynamic load balanced scheduling method based on real-time load of link | |
| CN108768876B (en) | Traffic scheduling method facing machine learning framework | |
| CN102664784B (en) | A kind of mapping method of virtual network of node link pressure weight adaptive equalization | |
| CN102196502A (en) | Congestion control method for wireless sensor network | |
| Lai et al. | CYBER-PHYSICAL LOGISTICS SYSTEM-BASED VEHICLE ROUTING OPTIMIZATION. | |
| CN110708736A (en) | Dynamic routing method and system based on energy efficiency relay selection | |
| US20160036689A1 (en) | Method and apparatus for reducing response time in information-centric networks | |
| CN105848238A (en) | IPv6 routing method of wireless sensor networks based on multiple parameters | |
| CN103297341A (en) | Method for configuring flow on nodes of internal router | |
| CN102427596B (en) | Routing method and scheduling method of node mobile network assisted by positioning information | |
| CN105610707A (en) | Implementation method of AntNet routing algorithm in two-dimensional mesh topology network-on-chip | |
| CN110460538A (en) | Data transmission method in a kind of Internet of Things of software definition | |
| CN109484244A (en) | Battery management system and electric car | |
| CN107682235A (en) | A kind of propagation delay time prediction data dispatching method based on Kalman filtering | |
| CN102917399B (en) | A kind of time-delay measuring method of wireless sensor network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |