CN106921584A - A kind of distributed network flow control method - Google Patents
A kind of distributed network flow control method Download PDFInfo
- Publication number
- CN106921584A CN106921584A CN201710204408.4A CN201710204408A CN106921584A CN 106921584 A CN106921584 A CN 106921584A CN 201710204408 A CN201710204408 A CN 201710204408A CN 106921584 A CN106921584 A CN 106921584A
- Authority
- CN
- China
- Prior art keywords
- flow control
- node
- flow
- information
- control node
- 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.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/20—Traffic policing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/16—Threshold monitoring
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/29—Flow control; Congestion control using a combination of thresholds
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention relates to a kind of distributed network flow control method, wherein, distributed network includes management node and connected multiple flow control nodes, and its methods described includes：Flow control node carries out flow information collection to respective channel, after report cycle arrival, the information of each respective channel is sent to management node by each flow control node, and management node is received after the information of the respective channel of each flow control node transmission, carries out statistical analysis；After management node statistical analysis, the new flow threshold whether judge respective channel has the passage is produced, and returns to flow control node.By the present invention, the flow control threshold value of each egress network is dynamically adjusted, go to intervene without artificial.
Description
Technical field
Present invention relates particularly to a kind of distributed network flow control method.
Background technology
With pc, mobile phone, the explosive growth of each class of electronic devices such as flat board, the network size of internet also expands rapidly
Greatly, the behavior and network application of user are also more and more diversified.Many non-key service applications are robbed and account for ours on network
Bandwidth so that many normal key businesses and application are affected.When user stays at home normally lookes at webpage, but go out existing network
During the situation of the network qualities such as network interim card reduction, the experience of user can have a greatly reduced quality.In order to lift Consumer's Experience, Logistics networks thing
Manage the unimpeded and regular traffic of passage.Then there has been flow control.Stream control scene conventional at present has the stream for single user
Amount control, the scene is mainly on user's couple in router, or is controlled on BRAS/BAS.But if will be to whole
Individual network carries out flow control by agreement, then generally require carries out the cloth of fluidic device in the exit of Metropolitan Area Network (MAN) or province's net
Put.
In order to avoid disaster, the influence to servicing such as failure, the access of Metropolitan Area Network (MAN) to the backbone network of operator can use many
Individual outlet, or even can be accessed in different computer rooms, flow typically carries out load balancing by stream between each access point.The opposing party
Face, as it currently stands, Metropolitan Area Network (MAN) or save net to backbone network access bandwidth can to hundreds of or even thousands of G, but also
In continuing to increase.
Using the mode of traditional single node stream control, on the one hand physically it is difficult to accomplish all of flow all to converge to
In one equipment, it is also difficult on the other hand to carry out stream control according to agreement and user using the flow of single device Gs thousands of to hundreds of
With what is accomplished.
Conventional way is to be divided equally global threshold according to number of devices at present, but the distribution of existing network flow is general
It is uneven, and often fluctuates, on the whole, carry out stream control error using the method larger, or even normal industry can be influenceed
Business.
But, the state that distributed dynamic adjustment threshold value of the invention can be in real time according to network adjusts each stream control node
Threshold value, so adaptability to network is stronger, and the utilization rate to bandwidth is also higher.
The content of the invention
First, the concept of passage, passage are explained：Can be one or more protocol types or application type；
Can be source IP section or purpose IP section, or the two combine；Can also be certain link；Others can also be extended, his table
What is shown is session characteristics in network.This method is applied to need to carry out flow to some or multiple passages in certain network egress
Control, also, when having multiple outlets and being required for carrying out same or multiple passages flow control, can be using we
Method.
The present invention proposes a kind of distributed network flow control method to solve above mentioned problem of the prior art, its
In, distributed network includes management node and connected multiple flow control nodes, and methods described includes：
Step 101, flow control node carries out flow information collection to respective channel, and it includes that protocal analysis, passage are known
Not, flow information statistics, wherein, flow information includes byte number, bag number, fluxion；
Step 102, flow control node judges whether respective channel has setting flow threshold, and judges that flow now is
No if exceeding threshold value, then to carry out flow control more than threshold value, the flow control includes postponing forwarding or packet loss；Otherwise, turn
Enter step 103；
Step 103, after report cycle arrival, be sent to for the information of each respective channel by each flow control node
Management node, the information of transmission includes present flow rate, the flow of request, flows newly-built, number of dropped packets, passage ID and other networks letter
Breath；
Step 104, management node is received after the information of the respective channel of each flow control node transmission, is counted
Analysis；Wherein, management node is also configured as receiving the outside flow threshold for issuing or setting certain passage；
Step 105, after management node statistical analysis, the new flow threshold whether judge respective channel has the passage is produced,
If no, returning to flow control node of the default traffic threshold value to the passage；If so, being transferred to step 106；
Step 106, management node further determines whether it is the information for receiving the passage for the first time, if it is not, returning nearest
The flow threshold of the statistical computation passage out gives flow control node；If it is, return to default traffic threshold value giving
Flow control node.
Preferably, methods described also includes：The channel information that management node sends each flow control node, temporarily deposits
In internal memory, after the channel information of all flow control nodes of equivalent a cycle is reached, COMPREHENSIVE CALCULATING goes out newest each
The new flow threshold that flow control node respective channel should be configured, when flow control node reports channel information next time,
Beam back new flow threshold and give flow control node；Wherein, the channel information that each flow control node is reported is with passage as single
Position, the channel information of each passage is sent to management node by it after a cycle arrival.
Preferably, methods described also includes：Management node can also simultaneously as extension flow control node, now, pipe
The upper strata for managing node is connected with root node, when flow control node is issued the management section as extended flow control node by channel information
During point, channel information is then forwarded to root node by management node, and root node is received after channel information, carries out statistical analysis meter
Calculate；When root node receives channel information next time, this COMPREHENSIVE CALCULATING result is returned into management node, management node will be calculated
The new flow threshold for obtaining shares each flow again as total threshold value according to the network state information of each flow control node
Control node, and be transmitted across.
By the invention it is possible to the real network situation according to each network egress, dynamically adjusts each egress network
Flow control threshold value, go to intervene without artificial.
Brief description of the drawings
Fig. 1 is 2 layers of flow control structure figure of the invention.
Fig. 2 is 3 layers of flow control structure figure of the invention.
Fig. 3 is inventive flow control flow chart.
Specific embodiment
The present invention is described in further details below in conjunction with the accompanying drawings.
It is shown in Figure 1：It is 2 the most frequently used Rotating fields, is made up of a management node and several stream control nodes.
It is shown in Figure 2：It is the 3-tier architecture after extension, there is a root node (management node) and several middle nodes
Point (reporting functions of management node+control node) and several control nodes are constituted.
It is shown in Figure 3：It is flow control flow chart.Specially：
Step 101, network control node carries out flow information collection, including protocal analysis, channel recognition, to flow information
(byte number, bag number, fluxion) is counted.
Step 102, judges whether this passage has a setting threshold value, and whether flow (up-downgoing speed) now exceedes threshold
Value, if exceeding threshold value, then carries out stream control, including postpone forwarding or packet loss
Step 103, after report cycle arrival, flow control node being sent to one by one by the information of each passage
Management node, the data of transmission include but is not limited to present flow rate, and the flow of request flows newly-built, number of dropped packets, passage ID etc. various
The network information.
Step 104, management node is received after the information of certain passage that stream control node sends, and carries out statistical analysis.Pipe
Reason node is possible to receive the outside threshold value for issuing or setting certain passage.
Step 105, determines whether the threshold value of the passage, if it's not true, returns to default threshold to control node.
Step 106, judges whether it is the data for receiving the passage for the first time, if it is not, returning to the last time
The threshold value of the statistical computation passage out.
Step 107, returns to default threshold to control node.
Wherein, management node will flow the channel status information that control node sends, and be temporarily present in internal memory, when same week
After the identical channel data of all nodes of phase is reached, COMPREHENSIVE CALCULATING goes out newest each respective channel of each stream control node should
The new threshold value of the configuration, in stream control node reported data next time, sends back to stream control node.What each node here was reported
Data are in units of passage, after a cycle arrival, the status information of each passage to be sent into management node.
Wherein, Fig. 2 is the extension of Fig. 1, and this method also supports that extension, i.e. management node can also simultaneously as stream control section
There is root node on point, the upper strata of management node, and at this moment, when data are issued management node by stream control node, management node is by number
According to root node is then forwarded to, when root node receives next secondary data, this COMPREHENSIVE CALCULATING result is returned into management node, managed
Node shares each stream control again using the threshold value of certain passage as total threshold value according to the network state information of each stream control node
Node sends the past.
Exception, in the stream control scene of multiple exit, synthesis is carried out by the information that others stream controls node, does calculating analysis,
Again new stream control threshold value is fed back, this dynamic adjustment is done.And it is anti-for reporting for this multi-client one service end of correspondence
Feedback pattern, is also extended including this pattern, develops into tree-like, netted structure.And the letter of this real-time father and son's node
Message communication feedback model can be applied equally in other fields, such as water consumption, power consumption, the wagon flow of traffic control, people
Stream etc..
Embodiment described above is only that the preferred embodiment of the present invention is described, not to model of the invention
Enclose and be defined, on the premise of design spirit of the present invention is not departed from, this area ordinary skill technical staff is to skill of the invention
Various modifications and improvement that art scheme is made, all should fall into the protection domain of claims of the present invention determination.
Claims (3)
1. a kind of distributed network flow control method, wherein, distributed network includes management node and connected many
Individual flow control node, it is characterised in that methods described includes：
Step 101, flow control node carries out flow information collection to respective channel, and it includes protocal analysis, channel recognition, stream
Amount Information Statistics, wherein, flow information includes byte number, bag number, fluxion；
Step 102, flow control node judges whether respective channel has setting flow threshold, and judges whether flow now surpasses
Threshold value is crossed, if exceeding threshold value, then flow control is carried out, the flow control includes postponing forwarding or packet loss；Otherwise, it is transferred to step
Rapid 103；
Step 103, after report cycle arrival, the information of each respective channel is sent to management by each flow control node
Node, the information of transmission includes present flow rate, the flow of request, flows newly-built, number of dropped packets, passage ID and other network informations；
Step 104, management node is received after the information of the respective channel of each flow control node transmission, carries out statistical
Analysis；Wherein, management node is also configured as receiving the outside flow threshold for issuing or setting certain passage；
Step 105, after management node statistical analysis, the new flow threshold whether judge respective channel has the passage is produced, if not having
Have, return to flow control node of the default traffic threshold value to the passage；If so, being transferred to step 106；
Step 106, management node further determines whether it is the information for receiving the passage for the first time, if it is not, returning the last
The flow threshold of the statistical computation passage out gives flow control node；If it is, returning to default traffic threshold value to flow
Control node.
2. method according to claim 1, it is characterised in that methods described also includes：Management node is by each flow control
The channel information that node processed sends, is temporarily present in internal memory, the channel information of all flow control nodes of equivalent a cycle
After arrival, COMPREHENSIVE CALCULATING goes out the new flow threshold that each newest flow control node respective channel should be configured, next time
When flow control node reports channel information, beam back new flow threshold and give flow control node；Wherein, each flow control node
The channel information for reporting is that in units of passage, be sent to for the channel information of each passage after a cycle arrival by it
Management node.
3. method according to claim 1, it is characterised in that methods described also includes：Management node can also be made simultaneously
It is extension flow control node, now, the upper strata of management node is connected with root node, when flow control node sends out channel information
During to management node as extended flow control node, channel information is then forwarded to root node by management node, and root node receives logical
After road information, statistical analysis calculating is carried out；When root node receives channel information next time, this COMPREHENSIVE CALCULATING result is returned
To management node, the new flow threshold that management node will be calculated is as total threshold value according to the network of each flow control node
Status information shares each flow control node again, and is transmitted across.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710204408.4A CN106921584A (en) | 2017-03-31 | 2017-03-31 | A kind of distributed network flow control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710204408.4A CN106921584A (en) | 2017-03-31 | 2017-03-31 | A kind of distributed network flow control method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106921584A true CN106921584A (en) | 2017-07-04 |
Family
ID=59461992
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710204408.4A Pending CN106921584A (en) | 2017-03-31 | 2017-03-31 | A kind of distributed network flow control method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106921584A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107948084A (en) * | 2017-12-15 | 2018-04-20 | 中国银联股份有限公司 | A kind of current-limiting method and device |
| CN108235804A (en) * | 2017-12-27 | 2018-06-29 | 深圳前海达闼云端智能科技有限公司 | A kind of network speed limit method, device and server |
| CN108683598A (en) * | 2018-04-20 | 2018-10-19 | 武汉绿色网络信息服务有限责任公司 | A kind of asymmetrical network flow processing method and processing unit |
| CN109150740A (en) * | 2018-08-01 | 2019-01-04 | 武汉绿色网络信息服务有限责任公司 | A distributed network flow control method and device |
| CN109787906A (en) * | 2017-11-13 | 2019-05-21 | 瑞昱半导体股份有限公司 | Control of network flow quantity device and system |
| CN110460676A (en) * | 2019-08-22 | 2019-11-15 | 腾讯科技（深圳）有限公司 | Request quantity control method and device, storage medium and computer equipment |
| CN110858808A (en) * | 2018-08-23 | 2020-03-03 | 阿里巴巴集团控股有限公司 | Method, system and device for realizing speed limit |
| CN111262793A (en) * | 2020-01-20 | 2020-06-09 | 福建奇点时空数字科技有限公司 | System flow overall control method based on dual flow control mechanism |
| CN111585798A (en) * | 2020-04-28 | 2020-08-25 | 腾讯科技（深圳）有限公司 | A kind of network resource parameter configuration method, apparatus and computer readable storage medium |
| CN113328906A (en) * | 2021-04-22 | 2021-08-31 | 成都欧珀通信科技有限公司 | Flow real-time monitoring method and device, storage medium and electronic equipment |
| CN115348208A (en) * | 2021-04-27 | 2022-11-15 | 中移(苏州)软件技术有限公司 | A flow control method, device, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101166308A (en) * | 2007-08-07 | 2008-04-23 | 中兴通讯股份有限公司 | A method for dynamically controlling information traffic from local device to device at other end |
| CN101286937A (en) * | 2008-05-16 | 2008-10-15 | 华为技术有限公司 | A network traffic control method, device and system |
| CN102035737A (en) * | 2010-12-08 | 2011-04-27 | 北京交通大学 | Adaptive load balancing method and device based on cognitive network |
| CN105991474A (en) * | 2015-05-18 | 2016-10-05 | 杭州迪普科技有限公司 | Traffic scheduling method, traffic scheduling device and network device |
-
2017
- 2017-03-31 CN CN201710204408.4A patent/CN106921584A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101166308A (en) * | 2007-08-07 | 2008-04-23 | 中兴通讯股份有限公司 | A method for dynamically controlling information traffic from local device to device at other end |
| CN101286937A (en) * | 2008-05-16 | 2008-10-15 | 华为技术有限公司 | A network traffic control method, device and system |
| CN102035737A (en) * | 2010-12-08 | 2011-04-27 | 北京交通大学 | Adaptive load balancing method and device based on cognitive network |
| CN105991474A (en) * | 2015-05-18 | 2016-10-05 | 杭州迪普科技有限公司 | Traffic scheduling method, traffic scheduling device and network device |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109787906A (en) * | 2017-11-13 | 2019-05-21 | 瑞昱半导体股份有限公司 | Control of network flow quantity device and system |
| CN107948084A (en) * | 2017-12-15 | 2018-04-20 | 中国银联股份有限公司 | A kind of current-limiting method and device |
| CN107948084B (en) * | 2017-12-15 | 2021-07-13 | 中国银联股份有限公司 | A current limiting method and device |
| CN108235804A (en) * | 2017-12-27 | 2018-06-29 | 深圳前海达闼云端智能科技有限公司 | A kind of network speed limit method, device and server |
| CN108235804B (en) * | 2017-12-27 | 2021-12-31 | 达闼机器人有限公司 | Network speed limiting method and device and server |
| CN108683598B (en) * | 2018-04-20 | 2020-04-10 | 武汉绿色网络信息服务有限责任公司 | Asymmetric network traffic processing method and processing device |
| CN108683598A (en) * | 2018-04-20 | 2018-10-19 | 武汉绿色网络信息服务有限责任公司 | A kind of asymmetrical network flow processing method and processing unit |
| CN110224944A (en) * | 2018-08-01 | 2019-09-10 | 武汉绿色网络信息服务有限责任公司 | A kind of distributed network flow control method and device |
| CN109150740A (en) * | 2018-08-01 | 2019-01-04 | 武汉绿色网络信息服务有限责任公司 | A distributed network flow control method and device |
| CN109150740B (en) * | 2018-08-01 | 2019-09-10 | 武汉绿色网络信息服务有限责任公司 | A distributed network flow control method and device |
| CN110224944B (en) * | 2018-08-01 | 2021-10-08 | 武汉绿色网络信息服务有限责任公司 | A distributed network flow control method and device |
| CN110858808A (en) * | 2018-08-23 | 2020-03-03 | 阿里巴巴集团控股有限公司 | Method, system and device for realizing speed limit |
| CN110460676A (en) * | 2019-08-22 | 2019-11-15 | 腾讯科技（深圳）有限公司 | Request quantity control method and device, storage medium and computer equipment |
| CN110493362A (en) * | 2019-08-22 | 2019-11-22 | 腾讯科技（深圳）有限公司 | Request amount control method, device, storage medium and computer equipment |
| CN110493362B (en) * | 2019-08-22 | 2022-04-29 | 腾讯科技（深圳）有限公司 | Request quantity control method and device, storage medium and computer equipment |
| CN110460676B (en) * | 2019-08-22 | 2022-03-25 | 腾讯科技（深圳）有限公司 | Request quantity control method and device, storage medium and computer equipment |
| CN111262793A (en) * | 2020-01-20 | 2020-06-09 | 福建奇点时空数字科技有限公司 | System flow overall control method based on dual flow control mechanism |
| CN111585798A (en) * | 2020-04-28 | 2020-08-25 | 腾讯科技（深圳）有限公司 | A kind of network resource parameter configuration method, apparatus and computer readable storage medium |
| CN111585798B (en) * | 2020-04-28 | 2024-03-15 | 腾讯科技（深圳）有限公司 | Network resource parameter configuration method, device and computer readable storage medium |
| CN113328906A (en) * | 2021-04-22 | 2021-08-31 | 成都欧珀通信科技有限公司 | Flow real-time monitoring method and device, storage medium and electronic equipment |
| CN113328906B (en) * | 2021-04-22 | 2023-01-06 | 成都欧珀通信科技有限公司 | Flow real-time monitoring method and device, storage medium and electronic equipment |
| CN115348208A (en) * | 2021-04-27 | 2022-11-15 | 中移(苏州)软件技术有限公司 | A flow control method, device, electronic equipment and storage medium |
| CN115348208B (en) * | 2021-04-27 | 2024-04-09 | 中移(苏州)软件技术有限公司 | Flow control method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106921584A (en) | A kind of distributed network flow control method | |
| Bouillet et al. | The structure and management of service level agreements in networks | |
| US10554570B2 (en) | Method and device for implementing QoS in OpenFlow network | |
| CN105556916B (en) | Method and device for information statistics of network flow | |
| CN110572280B (en) | A network monitoring method and system | |
| CN104580018B (en) | Bandwidth adjusting method and device in a kind of software defined network | |
| CN100592707C (en) | Method and apparatus for determining token bucket parameters for guaranteed services in a data network | |
| CN104468408B (en) | For dynamically adjusting the method and control centre's server of service bandwidth | |
| CN110191065A (en) | High performance load balancing system and method based on software defined network | |
| CN103532869A (en) | Data flow sharing method and equipment | |
| CN107919982A (en) | A kind of DCI management platforms and its management method | |
| CN109150740A (en) | A distributed network flow control method and device | |
| CN115277574B (en) | Data center network load balancing method under SDN architecture | |
| WO2021244247A1 (en) | Data message forwarding method, network node, system, and storage medium | |
| CN109787910A (en) | A traffic management method, system and fabric network processor | |
| CN105991456B (en) | A kind of OpenFlow interchanger, network system and bandwidth sharing method | |
| CN110769023A (en) | A point-to-point content distribution network system based on intelligent home gateway | |
| CN112583636A (en) | Construction method of government affair network slice, electronic device and storage medium | |
| CN104539558A (en) | Capacity-expansible IP telephone exchange blade mechanism frame and automatic capacity expansion method | |
| CN114448899A (en) | A method for balancing network load in data center | |
| CN109309624A (en) | Traffic scheduling method and system, software defined network controller | |
| CN107248959A (en) | A kind of flow optimization method and device | |
| CN106341339A (en) | End-to-end network flow control method | |
| CN108199975A (en) | A kind of flow control methods and device | |
| CN101022358A (en) | Method for realizing resource cluster and management control in IP telecommunication network system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170704 |