CN101977158A - Method for realizing congestion control in multi-source application layer multicast - Google Patents
Method for realizing congestion control in multi-source application layer multicast Download PDFInfo
- Publication number
- CN101977158A CN101977158A CN2010105582674A CN201010558267A CN101977158A CN 101977158 A CN101977158 A CN 101977158A CN 2010105582674 A CN2010105582674 A CN 2010105582674A CN 201010558267 A CN201010558267 A CN 201010558267A CN 101977158 A CN101977158 A CN 101977158A
- Authority
- CN
- China
- Prior art keywords
- data
- node
- application layer
- layer multicast
- output buffer
- 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
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention relates to a method for realizing congestion control in multi-source application layer multicast. The method comprises the following steps that: 1, an application layer multicast node reserves an input buffer region and an output buffer region for each neighboring node; 2, the application layer multicast node locally records one piece of congestion control information for each subnode of each multicast data flow; 3, the application layer multicast node locally and periodically calculates the congestion feedback information of each data flow and transmits the information to a corresponding parent node; 4, when needing to transmit data to the subnode, the node first calculates the maximum allowable transfer data quantity and writes all data to be transmitted into the output buffer region if the data quantity to be transmitted is less than the maximum allowable transfer data quantity; and 5, a transmission process reads data from the output buffer region in real time and transmits the data to the subnode. By the method, the buffer region is shared by different multicast data flows, the utilization ratio of buffer can be improved, and the congestion control and reliable transmission are realized.
Description
Technical field
The present invention relates to a kind of method for controlling network congestion, be specifically related to a kind of jamming control method that is applied to the multi-source application layer multicast, belong to application layer multicast reliable transmission technical field.
Background technology
The multi-source application layer multicast be a kind of multiple spot to the multiple spot data transfer mode, wherein any one group membership all may have oneself as data source multicast distribution tree data are multicast to every other node in the group; And any group membership also will receive the multicast packet that send in all multicast packet sources.Because this multicast is to realize on application layer, so group membership's node is also being undertaken the task that multicast packet duplicates and transmits.Make up the multi-source application layer multicast, at first need on the basis of existing physical network, set up nerve of a covering, on nerve of a covering, make up multicast tree then for each data source.If have nerve of a covering logical links between A and the B at 2, just may there be two-way data path between 2, thereby A is upstream (father) node of B in certain forwarding tree, A then is downstream (son) node of B on the another one forwarding tree, so only at the multicast tree of determining, the upstream and downstream of description node relation is just meaningful.Local node can be input block of each neighbours' link establishment and an output buffer at intra-node, and different multicast traffic streams can be shared the buffering area that inputs or outputs of this link simultaneously when sharing the same logical links of local node.Because multicast traffic stream is characterised in that the many outputs of single input, all may there be the data flow bifurcated in each intermediate node, and on nerve of a covering, have the formed a plurality of tree-like multicast traffic stream mutual superposition of a plurality of data sources simultaneously, and can make that data flow is intricate, data volume explodes.
The transmission means of multicast makes multicast packet mode with multiple in network increase, the duplicated more at most packet of group membership is many more, and traditional IP multicast does not provide rate controlled, so flux of multicast is easy to occupy the all-network resource, and then causes congested.Application layer multicast more is easy to generate congested, is because the processing speed of application layer node is slow, may pile up mass data in the buffering area, if do not controlled the transmission rate of source node this moment, will cause buffering area to overflow, and then produces packet loss; In addition, a plurality of processes on the application layer node need be competed the unstable properties that host resource causes end host, can't realize that the linearity of packet is transmitted, and also can cause congested.Each node in the multi-source application layer multicast all is potential data source, is easy to the situation that a plurality of data sources are asked the forwarded service simultaneously takes place, if scheduling is not controlled and distinguished to each source node, then more is easy to generate congested.
Summary of the invention
The purpose of this invention is to provide a kind of multi-source application layer multicast jamming control method, it can make the application layer multicast source node carry out effective rate controlled according to its buffer state, guarantees effective reliable transmission of data.Effectively adjust according to the congestion feedback information of each data flow in addition and share the buffering distribution, can guarantee the fairness between the multiple data stream.
The invention provides and a kind ofly in the multi-source application layer multicast, realize congestion control method, may further comprise the steps:
Step 1, application layer multicast node are set up TCP in application layer for each neighbor node and are connected, and reserve input block and output buffer; After node receives the data of father node, if the input block less than, just in real time data are deposited into the input block; Treatment progress is from reading of data wherein, and removes data processed bag;
Step 2, application layer multicast node are congestion control information of each child node record of every multicast traffic stream in this locality, this value is initially set to the maximum that this record can be represented, after receiving the congestion control information of the data flow j that a child node is sent, promptly upgrade local record;
Step 3, application layer multicast node regularly calculate the congestion feedback information of each data flow in this locality, and this congestion feedback information is sent to the corresponding father node of each data flow; The concrete grammar that calculates congestion feedback information is: the application layer multicast node is selected the congestion feedback information P of minimum value as data flow j from all congestion control information of the data flow j of this locality storage
j, and with P
jCompare with the idle size of all output buffers of data flow j, getting minimum value is new congestion feedback information P
j
Step 4, when node will be when child node sends data, at first calculate the maximum transferring data amount that allows, allow the transferring data amount if need to send data volume less than maximum, then all need be sent data and write output buffer; Otherwise, only shift the maximum data volume that allows transfer to output buffer;
Calculating the maximum method of transferring data amount that allows is:
(1) the application layer multicast node is according to formula W
j=α * C
j+ β * P
jThe transfer weight of calculated data stream j, wherein C
jBe the data volume of existing data flow j in the current buffering area, α, β is a weight factor, gets the real number between 0 to 1;
(2) the application layer multicast node is shared the promptly maximum transferring data amount that allows of free space of buffering according to shifting weight for each distribution of flows:
S wherein
jThe maximum transferring data amount that allows of expression, φ represents to share with data flow j the every other set of data flows of same output buffer, B represents the idle size of corresponding output buffer;
Step 5, the transmission process is from the real-time reading of data of output buffer and send to child node.
Beneficial effect
The reliability that this method utilizes TCP to connect guarantees that data do not send and can lose between node; Utilize simultaneously local buffer guarantee to connect the transfer of data of interrupting causing loses can be by fast quick-recovery; And at intra-node reliably from the input block to the output buffer transferring data, therefore can guarantee that the transmission course of whole multicast data flow loss of data can not take place.Make different multicast traffic stream shared buffers can improve the buffering utilance, and limit the flow of every multicast traffic stream, reach the purpose of congested control by the distributing buffer space.Therefore, use this method can reach reliable transmission, strengthen the purpose of buffering utilance.
Description of drawings
Fig. 1 is the flow chart of the method for the invention;
Embodiment
Below in conjunction with accompanying drawing, specify preferred implementation of the present invention.
Present embodiment has realized a kind ofly realizing congestion control method in the multi-source application layer multicast.
At first some terms that present embodiment is used make an explanation:
The input block---refer to the memory headroom that multicast node is reserved for the reception packet in application layer.Node is reading of data from TCP, and the data of reading are kept in the input block, and treatment progress is from reading of data wherein, and removes data processed bag;
Output buffer---refer to that multicast node is at the memory headroom of application layer for reserving to child node forwarding packet, treatment progress is transferred to output buffer with data from the input block, the transmission process is given the TCP layer and is handled from the real-time reading of data of output buffer.
Congestion control information---refer to the congestion control parameter safeguarded for each data flow on node, this parameter is calculated by child node and is generated and transmit to father node.
Shift weight---refer in the standard of local node, promptly shift the free space of the shared buffering of pro rate of weight according to each data flow for the free space institute foundation of the shared buffering of each distribution of flows;
Weight factor---the weighted value when referring to the different parameters weighted sum, big more this parameter of expression of this value is also big more to the influence of final result.
Fig. 1 is the flow chart of present embodiment institute using method.As shown in the figure, present embodiment comprises the steps:
Step 1, application layer multicast node are set up TCP in application layer for each neighbor node and are connected, and reserve input block and output buffer; After node receives the data of father node, if the input block less than, just in real time data are deposited into the input block; Treatment progress is from reading of data wherein, and removes data processed bag;
Step 2, application layer multicast node are congestion control information of each child node record of every multicast traffic stream in this locality, this value is initially set to the maximum that this record can be represented, after receiving the congestion control information of the data flow j that a child node is sent, promptly upgrade local record;
Step 3, application layer multicast node regularly calculate the congestion feedback information of each data flow in this locality, and this congestion feedback information is sent to the corresponding father node of each data flow; The concrete grammar that calculates congestion feedback information is: the application layer multicast node is selected the congestion feedback information P of minimum value as data flow j from all congestion control information of the data flow j of this locality storage
j, and with P
jCompare with the idle size of all output buffers of data flow j, getting minimum value is new congestion feedback information P
j
Step 4, when node will be when child node sends data, at first calculate the maximum transferring data amount that allows, allow the transferring data amount if need to send data volume less than maximum, then all need be sent data and write output buffer; Otherwise, only shift the maximum data volume that allows transfer to output buffer;
Calculating the maximum method of transferring data amount that allows is:
(1) the application layer multicast node is according to formula W
j=α * C
j+ β * P
jThe transfer weight of calculated data stream j, wherein C
jBe the data volume of existing data flow j in the current buffering area, α, β is a weight factor, gets the real number between 0 to 1;
(2) the application layer multicast node is shared the promptly maximum transferring data amount that allows of free space of buffering according to shifting weight for each distribution of flows:
S wherein
jThe maximum transferring data amount that allows of expression, φ represents to share with data flow j the every other set of data flows of same output buffer, B represents the idle size of corresponding output buffer;
Step 5, the transmission process is from the real-time reading of data of output buffer and send to child node.
It should be noted last that, above embodiment is only unrestricted in order to technical scheme of the present invention to be described, although the present invention is had been described in detail with reference to preferred embodiment, those of ordinary skill in the art is to be understood that, can make amendment or be equal to replacement technical scheme of the present invention, and not break away from the spirit and scope of technical solution of the present invention.
Claims (1)
1. in the multi-source application layer multicast, realize congestion control method for one kind, it is characterized in that may further comprise the steps:
Step 1, application layer multicast node are set up TCP in application layer for each neighbor node and are connected, and reserve input block and output buffer; After node receives the data of father node, if the input block less than, just in real time data are deposited into the input block; Treatment progress is from reading of data wherein, and removes data processed bag;
Step 2, application layer multicast node are congestion control information of each child node record of every multicast traffic stream in this locality, this value is initially set to the maximum that this record can be represented, after receiving the congestion control information of the data flow j that a child node is sent, promptly upgrade local record;
Step 3, application layer multicast node regularly calculate the congestion feedback information of each data flow in this locality, and this congestion feedback information is sent to the corresponding father node of each data flow; The concrete grammar that calculates congestion feedback information is: the application layer multicast node is selected the congestion feedback information P of minimum value as data flow j from all congestion control information of the data flow j of this locality storage
j, and with P
jCompare with the idle size of all output buffers of data flow j, getting minimum value is new congestion feedback information P
j
Step 4, when node will be when child node sends data, at first calculate the maximum transferring data amount that allows, allow the transferring data amount if need to send data volume less than maximum, then all need be sent data and write output buffer; Otherwise, only shift the maximum data volume that allows transfer to output buffer;
Calculating the maximum method of transferring data amount that allows is:
(1) the application layer multicast node is according to formula W
j=α * C
j+ β * P
jThe transfer weight of calculated data stream j, wherein C
jBe the data volume of existing data flow j in the current buffering area, α, β is a weight factor, gets the real number between 0 to 1;
(2) the application layer multicast node is shared the promptly maximum transferring data amount that allows of free space of buffering according to shifting weight for each distribution of flows:
S wherein
jThe maximum transferring data amount that allows of expression, φ represents to share with data flow j the every other set of data flows of same output buffer, B represents the idle size of corresponding output buffer;
Step 5, the transmission process is from the real-time reading of data of output buffer and send to child node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010105582674A CN101977158A (en) | 2010-11-22 | 2010-11-22 | Method for realizing congestion control in multi-source application layer multicast |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010105582674A CN101977158A (en) | 2010-11-22 | 2010-11-22 | Method for realizing congestion control in multi-source application layer multicast |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101977158A true CN101977158A (en) | 2011-02-16 |
Family
ID=43577005
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010105582674A Pending CN101977158A (en) | 2010-11-22 | 2010-11-22 | Method for realizing congestion control in multi-source application layer multicast |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101977158A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104507119A (en) * | 2014-11-26 | 2015-04-08 | 无锡儒安科技有限公司 | Concurrent transmission based wireless sensor network data collection method |
| CN109388609A (en) * | 2018-09-30 | 2019-02-26 | 中科驭数（北京）科技有限公司 | Based on the data processing method and device for accelerating core |
| CN112769710A (en) * | 2021-01-21 | 2021-05-07 | 中国电子科技集团公司电子科学研究院 | Router and operation method thereof |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101150521A (en) * | 2007-11-05 | 2008-03-26 | 北京航空航天大学 | Congestion control method for multi-to-multi reliable multicast |
| CN101478483A (en) * | 2009-01-08 | 2009-07-08 | 中国人民解放军信息工程大学 | Method for implementing packet scheduling in switch equipment and switch equipment |
-
2010
- 2010-11-22 CN CN2010105582674A patent/CN101977158A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101150521A (en) * | 2007-11-05 | 2008-03-26 | 北京航空航天大学 | Congestion control method for multi-to-multi reliable multicast |
| CN101478483A (en) * | 2009-01-08 | 2009-07-08 | 中国人民解放军信息工程大学 | Method for implementing packet scheduling in switch equipment and switch equipment |
Non-Patent Citations (1)
| Title |
|---|
| 《中国优秀硕士论文全文数据库》 20100920 刘靖 《应用层多源组播拥塞控制研究》 , 2 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104507119A (en) * | 2014-11-26 | 2015-04-08 | 无锡儒安科技有限公司 | Concurrent transmission based wireless sensor network data collection method |
| CN104507119B (en) * | 2014-11-26 | 2018-09-14 | 无锡儒安科技有限公司 | A kind of radio sensor network data collection method based on concurrent transmission |
| CN109388609A (en) * | 2018-09-30 | 2019-02-26 | 中科驭数（北京）科技有限公司 | Based on the data processing method and device for accelerating core |
| CN112769710A (en) * | 2021-01-21 | 2021-05-07 | 中国电子科技集团公司电子科学研究院 | Router and operation method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105227481B (en) | SDN congestion control method for routing based on path cost and stream scheduling cost minimization | |
| CN106533970B (en) | Difference towards cloud computing data center network spreads transmission control method and device | |
| CN111580970B (en) | Transmission scheduling method for model distribution and aggregation of federated learning | |
| CN108092888A (en) | A kind of transmission method based on Overlay networks, gateway and Transmission system | |
| CN101729589A (en) | Method and system for improving end-to-end data transmission rate | |
| CN103179046A (en) | Data center flow control method and data center flow control system based on openflow | |
| CN106302228B (en) | The transfer control method of task based access control perception in a kind of data center network | |
| CN106685835A (en) | Method for achieving high-speed distributed routing between computational nodes of data center | |
| CN102571571A (en) | Multilayer effective routing method applied to delay tolerant network (DTN) | |
| CN102752192B (en) | Bandwidth allocation method of forwarding and control element separation (ForCES) transmission mapping layer based on stream control transmission protocol (SCTP) | |
| CN107948103A (en) | A kind of interchanger PFC control methods and control system based on prediction | |
| CN104767695B (en) | A kind of stream scheduling method of task rank in data center | |
| Jiang et al. | Enhancing traffic capacity of scale-free networks by employing hybrid routing strategy | |
| CN106161254A (en) | A kind of many purposes data transmission network road route device, method, chip, router | |
| CN115277574B (en) | Data center network load balancing method under SDN architecture | |
| CN114398174A (en) | Dynamic energy-saving scheduling method and system for data center network in SDN environment | |
| CN101977158A (en) | Method for realizing congestion control in multi-source application layer multicast | |
| Wang et al. | Load balancing for heterogeneous traffic in datacenter networks | |
| CN102231711B (en) | Route control method for dynamically regulating congestion level of nodes based on Wiener prediction | |
| CN102223311A (en) | Queue scheduling method and device | |
| CN104869113A (en) | Network transmission method and system based on minimum completion time | |
| CN107046507B (en) | A kind of jamming control method for multiplexed transport in DCN | |
| CN106372013B (en) | Long-distance inner access method, device and system | |
| CN106209683B (en) | Data transmission method and system based on data center's wide area network | |
| CN105072046B (en) | A kind of delay-tolerant network congestion-preventing approach based on the forwarding of Token Control node concurrent data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20110216 |