CN100413283C - Flow equilization processing method and device based on connection pair - Google Patents
Flow equilization processing method and device based on connection pair Download PDFInfo
- Publication number
- CN100413283C CN100413283C CNB2004100095934A CN200410009593A CN100413283C CN 100413283 C CN100413283 C CN 100413283C CN B2004100095934 A CNB2004100095934 A CN B2004100095934A CN 200410009593 A CN200410009593 A CN 200410009593A CN 100413283 C CN100413283 C CN 100413283C
- Authority
- CN
- China
- Prior art keywords
- data
- flow
- osi
- layer
- buffer area
- 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.)
- Expired - Lifetime
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 12
- 238000000034 method Methods 0.000 claims abstract description 15
- 230000008569 process Effects 0.000 claims abstract description 7
- 238000012545 processing Methods 0.000 claims description 21
- 230000006854 communication Effects 0.000 claims description 19
- 238000004891 communication Methods 0.000 claims description 14
- 230000005540 biological transmission Effects 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 7
- 238000012856 packing Methods 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 239000000284 extract Substances 0.000 claims description 5
- 238000005538 encapsulation Methods 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 3
- 238000003825 pressing Methods 0.000 claims description 2
- 230000008878 coupling Effects 0.000 claims 1
- 238000010168 coupling process Methods 0.000 claims 1
- 238000005859 coupling reaction Methods 0.000 claims 1
- 230000003139 buffering effect Effects 0.000 description 27
- 238000003860 storage Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 230000002457 bidirectional effect Effects 0.000 description 2
- 230000005574 cross-species transmission Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011144 upstream manufacturing Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention discloses a flow equalization processing method and a device based on a connection pair. The method comprises the following steps: up data and down data transmitted by a backbone network are copied in an image mode, and the access of a first layer of data of OSI is completed; the data is converted to a second layer of the OSI; the second layer of the OSI processes the data in a protocol mode, and the data is divided into bale units; SIP information and DIP information of the bale units are extracted, and a predetermined link distribution chart is inquired according to the SIP information and the DIP information; the data is cached into caching regions of the outlets according to allocation outlets in the link distribution chart; the data in the caching regions is packed and encapsulated according to the protocol of the second layer of the OSI; after the second layer of the OSI is transformed to the first layer, the data is transmitted on the first layer of the OSI. The present invention better solves the problem of equalizing shunting from the backbone network to a lower level network circuit, ensures the instant equalization of each lower level circuit, and ensures flow equalizing distribution between different bandwidths.
Description
Technical field
The present invention relates to a kind of flow equalization processing method and device, relate in particular to a kind of based on connecting right flow equalization processing method and device.
Background technology
In the internet, point-to-point communication by IP to realizing, setting up in the process of a communication, such as sending an envelope Email from server A, must the transmission request be proposed earlier by server A, if server B has resource that reception is provided to server B, so just sending response tells server A to send, server A sends to server B according to the desired form of Network Transmission with Email then, and last server A is reminded to send and finished, and server B discharges resource.Below as can be seen, no matter the transmission of valid data is to receive or send out to a side, this communication process itself be mutual, be the process of a duplex signaling.As shown in Figure 1, be the communication process schematic diagram between two users.Its own unique on network of sign is exactly its IP address for server A, such as 1234, be 4321 for server B, so in communication process, when data flow during from server A to server B the source address SIP of data flow be exactly 1234, destination address is 4321; When data flow during from server B to server A the source address SIP of data flow be exactly 4321, destination address is 1234.Regardless of data flow direction, 1234 and 4321 these pair of IP of forming are to being thick-and-thin.Based on the bidirectional traffic of this pair of IP, it is right to be referred to as to connect.
In order to set up the network system of communication between all types of computer systems, osi model provides the agreement of seven levels to realize, this is the Fundamentals of Mathematics that current most of network is set up.As shown in Figure 2, be the basic model of modern communications.The ground floor physical layer is used for carrying out actual transfer of data, and it is the light or the signal of telecommunication.The second layer is a data link layer, and it is responsible for data cell (bit group) zero defect sent to next adjacent physical address from a last physical address.And add that significant bit forms heading and message trailer comes packet is encapsulated, be referred to as frame.The three-layer network layer includes agreement (PROTOCOL), source IP address (SIP), purpose IP address (DIP), source port address (SP), destination interface address (DP) and the content etc. of message.
Setting up in the process of the Internet infrastructure, because the difference and the application oriented object of economic development and its construction level are different, thereby on the different circuits of different regions or areal, adopted the network bandwidth of different brackets, corresponding infrastructure can not be directly compatible, as direct compatibility just between 10G, 2.5G, 622M, 155M or the like.In the time need moving the data of on the 10G backbone network, transmitting to the 2.5G backbone network, just need 1: 4 ratio to build system at least, the situation of the 2.5G line idle that the 2.5G circuit overload that does not have has simultaneously but the data flow that how to guarantee a 10G flow can be spread on 4 2.5G circuits equably just need be distributed based on the equilibrium that guarantees flow someway.Same, as shown in Figure 3, even on the circuit of same grade, also may flow evenly need be divided into the transmission of several roads, to guarantee to be no more than the limit of back-end processing ability because of some application.As seen, the flow shunting is to compare common phenomena in the network service.
At present, generally be that source, destination address according to packet connects identification based on the flow equalization method of TCP/IP, and connection be assigned on the different nodes according to equalization algorithm.Obviously, what present algorithm was primarily aimed at is the node forwarding problems, and switches still not really ripe at flow between the different bandwidth.Corresponding flow equalization algorithm neither be very ripe, and implementation method also is not quite similar, and its assignment of traffic is not really even.
Publication number discloses a kind of packet switching equipment and method of packet communication network for the special Japan Patent " packet exchange device " of opening 2000-295274, at first, an IP bag reaches one of them low layer processing module, this module is connected to external network by a network interface, this module is carried out the bag of data link layer (the 2nd layer) and physical layer (the 1st layer) and is handled, send packet handing module after handling to, be saved in the bag memory block, afterwards, packet handing module extracts its IP destination address and IP source address from the packet of storage; Packet handing module produces a search instruction, and IP destination address and IP source address are as search parameter, and search parameter is used for searching at the IP flow chart; Packet handing module receives Search Results from the search processing module: mac source address, MAC destination address and a port numbers, packet handing module sends to the low layer processing module that is connected with the physics output port of Search Results appointment with this bag then, the CRC that the bottom layer treatment module is carried out whole bag calculates, and adding result of calculation to bag, the low layer processing module sends this bag by connecting network interface then.Yet at the flow equalization problem of swap data between different bandwidth, this technical scheme does not provide the method for solution.
Summary of the invention
At existing problem and shortage in the above-mentioned existing network communication flows equalization process, the purpose of this invention is to provide a kind of based on connecting right flow equalization processing method and device.
The present invention is achieved in that a kind of based on connecting right flow equalization processing method, may further comprise the steps:
1) the uplink and downlink data of backbone network transmission are done mirror-image copies, finish the access of the ground floor data of OSI, and with the second layer of described uplink and downlink data transaction to OSI;
2) second layer of described OSI carries out protocol processes to described uplink and downlink data, and described data separating is the bag unit;
3) source address (SIP) and destination address (DIP) information of the described bag of extraction unit, and according to this SIP and DIP information inquiry preset link distributing list, and the distribution of pressing in the link distributing list exports the buffer area of described bag unit caches to this outlet, to the encapsulation of packing of the bag unit of described buffer area, the second layer that carries out described OSI again ground floor at this OSI after the conversion of ground floor transmits by the second layer agreement of described OSI.
Further, default link distributing list in the described step 3) is to set up like this: if data source need be divided into the n road of flow equalization, then the IP address field with communication is mapped to the n road successively by adjacent mode of not going the same way, and the data traffic that is about to described backbone network is divided into n flow piece.
Further, this method also comprises, described n flow piece detected, when a certain flow piece flow surpasses threshold value, reduce the IP address that this flow piece is comprised, that is to say, reduce the IP address space that this flow piece is comprised, the IP address that increases other flow pieces simultaneously comprises quantity, promptly increases the IP address space of other flow pieces.
Further, described buffer area is a plurality of, adopts the repeating query mode to carry out when the buffer area data are encapsulated.
Further, when described buffer area overflows, this buffer area that overflows of priority treatment then.
According to another object of the present invention, comprise based on connecting right flow equalization processing unit:
OSI ground floor data processing and forwarding module are used for the access mirror-image copies of the ground floor data of OSI, and with described data forwarding to the OSI second layer;
OSI second layer data processing and forwarding module are used for by related protocol the ground floor data of OSI being handled, and described data separating is the bag unit;
The bag buffer area is used for data cached;
SIP and DIP information extraction modules are used for extracting the bag unit from described bag buffer area, and therefrom extract the SIP and the DIP information of this bag;
Calculating and comparing module, utilize SIP and the DIP information matches link distributing list extracted, after finding the path, distribute outlet described bag unit to be sent to the buffer area of this outlet by it, to the encapsulation of packing of the bag unit of described buffer area, the second layer that carries out OSI again transmits at ground floor after the conversion of ground floor by the second layer agreement of OSI.
Further, described link distributing list is to set up like this: if data source need be divided into the n road of flow equalization, then the IP address field with communication is mapped to the n road successively by adjacent mode of not going the same way, and the data traffic that is about to described backbone network is divided into n flow piece.
Further, described link distributing list is stored in the memory of this device.
Further, this device also comprises the flow detection module, is used for the bulk flow of each flow piece is detected, when surpassing threshold value as if a certain flow piece flow, reduce the IP address space scope that this flow piece is comprised, increase the IP address space scope of other flow pieces simultaneously.
Further, described buffer area is a plurality of, adopts the repeating query mode to carry out when the buffer area data are encapsulated; When buffer area overflows, this buffer area that overflows of priority treatment then.
The present invention is by being mapped to the IP address field of communication successively by adjacent mode of not going the same way the link of pre-distribution in communication process, make each link form the flow piece, substantially realized flow equalization, and, because communication process is based on connection to carrying out, therefore can guarantee that the bidirectional transmit-receive of communication data carries out on the same line road, data can not broken.The present invention is assigned to source traffic in some subordinate's circuits uniformly according to the purpose IP address, source of each packet just, thereby has realized the flow equalization distribution.
In addition, for avoiding the actual flow inequality of flow piece, promptly other flow piece flows are too small in order to avoid a certain flow piece overload, the present invention also is provided with the flow detection mode, and each flow piece is detected in good time, and promptly the flow piece feeds back its flow, when the flow that detects a certain IP flow piece is excessive, dwindle this IP flow piece and comprise the scope of concrete IP value, thereby reduce flow, realize instant flow equalization.The present invention has solved the equilibrium shunting of backbone network to the undernet circuit well, and can guarantee the instant equilibrium of each subordinate's circuit, has guaranteed the flow equalization distribution between the different bandwidth.
Description of drawings
Below in conjunction with accompanying drawing, the present invention is made detailed description.
Fig. 1 is general communication structure schematic diagram;
Fig. 2 is a Network Transmission model structure schematic diagram;
Fig. 3 is existing flow equalization structural representation;
Fig. 4 is a data processing structure schematic diagram of the present invention.
Embodiment
As shown in Figure 4, the uplink and downlink data of backbone network transmission are done the access that mirror-image copies has just been finished the ground floor data of OSI.Ground floor at OSI, what transmit all is the signal of telecommunication or the light signal of simulation or numeral, though the signal of this layer itself also comprises SIP and DIP information, but these signals all are to be delivered to the other end with the high-speed of serial from an end, the present invention does not do any improvement to the ground floor of OSI, directly receive by this layer data and processing module to signal processing, guarantee that the high speed of its signal is unimpeded.Because if want from the ground floor of OSI, directly to parse SIP and DIP, be very difficult.At first will isolate packet unit, be the unit caches data with the packet again, then data analyzed, and extracts correct SIP and DIP according to different data type, and then calculate the flow direction after its shunting according to SIP and DIP.This is to waste time and energy and uneconomic method.
And the second layer of OSI, what import is the digital electric signal of the protocol compliant regulation of ground floor output, this signal of telecommunication is confusing physically, but processing through the second layer, isolate packet unit one by one the digital electric signal of the protocol compliant after handling through ground floor, and with its head and the tail of mode sign of independent control signal, to mistake, verification etc.Its output be exactly bag unit one by one, and erased bag tail tag bit position, packet header necessary when physical layer transmission, and when sending packet, unwrapped the beginning and finish to provide synchronously independently control signal at number to the 3rd layer.The Data Receiving of the second layer and processing module can be ready-made asic chips, also can be to design embedded chip voluntarily by the user, but must satisfy relevant protocol type, these agreements be according to different Network Transmission bandwidth, switch, route etc. have nothing in common with each other.
The packet of bag buffering area one after with second layer protocol analysis is that unit is stored in the buffering area with the bag.Buffering area of the present invention can be a storage medium arbitrarily such as RAM, FIFO, FLASH, as long as its read or write speed must adapt to the requirement on the whole system speed.Its storage size needs and can set arbitrarily according to system.The SIP/DIP information extraction modules is analyzed the affiliated protocol type and the SIP/DIP information of each bag from bag buffering area one, and with SIP/DIP information together with the first address that wraps in storage in the buffering area one, information such as length are exported.Because the position that different its SIP/DIP of data of agreement stores in message also is not quite similar.Thereby when analyzing data SIP/DIP, at first to analyze the protocol type of packet, extract the SIP/DIP information of packet again in the packet relevant position according to the data pack protocol type, the record bag is long simultaneously, wrap in first address information of storing in the buffering area etc., send to downstream module together, use during for moving data.
Calculating and comparing module receives that to call IP after the SIP/DIP information that information extraction modules sends right---bag output is pointed in the corresponding list and is inquired about, obtain the direction of this packet outlet, then together with packet address stored in buffering area one, information such as length send to the data-moving module together.Here, IP is right---and it promptly is the link distributing list that corresponding list is pointed in bag output.Here, the link distributing list is to set up like this: because the IP address is a Normal Distribution on the network, that is to say it is that macroscopic view is evenly continuous.The present invention adopts the IP address to be mapped on the link successively by adjacent mode of not going the same way.Soon all IP from 0.0.0.0 to 255.255.255.255 guarantee between each piece roughly the same to defined data flow by IP to being dispersed as several bulks simultaneously.With 0.0.0.0 to all IP of 255.255.255.255 according to adjacent different principle, define the dispense flow rate direction of each IP.Such as, if the data source of backbone network need be divided into 5 tunnel of flow equalization, each IP will shine upon in such a way so: IP is that the data of 0.0.0.0 output to the first via, 0.0.0.1 to the second the tunnel, 0.0.0.2 is to Third Road, 0.0.0.3 to the four the tunnel, 0.0.0.4 to the five the tunnel, 0.0.0.5 to the first via, 0.0.0.6 is to the second the tunnel ... 255.255.255.254 to the five the tunnel, 255.255.255.255 is to the first via.Because the source destination address of communication two party exchanges, so that the present invention has guaranteed simultaneously that also uniform flow rate is based on connection is right.Because its IP of data flow of network delivery is normal distribution, macroscopic view is uniform, so this processing method can guarantee the flow basis equalization of the IP piece that each is big.
IP of the present invention pair is pointed to corresponding list with bag output and has set up the IP pair of correspondence table of pointing to outlet that covers entire I P address scope.Each IP points to corresponding outlet is arranged.
Data-moving module one is according to the first address that wraps in storage in the buffering area one, and length information is moved packet the buffering area two from buffering area one, simultaneously the output directional information of additional packets.Here, SIP/DIP information extraction modules, calculating and comparing module, data-moving module can be by embedded design, and also available ASIC realizes.
Two storages of bag buffering area have the packet unit that exports directional information.
Data-moving module two according to certain strategy, is moved whole bag successively, and is write definite subordinate's buffering area according to their sensing from two bag buffering areas two from two buffering areas---bag buffering area three.The strategy of moving here is specially, from two bag buffering areas two repeating query (table tennis) move bag, and, when one of them wraps buffering area two spill-overs, the bag buffering area two of this spill-over of priority treatment.In this module, broken up from the up-downgoing data that data source is come, moved separately respectively according to the difference of right difference of IP rather than up-downgoing again and sent in the data buffer zone.Each sends data in buffering area and had both comprised upstream data and also comprise downlink data, and only to be whether be based on IP right to relevant connection in the difference of packet between different buffering areas.Can realize by embedded system or ASIC.
Wrap buffering area three storage back each self-corresponding all packet of each outlet line of level, and erase the bag directional information, determine that the buffering area that points to has suffered because wrap this moment to be stored in.Independently wrap tail information in control signal sync mark packet header at dateout Bao Shiyong, use when reading for downstream module.Buffering area three and thereafter the number of each processing module of polyphone decide according to the ratio of Data Receiving source and transmitting terminal.Be transformed into the 622M circuit such as 1 road 2.5G upstream data, just must 4 send buffering area and its rear end polyphone module chain at least.
Second layer packing sending module is packed packet in the buffering area and is exported according to second layer protocol requirement.The dateout form meets the requirement of second layer data output protocol.Here, the packing sending module is data processing and forwarding module.The present invention also comprises the traffic statistics module that each shunt link flow is detected, detect by the data traffic that each second layer packing sending module is handled, can obtain each flow information of link along separate routes, if surpassed threshold value, then dwindle this IP flow piece and comprise the scope of concrete IP value, thereby the reduction flow is realized instant flow equalization.
Like this, the present invention has realized the equilibrium distribution of flow, just the rate of discharge of each IP piece correspondence should approximate equality, link distributing list of the present invention is a dynamic list, because flow is always unsettled, although meet normal distribution in the distribution of flow on the macroscopic view in the IP address, but in local time, but might produce unstable unbalanced, so IP is right-and corresponding list is pointed in outlet must can be according to the variation of each rate of discharge, time update respectively exports the right coverage of corresponding IP, both can satisfy total coverage rate 100%, can realize fine setting to interblock at each sub-IP again.
Ground floor conversion sending module carries out the data transaction of ground floor, and machinery and the electrical characteristic transmission Bit data according to interface flows on the circuit then.
The present invention can realize that the flow equalization of 1 road 10G when 4 road 2.5G flows are changed handle; 1 road 2.5G handles to the flow cycled flow equalization of 4 road 622M; 1 road 622M handles or the like to the flow equalization of 4 road 155M.
The present invention does not consider the difference of up-downgoing, puts on an equal footing for the up-downgoing data that data source is come, and each data flow of output is propped up the difference of also not doing up-downgoing.If need also consider the difference of up-downgoing when using desire of the present invention to realize flow equalization at the dateout end, then need before data-moving module two is carried out data-moving, the difference (difference of up-downgoing) according to data buffer zone, the left and right sides carry out mark, put into the buffering area of appointment according to the difference of up-downgoing mark and output sensing when moving.The realization number that sends buffering area three than present Duo one times get final product.
Claims (8)
1. one kind based on connecting right flow equalization processing method, may further comprise the steps;
1) the uplink and downlink data of backbone network transmission are done mirror-image copies, finish the access of the ground floor data of OSI, and with the second layer of described uplink and downlink data transaction to OSI;
2) second layer of OSI carries out protocol processes to described uplink and downlink data, and described data separating is the bag unit;
3) source address and the destination address information of the described bag of extraction unit, and according to this source address and destination address information inquiry preset link distributing list, and the distribution of pressing in the link distributing list exports the buffer area of described bag unit caches to this outlet, to the encapsulation of packing of the bag unit of described buffer area, the second layer that carries out OSI again ground floor at this OSI after the conversion of ground floor transmits by the second layer agreement of OSI;
It is characterized in that, default link distributing list in the described step 3) is to set up like this: if data source need be divided into the n road of flow equalization, then the IP address field with communication is mapped to the n road successively by adjacent mode of not going the same way, and the data traffic that is about to described backbone network is divided into n flow piece.
2. as claimed in claim 1 based on connecting right flow equalization processing method, it is characterized in that, this method also comprises step: described n flow piece detected, when a certain flow piece flow surpasses threshold value, reduce the IP address space scope that this flow piece is comprised, increase the IP address space scope of other flow pieces simultaneously.
3. as claimed in claim 1ly it is characterized in that based on connecting right flow equalization processing method described buffer area is a plurality of, when the buffer area data are encapsulated, adopt the repeating query mode to carry out.
4. as claimed in claim 3ly it is characterized in that based on connecting right flow equalization processing method, when described buffer area overflows, this buffer area that overflows of priority treatment then.
5. one kind based on connecting right flow equalization processing unit, comprising:
OSI ground floor data processing and forwarding module are used for the access mirror-image copies of the ground floor data of OSI, and with described data forwarding to the OSI second layer;
OSI second layer data processing and forwarding module are used for by related protocol the ground floor data of OSI being handled, and described data separating is the bag unit;
The bag buffer area is used for data cached;
Source address and destination address information extraction module are used for extracting the bag unit from described bag buffer area, and therefrom extract the source address and the destination address information of this bag;
Calculating and comparing module, utilize source address and the destination address information coupling link distributing list extracted, after finding the path, distribute outlet described bag unit to be sent to the buffer area of this outlet by it, to the encapsulation of packing of the bag unit of described buffer area, the second layer that carries out OSI again transmits at ground floor after the conversion of ground floor by the second layer agreement of OSI;
It is characterized in that, described link distributing list is to set up like this: if data source need be divided into the n road of flow equalization, then the IP address field with communication is mapped to the n road successively by adjacent mode of not going the same way, and the data traffic that is about to described backbone network is divided into n flow piece.
6. as claimed in claim 5 based on connecting right flow equalization processing unit, it is characterized in that described link distributing list is stored in the memory of this device.
7. as claimed in claim 5 based on connecting right flow equalization processing unit, it is characterized in that, this device also comprises the flow detection module, be used for the bulk flow of each flow piece is detected, when if a certain flow piece flow surpasses threshold value, reduce the IP address space scope that this flow piece is comprised, increase the IP address space scope of other flow pieces simultaneously.
8. as claimed in claim 5ly it is characterized in that based on connecting right flow equalization processing unit described buffer area is a plurality of, when the buffer area data are encapsulated, adopt the repeating query mode to carry out; When buffer area overflows, this buffer area that overflows of priority treatment then.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100095934A CN100413283C (en) | 2004-09-21 | 2004-09-21 | Flow equilization processing method and device based on connection pair |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100095934A CN100413283C (en) | 2004-09-21 | 2004-09-21 | Flow equilization processing method and device based on connection pair |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1599356A CN1599356A (en) | 2005-03-23 |
| CN100413283C true CN100413283C (en) | 2008-08-20 |
Family
ID=34662540
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2004100095934A Expired - Lifetime CN100413283C (en) | 2004-09-21 | 2004-09-21 | Flow equilization processing method and device based on connection pair |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100413283C (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100505684C (en) * | 2005-03-29 | 2009-06-24 | 国际商业机器公司 | Network system, flow equalization method, network monitoring device and host machine |
| CN100426773C (en) * | 2005-11-30 | 2008-10-15 | 中兴通讯股份有限公司 | Method for equalizing port flow while multiple-MAC-port inter-connecting |
| CN103067999B (en) * | 2009-01-23 | 2015-12-16 | 雷凌科技股份有限公司 | Package transmitting method and device |
| CN101635720B (en) * | 2009-08-31 | 2012-09-05 | 杭州华三通信技术有限公司 | Filtering method of unknown flow rate and bandwidth management equipment |
| CN102209035B (en) * | 2011-05-25 | 2014-10-15 | 杭州华三通信技术有限公司 | Traffic forwarding method and devices |
| CN102420752B (en) * | 2011-11-28 | 2015-02-04 | 曙光信息产业(北京)有限公司 | Dynamic distribution device under 10Gbps flow |
| CN102664789B (en) * | 2012-04-09 | 2016-08-17 | 北京百度网讯科技有限公司 | The processing method of a kind of large-scale data and system |
| CN105591965B (en) * | 2015-12-28 | 2018-12-14 | 北京锐安科技有限公司 | flow equalization output method and device |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000295274A (en) * | 1999-04-05 | 2000-10-20 | Nec Corp | Packet exchange |
| US20020161918A1 (en) * | 2001-04-27 | 2002-10-31 | Fujitsu Limited Of Kawasaki, Japan | Packet transmission system in which packet is transferred without replacing address in the packet |
| US20030067915A1 (en) * | 2001-10-04 | 2003-04-10 | Alcatel | Network nodes |
| CN1434604A (en) * | 2002-01-22 | 2003-08-06 | 瑞昱半导体股份有限公司 | Link layer communication protocol controlled swapping controller and control method thereof |
| US20040003116A1 (en) * | 1998-10-30 | 2004-01-01 | Science Applications International Corporation | Agile network protocol for secure communications with assured system availability |
| US6751221B1 (en) * | 1996-10-04 | 2004-06-15 | Kabushiki Kaisha Toshiba | Data transmitting node and network inter-connection node suitable for home network environment |
| CN2749188Y (en) * | 2004-09-08 | 2005-12-28 | 北京锐安科技有限公司 | Traffic balancing device based on connection couple |
-
2004
- 2004-09-21 CN CNB2004100095934A patent/CN100413283C/en not_active Expired - Lifetime
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6751221B1 (en) * | 1996-10-04 | 2004-06-15 | Kabushiki Kaisha Toshiba | Data transmitting node and network inter-connection node suitable for home network environment |
| US20040003116A1 (en) * | 1998-10-30 | 2004-01-01 | Science Applications International Corporation | Agile network protocol for secure communications with assured system availability |
| JP2000295274A (en) * | 1999-04-05 | 2000-10-20 | Nec Corp | Packet exchange |
| US20020161918A1 (en) * | 2001-04-27 | 2002-10-31 | Fujitsu Limited Of Kawasaki, Japan | Packet transmission system in which packet is transferred without replacing address in the packet |
| US20030067915A1 (en) * | 2001-10-04 | 2003-04-10 | Alcatel | Network nodes |
| CN1434604A (en) * | 2002-01-22 | 2003-08-06 | 瑞昱半导体股份有限公司 | Link layer communication protocol controlled swapping controller and control method thereof |
| CN2749188Y (en) * | 2004-09-08 | 2005-12-28 | 北京锐安科技有限公司 | Traffic balancing device based on connection couple |
Non-Patent Citations (3)
| Title |
|---|
| US 6 751 221 B1 2004.06.15 |
| 关于IP骨干网络设计中负载均衡的分析. 袁捷.电信工程技术与标准化，第4期. 2002 |
| 关于IP骨干网络设计中负载均衡的分析. 袁捷.电信工程技术与标准化，第4期. 2002 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1599356A (en) | 2005-03-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1166246C (en) | System supporting variable bandwidth asynchronous transfer mode network access for wireline and wireless communications | |
| CN100499872C (en) | Voice-over IP communication without echo cancellation | |
| CN102106125B (en) | A kind of multi-path network | |
| CN100574310C (en) | A kind of credit flow control method | |
| CN101939955B (en) | Method and system for controlling link saturation of synchronous data across packet networks | |
| CN102726009B (en) | Data flow exchange control method and relative device | |
| CN101534249B (en) | Method for transmitting data on bound link and network equipment thereof | |
| CN109302372A (en) | A kind of communication means, equipment and storage medium | |
| CN101258719A (en) | Method to extend the physical reach of an Infiniband network | |
| CN100477642C (en) | Ethernet access device and access method thereof | |
| CN101282305B (en) | Bandwidth control method for distributed system as well as service plate | |
| CN1893387B (en) | Encapsulation of e1-type frames under ethernet | |
| CN102883374A (en) | Resource distribution method for satellite communication system | |
| CN101626384A (en) | Method and system for transmitting data packet based on TPC protocol | |
| CN101189843A (en) | Electronic device and method of communication resource allocation | |
| CN100413283C (en) | Flow equilization processing method and device based on connection pair | |
| CN101616064B (en) | Method for managing data, mesh network system and associated device | |
| US7349401B2 (en) | Bonded G.shdsl links for ATM backhaul applications | |
| CN103152292A (en) | Device and method for framing XGEM (XGPON Encapsulation Method) in ONU (Optical Network Unit) in XGPON (XG-Passive Optical Network) system | |
| CA2312056C (en) | Communications channel synchronous micro-cell system for integrating circuit and packet data transmissions | |
| CN101064697B (en) | Apparatus and method for realizing asynchronous transmission mode network service quality control | |
| CN110351202A (en) | 5G core net traffic packet method, apparatus, equipment and computer storage medium | |
| CN109995608A (en) | Network rate calculation method and device | |
| CN105680928B (en) | Large capacity check-in signal captures and processing method and system | |
| CN101426152B (en) | Data transmission method for integrated mixed optical network based on burst exchange |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |