CN1889527A - Load sharing apparatus and method for realizing multi-chain circuit transmission - Google Patents
Load sharing apparatus and method for realizing multi-chain circuit transmission Download PDFInfo
- Publication number
- CN1889527A CN1889527A CNA2006101035696A CN200610103569A CN1889527A CN 1889527 A CN1889527 A CN 1889527A CN A2006101035696 A CNA2006101035696 A CN A2006101035696A CN 200610103569 A CN200610103569 A CN 200610103569A CN 1889527 A CN1889527 A CN 1889527A
- Authority
- CN
- China
- Prior art keywords
- link
- load sharing
- module
- ppp
- processing
- 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
A load sharing device of multilink transmission distributes load sharing dispatch module of usable link channel dynamically for IP data packet by adding fault detection module used for dynamically detecting quality of each link according to report result of fault detection module in order to let said device realize load sharing of multilink transmission for effectively relaxing load pressure on each single link.
Description
Technical field
The present invention relates to the multilink transmission technology, relate in particular to a kind of apparatus and method that realize the load sharing of multilink transmission.
Background technology
In communication system, (BE, BestEffort) mechanism have advantages such as flexibility and versatility to the IP technology, therefore obtain application more and more widely at the aspects such as transmission of signaling, data and voice based on address (IP Address) and forwarding as possible.In theory, the IP technology can be used on any physical media, but in fact, IP uses on Ethernet the earliest and at most still, because a MAC Address is arranged on the Ethernet card, just can be mapped one by one with the IP address.Like this, the link layer on the Ethernet is very simple, realizes a medium access control (MAC) function exactly, does some Channel Processing, the work of collision detection and so on.
Be not based on the address and be based on the physical media of connection if the IP technology is applied in, for example E1 and ATM(Asynchronous Transfer Mode) etc. just need be done some mappings at link layer, the notion of connection are mapped to the notion of address.Address is a point just, and PPP(Point-to-Point Protocol) promptly is for a kind of like this consideration, at first is used to provide a kind of method of standard, converts connection to address, so that the IP technology can be convenient to use on multiple different physical media.PPP a kind ofly can be used for data link layer protocol asynchronous or the synchronous serial medium, and its uses LCP (LCP) to set up and the service data link connects, and Network Control Protocol (NCP) allows to use the multiple network layer protocol in point-to-point connection.When PPP connected beginning, link through 3 session establishment phases was respectively: link establishment phase; Authentication phase; Network layer protocol phase.
Carrying the PPP transmission of the common employing of IP on E1 with E1 is PPP over E1 mode.This mode provides link establishment, authentication, error detection occurs, a whole set of mechanism of IP compression negotiation, thereby adopted in a large number.As shown in Figure 1, in this transmission means, exist a plurality of PPP to connect between local terminal and the far-end, each PPP connects corresponding different equipment routed ip address.The routed ip address of equipment is visible to outside transmission network, and, there are how many bar PPP to connect, what equipment routed ip address are just arranged.Under this mode, the bandwidth of every ppp link generally is limited in 64kbit/s, even if adopt High-Level Data Link Control (HDLC) time slot binding technology, link bandwidth also is no more than 2M.And along with the expansion of transmission capacity, the IP address is more and more, makes troubles for management and use.And because do not realize the software load sharing, like this, single link congested probability occurs and strengthens, and a transmission line failure can't remedy by other hardware.Though, in the Request for Comment of Internet engineering duty tissue (IETF RFC) MultiLink PPP transmission (PPP MP has been proposed, PPP Muti Link) agreement with the multilink bundle problem that solves the IP transmission, is promptly utilized multilink to bundle and is enlarged the IP transmission bandwidth.But it does not stipulate more there is not concrete implementation for load sharing how to realize multilink.
Summary of the invention
In view of this, first main purpose of the present invention provides a kind of device of realizing the load sharing of multilink transmission, uses this device can effectively alleviate the load pressure of single link, realizes the load sharing of multilink transmission.
Second main purpose of the present invention provides a kind of method that realizes the load sharing of multilink transmission, and this method can realize the load sharing of multilink transmission, effectively alleviates the load pressure of single link.
First main purpose of the present invention is achieved by the following technical solution:
A kind of device of realizing the load sharing of multilink transmission, comprising: many point-to-point ppp links, in addition, this device also comprises fault detection module and load sharing scheduler module.
Wherein, fault detection module is used for each bar ppp link quality of dynamic monitoring, and link quality information is reported to the load sharing scheduler module;
The load sharing scheduler module is used for according to the link quality information by the fault detection module report each link being carried out load sharing processing.
Wherein, described fault detection module can be arranged in the described load sharing scheduler module.
Wherein, described load sharing scheduler module comprises: load sharing computing module and point-to-point MultiLink PPP MP processing module.
Wherein, the load sharing computing module is used for according to the link quality information by the fault detection module report, each link is carried out load sharing calculate, and result of calculation is exported to PPP MP processing module.
PPP MP processing module is used to assemble many ppp links and forms single high-band wide beam MP bundle; PPP MP processing module is carried out the load sharing result calculated from the intrafascicular link of choosing transmission current IP packet of MP according to the output of load sharing computing module to link, thereby realizes the load balancing between each ppp link.
Wherein, described load sharing computing module comprises: computing module, user's Control Parameter memory module and link quality information memory module.
Wherein, user's Control Parameter memory module, what be used to store user's input calculates relevant Control Parameter with load sharing;
The link quality information memory module is used for the link quality information that the dynamic memory fault detection module is sent;
Computing module, be used for reading the IP address of current IP packet from PPP MP processing module, and read user's Control Parameter in user's Control Parameter memory module, then according to the link quality information in the link quality information memory module, each link is carried out load sharing to be calculated, for the current IP packet calculates current available link channel number, and export to PPP MP processing module.
Preferably, described load sharing computing module further comprises: IP address conversion module and chain line modular converter.
Wherein, the IP address conversion module is used for IP address transition with the described current IP packet that reads from PPP MP processing module and becomes IP allocation index number, exports to computing module;
Computing module receives this IP allocation index number, and in conjunction with the user's Control Parameter that has read, the link channel call number available for the current IP packet calculates according to link quality information exported to chain line modular converter;
Chain line modular converter is used for converting the link channel call number that computing module is exported to the chain line and exports to PPP MP processing module.
In the said apparatus, every ppp link of described many ppp links comprises: LCP processing module and NCP processing module; Perhaps every of described many ppp links ppp link includes only the LCP processing module, and many shared NCP processing modules of ppp link.Wherein, the shared NCP processing module of described many ppp links can be arranged in the described PPP MP processing module.
Second main purpose of the present invention is achieved by the following technical solution:
A kind of method that realizes the load sharing of multilink transmission, this method comprises the steps:
A, each bar ppp link quality of dynamic monitoring, and generate link quality information;
B, each bar link is carried out load sharing processing according to link quality information.
Wherein, link quality information comprises described in the step B: in link failure information, link congestion information, link-recovery information and the link normal information one or more than one combination in any; The then described load sharing processing of carrying out is: according to the above-mentioned quality information of each bar link, choosing current quality information is that link link normal or link-recovery distributes for the current IP packet as available link.
Wherein, described step B takes a step forward and comprises carrying out load sharing processing: read user's Control Parameter; Then describedly further comprise when carrying out load sharing processing: carry out load sharing according to user's Control Parameter and calculate.And described user's Control Parameter comprises: the link selection step-length; Then described when carrying out load sharing processing, if be the current link channel fault of IP allocation of packets or when congested, be next bar available link passage of current IP allocation of packets then according to this link selection step-length.
Preferably, described user's Control Parameter also comprises: the link initial offset; Then described when carrying out load sharing processing, select initial link circuit according to this link initial offset during for current IP allocation of packets available link.
Preferably, described user's Control Parameter also comprises: link assignment at interval; Then described when carrying out load sharing processing, when being different IP allocation of packets available link, select different initial link circuits at interval according to link assignment.
Preferably, further comprise among the described step B: the purpose IP address of input IP packet; Then described when carrying out load sharing processing, according to the IP address of input each link is carried out load sharing and calculate, be the current available link of the IP allocation of packets of this IP address.
Preferably, further comprise behind the input IP address among the described step B: with the IP address transition is the IP call number; Then described when carrying out load sharing processing, according to this IP call number each link is carried out load sharing and calculate, be the current available link of the IP allocation of packets of this IP address.
Preferably, when described step B carries out load sharing processing,, again link index numbers is converted to the chain line afterwards, distributes the available link passage according to the chain line for the current IP packet calculates available link index numbers earlier.
Preferably, set in advance chain travel permit number, and described link index numbers recycles, if the link index numbers of distributing surpasses the maximum link call number, then this link index numbers is carried out mould and remove chain travel permit number, make described link index numbers come back to legal interval.
Device provided by the present invention, by having increased each bar link-quality is carried out the fault detection module of dynamic monitoring and is the load sharing computing module of current IP packet dynamic assignment available link passage according to the monitoring result of fault detection module, thereby make this device realize the load sharing that multilink is transmitted, effectively alleviated the load pressure of single link; And when certain bar link breaks down by distributing next bar available link for current business, thereby can be implemented in certain bar link and break down the time with professional mean allocation to all the other normal links; And compare with PPP over E1 mode, this device is owing to removed the coupling of the quantity of IP number of addresses and transmission link, thereby can be at some independently on the link, according to the random one or more IP of the configuration address of user's needs, expanded the bandwidth of ppp link.
Method provided by the present invention, by each bar link-quality of dynamic monitoring, thereby the link channel that can use for current IP packet dynamic assignment, such as certain bar link failure or when congested, just be the available link of next bar of current IP allocation of packets, if certain bar chain route fault or congestion state return to normal condition, then this link continues the link selected as the current IP packet.Thereby realized the load sharing of multilink transmission.
Description of drawings
Fig. 1 is the transmission means structural representation of PPP over E1 in the prior art;
Fig. 2 is for realizing the structural representation of first preferred embodiment of device of the load sharing that multilink transmits among the present invention;
Fig. 3 is the internal structure schematic diagram of middle load sharing computing module embodiment illustrated in fig. 2;
Fig. 4 is for realizing the structural representation of second preferred embodiment of device of the load sharing that multilink transmits among the present invention;
Fig. 5 is for realizing the structural representation of the 3rd preferred embodiment of device of the load sharing that multilink transmits among the present invention;
Fig. 6 is for realizing the overview flow chart of the method embodiment of the load sharing that multilink transmits among the present invention;
Fig. 7 is the detail flowchart of middle step 602 embodiment illustrated in fig. 6.
Embodiment
For making the purpose, technical solutions and advantages of the present invention clearer, below in conjunction with embodiment and accompanying drawing, the present invention is described in more detail.
Basic thought of the present invention is: by the quality of each bar link of dynamic monitoring, i.e. and whether fault, congested or recovery and situation such as normal, the link channel that can use for current IP packet dynamic assignment, thus reach the equilibrium of load sharing.
Be described further below in conjunction with the device of specific embodiment the load sharing of realization multilink transmission of the present invention.
Embodiment one:
Referring to Fig. 2, the structural representation of first preferred embodiment of the device that Fig. 2 realizes multi-link load share for the present invention is a kind of.As shown in the figure, this device comprises: 8 ppp links, wherein every ppp link all has a LCP processing module 200, and shared NCP processing module of a plurality of LCP processing modules and multilink control protocol (MLCP) processing module, the common formation of NCP processing module and MLCP processing module point-to-point multilink (PPP MP) processing module 230, also comprise a fault detection module 210 and a load sharing computing module 220 in addition, wherein PPP MP processing module 230 and load sharing computing module 220 common formation load sharing scheduler modules.In addition, for the transmission course of complete description to data and voice, this device also comprises an IP protocol process module 240.
Wherein, LCP processing module 200 is used for setting up, disposes and stops PPP and is connected; According to configuration requirement, start verification process; Behind authentication success, notice PPP MP processing module 230 starts the negotiations process of network layer config option.In transmission of data packets, LCP processing module 200 is used to receive the data that PPP MP processing module sends and sends to far-end, and receives the data that far-end transmits and send to PPP MP processing module.
PPP MP processing module 230 is used to assemble multilink and forms single high-band wide beam, promptly produces the MP bundle that comprises many ppp links; After PPP MP processing module 230 receives the notice of the startup network layer config option negotiations process that LCP processing module 200 sends, configuration network layer protocol option on the ppp link of building up, after successfully finishing negotiation, informing network layer link arrives state, starts simultaneously to transmit and receive the network layer data bag.In transmission of data packets, after PPP MP processing module 230 receives the packet of IP protocol process module 240 transmissions, send inquiry to load sharing computing module 220, inquiry is the available link passage of this IP allocation of packets, this querying method can be realized by the function that calls load sharing computing module 220 and provide, then according to the result of load sharing computing module 220 output from the intrafascicular link that sends the current IP packet of choosing of MP, the IP packet is sent to the LCP processing module 200 of respective link; And the packet that will be received from LCP processing module 200 sends to IP protocol process module 240.
IP protocol process module 240 is used to handle Internet protocol, comprises receiving the IP packet that PPP MP processing module 230 sends over, and the route of IP packet is selected in the analysis purpose address, and packet is sent to this route corresponding port; Perhaps send the IP packet to PPP MP processing module 230.
Load sharing computing module 220 is used to be responsible for realize that load sharing calculates between the ppp link of many activities that PPP MP restraints.Promptly read user configured Control Parameter, and according to the link quality information of fault detection module 210 report, and the IP address of the current IP packet that from PPP MP processing module 230, receives, for the transmission of this IP packet is selected available, promptly do not break down or congested link channel, afterwards selection result is exported to PPP MP processing module 230.User's Control Parameter wherein can comprise chain travel permit number, is 8 in the present embodiment; Link initial offset, purpose are always to choose the less link of call number during for fear of original allocation, so beginning just increases a side-play amount, defaultly select 0; Link assignment at interval, purpose is for avoiding adjacent IP layer user too close on link selection, an assignment interval is set, and defaultly selects 1; And the link selection step-length, can jump to next bar available link etc. according to this step value.
Wherein, the internal structure of load sharing computing module 220 can comprise further as shown in Figure 3: IP address conversion module 221, computing module 222, link index modular converter 223, user's Control Parameter memory module 224 and link quality information memory module 225.
Wherein, user's Control Parameter memory module 224, what be used to store user's input calculates relevant user's Control Parameter with load sharing.
Link quality information memory module 225 is used for information such as relevant link failure that dynamic memory fault detection module 210 sends, congested or recovery.
IP address conversion module 221 is used for the IP address transition is become IP allocation index number, exports to computing module 222.
Chain line modular converter 223 is used for converting the link channel call number of computing module 222 outputs to the chain line, exports to PPP MP processing module 230.
If during specific to this end device, 220 of this load sharing computing modules carry out load sharing to multilink and calculate when data send, and when Data Receiving, then by the load sharing computing module in the end device being finished the link channel that the opposite end sends the IP packet, promptly local terminal receives the distribution of the link channel of this IP packet.Therefore, clearer for what describe, respectively the transmission of information between each module in process of transmitting and the receiving course is described in detail below.
Process of transmitting:
IP protocol process module 240 will need source address and the destination address to the data of opposite end transmission and data transmission, and parameters such as type of service are assembled into the IP packet, are transferred to PPP MP processing module 230.
PPP MP processing module 230 is called the function that load sharing computing module 220 provides after receiving this IP packet.
Load sharing computing module 220 reads user configured user's Control Parameter, and according to the link quality information of fault detection module 210 report, and the IP address of the current IP packet that from PPP MP processing module 230, receives, for the transmission of this IP packet is selected available, promptly do not break down or congested link channel, the result is returned to PPP MP processing module 230.
The result that PPP MP processing module 230 is returned according to load sharing computing module 220 finds out the link that sends the current IP packet, and the LCP processing module 200 that then this IP packet is sent to corresponding ppp link is handled.During this period, if MLCP processing module or NCP processing module need be sent control information to the other side, MLCP processing module or NCP processing module itself just can be initiated, and directly sends control data information to far-end, and data are sent to LCP processing module 200.
Receiving course:
Packet from the far-end to the local terminal after physical layer receives packet, sends to LCP processing module 200.After 200 pairs of data of LCP processing module are analyzed, handle according to RFC1661 regulation: if control information, LCP processing module 200 just can be handled itself, directly to far-end echo control data information or carry out controlling links and operate; If data message then parses packet wherein, send to PPP MP processing module 230.
After PPP MP processing module 230 receives packet, find, directly to far-end echo control data message or carry out the controlling links operation if the MLCP control information is then handled according to RFC1990; If the NCP control information then handles according to RFC1661, directly to far-end echo control data information or carry out the controlling links operation; If data message then parses packet wherein, send to IP protocol process module 240.
After IP protocol process module 240 receives data, resolve destination address wherein, find corresponding route, and packet is forwarded to designated port.
Can find that from said process when receiving data, 220 pairs of links of load sharing computing module are left intact, the load sharing computing module 220 by the opposite end is good according to the load sharing algorithm assigns because be used for receiving the link of this IP packet.
In the practical application, many shared NCP processing modules of ppp link also can not be arranged in the PPPMP processing module.
Embodiment two:
Referring to Fig. 4, Fig. 4 is the structural representation of a kind of second preferred embodiment of the device of realizing multi-link load share among the present invention.In the device of this embodiment, every ppp link of many ppp links that comprise comprises respectively: LCP processing module and NCP processing module, and form independent module 400, only comprise the MLCP processing module in the PPP MP processing module 410 in addition, the function of other module and annexation with embodiment illustrated in fig. 2 in identical.
Wherein, PPP MP processing module 410 is used to assemble multilink and forms single high-band wide beam, promptly produces the MP bundle that comprises many ppp links; After PPP MP processing module 410 receives data, send inquiry to load sharing computing module 220, inquiry is the available link passage of current IP allocation of packets, this querying method can be realized by the function that calls load sharing computing module 220 and provide, then according to the result of load sharing computing module 220 output from the intrafascicular link that sends the current IP bag of choosing of MP, the IP packet is sent to the module 400 of respective link, carry out transfer of data according to traditional mode by this module 400.
As seen, a variety of situations can be arranged, Fig. 2 and just wherein two kinds of specific implementations embodiment illustrated in fig. 4 about the way of realization of multilink transmission itself.
Embodiment three:
Referring to Fig. 5, Fig. 5 is the structural representation of a kind of the 3rd preferred embodiment of the device of realizing multi-link load share among the present invention.This embodiment for the transmission course of complete description to signaling, has increased a TCP/SCTP processing module 500 on basis embodiment illustrated in fig. 2, be used to realize data re-transmission, to be received with guaranteeing data integrity.When sending data, TCP/SCTP protocol process module 500 will need the packing data to the opposite end transmission, send to corresponding IP Process Module 240; When receiving data, the data that 500 pairs of IP Process Module that receive of TCP/SCTP protocol process module send over are resolved, and check whether Receive sequence number wherein is normal.In case finding has data-bag lost, then sends the data re-transmission request to the other side, up to receiving complete data; Abandon repeat requests after perhaps retransmitting the preassigned number of times of failure, then send alarm by Fault Management System.
Wherein, also can comprise signaling data in the data of transmission.Owing to increased TCP/SCTP processing module 500 in this device, had retransmission mechanism, thereby this device can be used for command transmitting.In addition, in order to realize the transmission of signaling, the TCP/SCTP processing module 500 in this device also can be other the processing module with retransmission mechanism.
Be described further below in conjunction with the method for specific embodiment the load sharing of realization multilink transmission of the present invention.
Embodiment four:
Referring to Fig. 6, Fig. 6 is for realizing the overview flow chart of the method embodiment of the load sharing that multilink transmits among the present invention.This flow process comprises the steps:
In this step, the quality condition of each link is in the dynamic monitoring always.Wherein, can carry out dynamic monitoring to each bar ppp link by fault detection module, and monitoring result is carried out dynamic memory, be used for step 602 and use.
Wherein, link quality information comprises: in link failure information, link congestion information, link-recovery information and the link normal information one or more than one combination in any.
Step 602 is carried out load sharing processing according to link quality information to each bar link.
In this step, according to the link quality information of each bar link of dynamic monitoring in the step 602, choose current available link and distribute, promptly choosing quality information is that link link normal or link-recovery distributes for the current IP packet as available link.
So far, the flow process of present embodiment finishes.Wherein, the specific implementation method that step 602 is carried out load sharing processing according to link quality information to each bar link has a variety of, enumerates a kind of preferable implementation method below.
Referring to Fig. 7, Fig. 7 is the detail flowchart of middle step 602 embodiment illustrated in fig. 6.This flow process comprises the steps:
Step 702 is the IP call number with this IP address transition.
In the present embodiment, in order to save the space and to improve computational speed, being provided with the IP address transition is the step of IP call number, in this step, be IP call number of each IP address setting, behind input IP address, according to the corresponding relation that is provided with, be the IP call number with the IP address transition.In the practical application,, can omit this step 702 if do not consider the problem of space and speed.
Step 703 reads user's Control Parameter of configuration.
In this step, the user configures user's Control Parameter, and these user's Control Parameter can comprise: chain travel permit number, get 8 in the present embodiment; Link initial offset, purpose are always to choose the less link of call number during for fear of original allocation, so beginning just increases a side-play amount, default value is 0, and promptly non-migration gets 1 in the present embodiment; Link assignment at interval, purpose is for avoiding adjacent IP layer user too close on link selection, an assignment interval is set, default value is 1, gets 2 in the present embodiment; And the link selection step-length, the user can jump to next bar available link according to this step value, gets 3 etc. in the present embodiment.
In the practical application, can need not to be provided with user's Control Parameter, also can only be provided with in above-mentioned user's Control Parameter parameter or more than one parameter.
Step 704 judges whether these user's Control Parameter are effective.If then execution in step 705 effectively; Otherwise this process stops, and the prompting mistake.
In this step, the judgement of validity can be according to data type, span etc.
Step 705 is carried out load sharing and is calculated, and chooses initial link circuit.
In this step, can earlier the link initial offset be composed to the link initial value, the link channel call number of distributing for each IP call number is then: the IP call number multiply by link assignment adds the link initial value at interval.Be 1 then according to link initial offset in the user's Control Parameter that reads, link assignment is spaced apart 3, then for the IP call number is that the link channel of 1 allocation of packets is that call number is 4 link channel, promptly link initial value 1 adds that IP call number 1 multiply by link assignment and 2 obtains 3 at interval; Then for the IP call number is that the link channel of 2 allocation of packets is that call number is 5 link channel, promptly link initial value 1 adds that IP call number 2 multiply by link assignment at interval 2, obtains 5.In addition, in this step, comprise also if the link index numbers that link index numbers allows greater than maximum that then it being carried out mould, to remove the chain travel permit several 8, make it come back to legal interval, promptly this link index numbers is reusable.For instance,,, then its mould is removed 8 and obtain 2, be its distribute links call number and be 2 path greater than largest index numbers 8 if the link index numbers of selecting is 10.
Perhaps in this step, suppose that the link initial value is 1, the link channel call number of distributing for each IP call number is then: link initial value+link initial offset+(IP call number-1) * link assignment at interval.Be 1 promptly according to link initial offset in the user's Control Parameter that reads, link assignment is spaced apart 3, be that the link channel of 1 allocation of packets is that call number is 2 link channel then for the IP call number, be link initial value 1+ link initial offset 1+ (IP call number 1-1) * link assignment interval 2, obtain 2; Then for the IP call number is that the link channel of 2 allocation of packets is that call number is 4 link channel, promptly link initial value 1+ link initial offset 1+ (IP call number 2-1) * link assignment at interval 2, obtain 4.
Therefore as seen, concrete computational methods have a variety ofly in the practical application, and which kind of computational methods above-mentioned wherein two kinds of just having enumerated adopt and be not intended to limit the scope of the invention.
Wherein,, then in this step, can be defaulted as link assignment and be spaced apart 1, and initial offset is 0, calculate then if link assignment is not set at interval and the link initial offset in the user's Control Parameter described in the step 703.
Step 706 reads the link quality information of this link.
In this step, read the monitoring result of the step 601 in the flow process shown in Figure 6, i.e. the link quality information of each link.
Step 707 is judged this link-quality situation, if the link-quality situation is normal, does not promptly have fault or congested, then link index numbers is converted to the output of chain line, and by carrying out transfer of data with this chain line link corresponding for the current IP packet; Otherwise execution in step 708.
Step 708 is chosen next bar available link.Return execution in step 706 afterwards, up to selecting the available link passage.
In this step, if be that link index numbers that 1 packet is chosen is that 3 link channel breaks down for the IP call number in the step 705, then this link index numbers 3 is added a link selection step-length, according to link selection step-length in the user's Control Parameter that reads is 3, then next bar available link of choosing for this IP packet is that call number is 6 link channel, return execution in step 706 afterwards, read link index numbers and be the state of 6 link, judge whether normally, normally then link index numbers is converted to the output of chain line; If it is undesired, it is added a link selection step-length 3, the selection link index numbers is 9 link channel, in this step, also comprise if the link index numbers that link index numbers allows greater than maximum is then carried out mould to it and removed the processing of chain travel permit number, make it come back to legal interval, promptly because 9 several 8, obtain 1 after then its mould being removed 8 greater than the chain travel permit in the present embodiment.Return execution in step 706 afterwards, read link index numbers and be the state of 1 link, judge whether normal, and the like, up to selecting an available link, output link identifies, and by carrying out transfer of data with this chain line link corresponding for the current IP packet.
In the practical application, if in the user's Control Parameter described in the step 703 the link selection step-length is not set, then in this step, can be defaulted as link link selection step-length is 1, calculates then.
So far, this flow process finishes.
The various embodiments described above are a kind of displaying of inventive concept; be not to be used to limit protection scope of the present invention; be understandable that in addition; the present invention is not subjected to the restriction of the way of realization of multilink transmission itself about the device and method of realizing the load sharing that multilink transmits; within the spirit and principles in the present invention all; any modification of being made, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (17)
1, a kind of device of realizing the load sharing of multilink transmission comprises: many point-to-point ppp links, it is characterized in that this device also comprises fault detection module and load sharing scheduler module, wherein,
Fault detection module is used for each bar ppp link quality of dynamic monitoring, and link quality information is reported to the load sharing scheduler module;
The load sharing scheduler module is used for according to the link quality information by the fault detection module report each link being carried out load sharing processing.
2, device as claimed in claim 1 is characterized in that, described fault detection module is arranged in the described load sharing scheduler module.
3, device as claimed in claim 1 or 2 is characterized in that, described load sharing scheduler module comprises: load sharing computing module and point-to-point MultiLink PPP MP processing module, wherein,
The load sharing computing module is used for according to the link quality information by the fault detection module report, each link is carried out load sharing calculate, and result of calculation is exported to PPP MP processing module;
PPP MP processing module is used to assemble many ppp links and forms single high-band wide beam MP bundle; PPP MP processing module is carried out the load sharing result calculated from the intrafascicular link of choosing transmission current IP packet of MP according to the output of load sharing computing module to link, thereby realizes the load balancing between each ppp link.
4, device as claimed in claim 3 is characterized in that, described load sharing computing module comprises: computing module, and user's Control Parameter memory module and link quality information memory module, wherein,
User's Control Parameter memory module, what be used to store user's input calculates relevant user's Control Parameter with load sharing;
The link quality information memory module is used for the link quality information that the dynamic memory fault detection module is sent;
Computing module, be used for reading the IP address of current IP packet from PPP MP processing module, and read user's Control Parameter in user's Control Parameter memory module, then according to the link quality information in the link quality information memory module, each link is carried out load sharing to be calculated, for the current IP packet calculates current available link channel number, and export to PPP MP processing module.
5, device as claimed in claim 4 is characterized in that, described load sharing computing module further comprises: IP address conversion module and chain line modular converter, wherein,
The IP address conversion module is used for IP address transition with the described current IP packet that reads from PPP MP processing module and becomes IP allocation index number, exports to computing module;
Computing module receives this IP allocation index number, and in conjunction with the user's Control Parameter that has read, the link channel call number available for the current IP packet calculates according to link quality information exported to chain line modular converter;
Chain line modular converter is used for converting the link channel call number of computing module output to the chain line, exports to PPP MP processing module.
6, device as claimed in claim 1 or 2 is characterized in that, every ppp link of described many ppp links comprises: LCP LCP processing module and Network Control Protocol NCP processing module; Perhaps every of described many ppp links ppp link includes only the LCP processing module, and many shared NCP processing modules of ppp link.
7, device as claimed in claim 6 is characterized in that, the shared NCP processing module of described many ppp links is arranged in the described PPP MP processing module.
8, a kind of method that realizes the load sharing of multilink transmission is characterized in that this method comprises the steps:
A, each bar ppp link quality of dynamic monitoring, and generate link quality information;
B, each bar link is carried out load sharing processing according to link quality information.
9, method as claimed in claim 8 is characterized in that, link quality information comprises described in the step B: in link failure information, link congestion information, link-recovery information and the link normal information one or more than one combination in any; The then described load sharing processing of carrying out is: according to the above-mentioned quality information of each bar link, choosing current quality information is that link link normal or link-recovery distributes for the current IP packet as available link.
10, method as claimed in claim 8 is characterized in that, described step B takes a step forward and comprises carrying out load sharing processing: read user's Control Parameter; Then describedly further comprise when carrying out load sharing processing: carry out load sharing according to user's Control Parameter and calculate.
11, method as claimed in claim 10 is characterized in that, described user's Control Parameter comprises: the link selection step-length; Then described when carrying out load sharing processing, if be the current link channel fault of IP allocation of packets or when congested, be next bar available link passage of current IP allocation of packets then according to this link selection step-length.
12, method as claimed in claim 11 is characterized in that, described user's Control Parameter also comprises: the link initial offset; Then described when carrying out load sharing processing, during for current IP allocation of packets available link, select initial link circuit according to this link initial offset.
As claim 10 or 11 described methods, it is characterized in that 13, described user's Control Parameter also comprises: link assignment at interval; Then described when carrying out load sharing processing, during for different IP allocation of packets available link, select different initial link circuits at interval according to link assignment.
14, method as claimed in claim 8 is characterized in that, further comprises among the described step B: the purpose IP address of input IP packet; Then described when carrying out load sharing processing, according to this IP address each link is carried out load sharing and calculate, be the current available link of this IP allocation of packets.
15, method as claimed in claim 14 is characterized in that, further comprise behind the input IP address among the described step B: with the IP address transition is the IP call number; Then described when carrying out load sharing processing, according to this IP call number each link is carried out load sharing and calculate, be the current available link of the IP allocation of packets of this IP address.
16, as claim 8,14 or 15 described methods, it is characterized in that, when described step B carries out load sharing processing, for the current IP packet calculates available link index numbers earlier, again link index numbers is converted to the chain line afterwards, according to chain line distribute links passage.
17, method as claimed in claim 16, it is characterized in that, in this method, set in advance chain travel permit number, and described link index numbers recycles, if the link index numbers of distributing surpasses the maximum link call number, then this link index numbers is carried out mould and remove chain travel permit number, make described link index numbers come back to legal interval.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006101035696A CN100433729C (en) | 2006-07-25 | 2006-07-25 | Load sharing apparatus and method for realizing multi-chain circuit transmission |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006101035696A CN100433729C (en) | 2006-07-25 | 2006-07-25 | Load sharing apparatus and method for realizing multi-chain circuit transmission |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1889527A true CN1889527A (en) | 2007-01-03 |
| CN100433729C CN100433729C (en) | 2008-11-12 |
Family
ID=37578786
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2006101035696A Expired - Fee Related CN100433729C (en) | 2006-07-25 | 2006-07-25 | Load sharing apparatus and method for realizing multi-chain circuit transmission |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100433729C (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008116399A1 (en) * | 2007-03-23 | 2008-10-02 | Huawei Technologies Co., Ltd. | Dynamically adjusting method and apparatus for link state and bundled link state |
| CN101795186A (en) * | 2010-03-04 | 2010-08-04 | 华为技术有限公司 | Linkage disconnecting method, system and equipment |
| CN101841480A (en) * | 2010-04-26 | 2010-09-22 | 华为技术有限公司 | Multilink message sending method, equipment and system |
| WO2010145578A1 (en) * | 2009-10-13 | 2010-12-23 | 中兴通讯股份有限公司 | Method and device for transmitting service data in packet switch domain |
| CN102098301A (en) * | 2011-01-06 | 2011-06-15 | 复旦大学 | Multilink self-adaptation data transmission method and system |
| CN102394878A (en) * | 2011-10-28 | 2012-03-28 | 杭州华三通信技术有限公司 | Cross board point-to-point multilink protocol binding method and device thereof |
| CN102696200A (en) * | 2009-07-17 | 2012-09-26 | 荷兰皇家Kpn电信集团 | Congestion control in a telecommunications network |
| CN102845022A (en) * | 2011-04-20 | 2012-12-26 | 华为技术有限公司 | Transmission quality detection method, apparatus and transmission system thereof |
| CN106850455A (en) * | 2016-12-27 | 2017-06-13 | 河北远东通信系统工程有限公司 | A kind of system of wireless multilink aggregated bandwidth |
| CN109862556A (en) * | 2019-02-12 | 2019-06-07 | Oppo广东移动通信有限公司 | Packet distribution method, device, mobile terminal and storage medium |
| CN113676405A (en) * | 2021-08-18 | 2021-11-19 | 上海晨驭信息科技有限公司 | Load sharing-based rapid link master-slave switching distributed system and method |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1748617A3 (en) * | 1998-09-11 | 2011-02-09 | Hitachi, Ltd. | IP packet communication apparatus |
| JP2003219027A (en) * | 2002-01-23 | 2003-07-31 | Matsushita Electric Ind Co Ltd | Internet phone system and method for generating bypass thereof |
| CN100388803C (en) * | 2003-03-04 | 2008-05-14 | 中兴通讯股份有限公司 | Short message center system with double-network double-plane structure |
-
2006
- 2006-07-25 CN CNB2006101035696A patent/CN100433729C/en not_active Expired - Fee Related
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101051957B (en) * | 2007-03-23 | 2010-11-10 | 华为技术有限公司 | Dynamically regulating method and device for link state and bundled link state |
| WO2008116399A1 (en) * | 2007-03-23 | 2008-10-02 | Huawei Technologies Co., Ltd. | Dynamically adjusting method and apparatus for link state and bundled link state |
| CN102696200B (en) * | 2009-07-17 | 2015-11-25 | 荷兰皇家Kpn电信集团 | Congestion control in communication network |
| CN102696200A (en) * | 2009-07-17 | 2012-09-26 | 荷兰皇家Kpn电信集团 | Congestion control in a telecommunications network |
| WO2010145578A1 (en) * | 2009-10-13 | 2010-12-23 | 中兴通讯股份有限公司 | Method and device for transmitting service data in packet switch domain |
| US8787370B2 (en) | 2009-10-13 | 2014-07-22 | Zte Corporation | Method and apparatus for transmitting packet switch domain service data |
| CN101795186B (en) * | 2010-03-04 | 2013-01-23 | 华为技术有限公司 | Linkage disconnecting method, system and equipment |
| CN101795186A (en) * | 2010-03-04 | 2010-08-04 | 华为技术有限公司 | Linkage disconnecting method, system and equipment |
| CN101841480A (en) * | 2010-04-26 | 2010-09-22 | 华为技术有限公司 | Multilink message sending method, equipment and system |
| CN102098301A (en) * | 2011-01-06 | 2011-06-15 | 复旦大学 | Multilink self-adaptation data transmission method and system |
| CN102098301B (en) * | 2011-01-06 | 2015-07-29 | 复旦大学 | The adaptive data transmission method of multilink and system |
| CN102845022A (en) * | 2011-04-20 | 2012-12-26 | 华为技术有限公司 | Transmission quality detection method, apparatus and transmission system thereof |
| WO2013060193A1 (en) * | 2011-10-28 | 2013-05-02 | Hangzhou H3C Technologies Co., Ltd. | Method for bundling cross-board multilink protocol |
| GB2506316A (en) * | 2011-10-28 | 2014-03-26 | Hangzhou H3C Tech Co Ltd | Method for bundling cross-board multilink protocol |
| CN102394878B (en) * | 2011-10-28 | 2015-02-18 | 杭州华三通信技术有限公司 | Cross board point-to-point multilink protocol binding method and device thereof |
| CN102394878A (en) * | 2011-10-28 | 2012-03-28 | 杭州华三通信技术有限公司 | Cross board point-to-point multilink protocol binding method and device thereof |
| CN106850455A (en) * | 2016-12-27 | 2017-06-13 | 河北远东通信系统工程有限公司 | A kind of system of wireless multilink aggregated bandwidth |
| CN106850455B (en) * | 2016-12-27 | 2020-01-03 | 河北远东通信系统工程有限公司 | Wireless multilink bandwidth aggregation system |
| CN109862556A (en) * | 2019-02-12 | 2019-06-07 | Oppo广东移动通信有限公司 | Packet distribution method, device, mobile terminal and storage medium |
| CN113676405A (en) * | 2021-08-18 | 2021-11-19 | 上海晨驭信息科技有限公司 | Load sharing-based rapid link master-slave switching distributed system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100433729C (en) | 2008-11-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1889527A (en) | Load sharing apparatus and method for realizing multi-chain circuit transmission | |
| CN1208935C (en) | Processing data packets | |
| CN1929472A (en) | Method, system, signal and medium for managing data transmission in a data network | |
| CN101047618A (en) | Method and system for acquiring network route information | |
| CN101064625A (en) | Method for managing user side equipment through network address translation gateway | |
| CN1921457A (en) | Network equipment and message transferring method based on multiple-core processor | |
| JP2004236332A (en) | Identification of packet data flow for multiplexing | |
| CN101030946A (en) | Method and system for realizing data service | |
| CN1431805A (en) | Data pack exchanging system, method, route appts. data pack structure and generation method | |
| CN1960338A (en) | Communication controlling method | |
| CN101060493A (en) | A method of private network user access the server in a private network through domain name | |
| CN101039224A (en) | Method and equipment for detecting periodically link state | |
| CN1878130A (en) | Data forwarding method | |
| CN101039273A (en) | Communication equipment, transmission control method and transmission control program | |
| CN1773993A (en) | Session relay equipment and session relay method | |
| CN100346601C (en) | Access server with function of collecting communication statistics information | |
| CN1874319A (en) | Admission control system and admission control method in load supporting interface of communication network | |
| CN101075924A (en) | Method for accessing server by customer end | |
| CN1612501A (en) | Transmitting data using multi-frames | |
| CN1571418A (en) | A method and system for implementing data transmission in flow control transmission protocol | |
| CN1643858A (en) | Quality of service request correlation | |
| CN1890928A (en) | Network, router device, route updating suppression method used for the same, and program thereof | |
| CN1848798A (en) | IP data stream guiding system and method based on business trigging | |
| CN1893504A (en) | Network interface apparatus for controlling to forward image data according to enable service setup every address and method thereof | |
| CN1665239A (en) | A method for implementing information transmission between equipments of different communication protocols |
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 | ||
| ASS | Succession or assignment of patent right |
Owner name: CHANGZHOU XIAOGUO INFORMATION SERVICE CO., LTD. Free format text: FORMER OWNER: HUAWEI TECHNOLOGY CO., LTD. Effective date: 20140313 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| COR | Change of bibliographic data |
Free format text: CORRECT: ADDRESS; FROM: 518129 SHENZHEN, GUANGDONG PROVINCE TO: 213164 CHANGZHOU, JIANGSU PROVINCE |
|
| TR01 | Transfer of patent right |
Effective date of registration: 20140313 Address after: 213164 building C, building 407-2-6, Tian An Digital City, 588 Chang Wu Road, Wujin hi tech Industrial Development Zone, Changzhou, Jiangsu, China Patentee after: Changzhou Xiaoguo Information Service Co., Ltd. Address before: 518129 Bantian HUAWEI headquarters office building, Longgang District, Guangdong, Shenzhen Patentee before: Huawei Technologies Co., Ltd. |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20081112 Termination date: 20160725 |