CN100563238C - A kind of transmission method of message - Google Patents
A kind of transmission method of message Download PDFInfo
- Publication number
- CN100563238C CN100563238C CNB2006100073972A CN200610007397A CN100563238C CN 100563238 C CN100563238 C CN 100563238C CN B2006100073972 A CNB2006100073972 A CN B2006100073972A CN 200610007397 A CN200610007397 A CN 200610007397A CN 100563238 C CN100563238 C CN 100563238C
- Authority
- CN
- China
- Prior art keywords
- message
- link
- formation
- interface
- buffer memory
- 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 - Fee Related
Links
Abstract
The invention discloses a kind of transmission method of message, be applied on the binding multiple linkage interface, this method comprises: the number to message in the message queue of transmitting terminal buffer memory detects in real time; The number of message was positioned at the message of first position greater than 0 o'clock in the formation by link L1 transmission current cache message in the formation that detects the buffer memory message; The number of message is when seeing bottom right formula (1) in the formation that detects the buffer memory message, sends by link Li to be positioned at the message of seeing the individual position of bottom right formula (2) in the formation of current cache message.The present invention can guarantee that the message of transmitting terminal arrives receiving terminal according to correct order in the transmission bandwidth that increases multilink interface, avoid the problem of receiving terminal packet out-ordering.
Description
Technical field
The present invention relates to the multilink bundle technology, particularly relate to a kind of have in binding send method of message on the multiple linkage interface.
Background technology
The multilink bundle technology is that a plurality of physical links are tied up, and makes to comprise many physical links in the multilink interface, thereby reaches the purpose that increases bandwidth.Present multilink bundle technology mainly comprises multilink frame relay (MFR) technology and multilink point-to-point (NP) technology, when adopting MFR technology or NP technology to increase transmission bandwidth, prior art sends message on MFR interface or NP interface mode mainly comprises:
Mode one, ping-pong.
When adopting ping-pong, transmit leg uses each the bar physical link in MFR interface or the NP interface to send message in turn, that is to say, if this uses N bar physical link to send message, then use next time N+1 bar physical link to send message, the rest may be inferred, when the last item physical link is finished the message transmission, begins to carry out sending again from article one physical link.
Mode two, token send mode.
When the token send mode of employing mode two, be token value for each the bar physical link in MFR interface or the NP interface is provided with the byte number that allows to send in advance, when a physical link in MFR interface or the NP interface is carried out transmission, message of every transmission just deducts the byte length of the literary composition of transmitting messages from the token value of this physical link, token value up to this physical link is less than or equal at 0 o'clock, then carries out transmission by next bar physical link.
Need to prove, no matter adopting aforesaid way one still is mode two, in the prior art, when transmitting terminal takes out message and sends by a physical link in MFR interface or the NP interface from self formation, all send according to the principle that sends earlier first, that is to say that transmitting terminal takes out message successively according to the order of message in formation and sends.
Yet, owing to tending to exist bigger speed difference between many physical links in a MFR interface or the NP interface, such as, comprise 1 T3 link in the MFR interface, article 1, T1 link and 1 E1 link, the speed of T3 link can reach tens usually, such as 44M, and the speed of T1 link and E1 link has only several million, such as being respectively 2M and 1.5M, like this, when the principle that sends earlier first according to prior art and the ping-pong of employing mode one, though can guarantee to increase the transmission bandwidth of MFR interface or NP interface, but when being sent by rapid link, the message of sending out in back on the rapid physical link tends to arrive receiving terminal earlier than the message of sending out earlier before on physical link at a slow speed, thereby causes occurring at receiving terminal the problem of a large amount of packet out-orderings; When the principle that sends earlier first according to prior art and the token send mode of employing mode two, though also can guarantee to increase the transmission bandwidth of MFR interface or NP interface, but the token value on physical link at a slow speed is less than or equal to 0 and when switching on the rapid physical link, the message of sending out in back on the rapid physical link also tends to arrive receiving terminal earlier than the message of sending out earlier before on physical link at a slow speed, thereby causes occurring at receiving terminal the problem of a large amount of packet out-orderings.
By above description as can be seen, owing to often there being bigger speed difference between many physical links in a MFR interface or the NP interface, though and the mode that principle that prior art sends earlier first and the mode that alternately sends and token send has guaranteed to increase the transmission bandwidth of MFR interface or NP interface, but but can't remedy the speed difference between each bar physical link, can't avoid receiving terminal the problem of a large amount of packet out-orderings to occur, thereby bring great inconvenience for the subsequent packet work of treatment of receiving terminal.Such as, receiving terminal must carry out buffered to a large amount of out of order messages, and when the quantity of out of order message exceeded the length in rx-side buffering district, receiving terminal then can be with out of order packet loss, thereby greatly reduced the QoS of network when adopting MFR technology or NP technology.
Summary of the invention
In view of this, main purpose of the present invention is to provide a kind of transmission method of message, so that farthest avoid the problem of receiving terminal packet out-ordering when guaranteeing to increase transmission bandwidth.
In order to achieve the above object, technical scheme of the present invention is achieved in that
A kind of transmission method of message be applied on the binding multiple linkage interface, and message length is fixed;
This method sorts to each the bar link in the multilink interface according to speed order from big to small;
For the fastest link L1 of multilink interface medium velocity, the entry condition that this fastest link L1 is set is that in the formation of buffer memory message, the number of message is greater than 0;
For any link Li except that fastest link L1 in the multilink interface, i is the natural number greater than 1, and the entry condition of this Li is: in the formation of buffer memory message, the number of message greater than
Wherein, Vk, Vi are respectively the speed of k bar and i bar link in the multilink interface;
This method also comprises:
A, the number of message in the message queue of transmitting terminal buffer memory is detected in real time;
B, the number of message sent the message that is positioned at first position in the formation of current cache message greater than 0 o'clock by link L1 in the formation that detects the buffer memory message; In the formation that detects the buffer memory message number of message greater than
The time, send by link Li and be positioned at the in the formation of current cache message
The message of individual position;
When described
Value when being decimal, determined
The message of individual position is
Value round up the back position message.
Wherein, described multilink interface is multilink frame relay MFR interface or multilink point-to-point interface.
The another kind of method of message that sends provided by the invention be applied on the binding multiple linkage interface, and message length is fixed;
This method sorts to each the bar link in the multilink interface according to speed order from big to small;
For the fastest link L1 of multilink interface medium velocity, the entry condition that this fastest link L1 is set is that in the formation of buffer memory message, the total length of each message is greater than 0;
For any link Li except that fastest link L1 in the multilink interface, i is the natural number greater than 1, and the entry condition of this Li is: in the formation of buffer memory message, the total length of each message greater than
Wherein, Vk, Vi are respectively the speed of k bar and i bar link in the multilink interface, and S is the length of each message;
This method also comprises:
A, the total length of each message in the message queue of transmitting terminal buffer memory is detected in real time;
B, the total length of each message sent the message that is positioned at first position in the formation of current cache message greater than 0 o'clock by link L1 in the formation that detects the buffer memory message; In the formation that detects the buffer memory message total length of each message greater than
The time, send by link Li and be positioned at the in the formation of current cache message
The message of individual position;
When described
Value when being decimal, determined
The message of individual position is
Value round up the back position message.
Wherein, described multilink interface is multilink frame relay MFR interface or multilink point-to-point interface.
Another transmission method of message provided by the invention is applied on the binding multiple linkage interface;
This method sorts to each the bar link in the multilink interface according to speed order from big to small;
For the fastest link L1 of multilink interface medium velocity, the entry condition that this fastest link L1 is set is that in the formation of buffer memory message, the total length of each message is greater than 0;
For any link Li except that fastest link L1 in the multilink interface, i is the natural number greater than 1, and the entry condition that this Li is set is: in the formation of buffer memory message, exist message arrangement position X to satisfy inequality
Wherein, Vk, Vi are respectively the speed of k bar and i bar link in the multilink interface, S
kAnd S
XThe message length of representing k position and X position in the formation of current cache message respectively;
This method also comprises:
A, the total length of each message in the message queue of transmitting terminal buffer memory is detected in real time;
B, the total length of each message sent the message that is positioned at first position in the formation of current cache message greater than 0 o'clock by link L1 in the formation that detects the buffer memory message; In the formation that detects the buffer memory message, exist message arrangement position X to satisfy inequality
The time, get the minimum value X ' of X, send the message that is positioned at the individual position of X ' in the formation of current cache message by link Li.
When detecting for i is the different links of different value, and when having identical message arrangement position X ', the link Li of the individual position of described transmission X ' message is the fastest link of this difference link medium velocity.
When transmitting terminal message length not fixedly the time, this method further comprises: the length that has a message when transmitting terminal is during greater than the burst length threshold that sets in advance, when this message is put into the formation of buffer memory message, this message from first to last is divided into equal-sized a plurality of message, and a plurality of messages that will tell are then put into the formation of buffer memory message successively according to cephalocaudal order.
Described multilink interface is MFR interface or multilink point-to-point interface.
This shows, the present invention has made full use of the speed difference between each bar link in the multilink interface, determine to start the opportunity of each bar link according to the speed of each bar link, and determine the position of message in the buffer memory message queue that each bar link is sent according to the speed of each bar link, like this, the message that then can guarantee transmitting terminal arrives receiving terminal according to correct order, thereby has avoided the problem of receiving terminal packet out-ordering.In addition, it is the purposes that increase bandwidth in order to reach that the multilink bundle technology ties up many physical links, that is to say, be for the transmission speed that makes a multilink interface reach as much as possible each bar physical link in this multilink interface speed and, the present invention is when starting each bar link transmission message, elder generation's operating speed link faster sends message, when speed can't satisfy transmission demand than faster link, link that the ability toggle speed is slower and speed link faster send message simultaneously, therefore, when guaranteeing that message arrives receiving terminal with correct order, farthest increased the transmission bandwidth of multilink interface.
Description of drawings
Fig. 1 is a flow chart of realizing embodiments of the invention.
Embodiment
When on multilink interface such as MFR interface or NP interface, sending message, the mode that the principle that prior art just sends earlier according to message simply first adopts alternately transmission or token to send sends, and do not take into full account and utilize the speed difference between each bar physical link in the multilink interface, thereby tend to cause receiving terminal the problem of a large amount of packet out-orderings to occur.The process that multilink interface is sent message is analyzed as can be known, remedy the speed difference between each bar physical link in the multilink interface, to guarantee that message arrives receiving terminal according to the order that it reaches transmitting terminal, then should be by the message of position in the slower link transmit queue of multilink interface medium velocity after, simultaneously by the speed more preceding message in position in the link transmit queue faster, because it is longer that slow link sends the required time of message, speed faster link to send the required time of message shorter, thereby can guarantee that the more preceding message in position can arrive receiving terminal earlier by the transmission of rapid link, the message of position after arrives after arriving receiving terminal by the more preceding message in position that is sent in of slow speed link again, thereby guaranteed the correctness of the received message sequence of receiving terminal.Wherein, which the position message after slow speed link is specifically got in the formation then needs to determine according to the speed of rapid link and slow speed link.
At These characteristics, the present invention proposes a kind of transmission method of message, its core concept is: the entry condition that each bar link in the multilink interface is set; Formation to transmitting terminal buffer memory message detects in real time; When detecting the entry condition that satisfies any link in the multilink interface, determine that the current pairing message of link that satisfies entry condition obtains the position, and send by the current link that satisfies entry condition be positioned in the formation of current cache message definite message obtain the message of position.
Wherein, it is quantity or length according to message in the formation of buffer memory message that the entry condition that is detected and pairing message of link obtain the position, and the speed of each bar link is determined.And described multilink interface can be any one interface that binding has multilink, such as, MFR interface or NP interface.
For making the purpose, technical solutions and advantages of the present invention clearer, the present invention is described in further detail below in conjunction with drawings and the specific embodiments.
In following embodiment, be that example illustrates implementation procedure of the present invention as described multilink interface with the MFR interface.
Fig. 1 is a flow chart of realizing embodiments of the invention.Referring to Fig. 1, the present invention is implemented in the process that sends message on the MFR interface and may further comprise the steps:
Step 101:, each the bar link in the MFR interface is sorted according to speed order from big to small at transmitting terminal.
Here, such as, a MFR interface has been bound n bar link L1 to Ln, and its speed is respectively V
1To V
n, and V
1＞...＞V
n, wherein, n is the natural number more than or equal to 2, and like this, in this step, to after L1 to the Ln ordering, obtaining article one link is the L1 link according to speed order from big to small, and the second link is the L2 link, and the rest may be inferred, is the Ln link until n bar link.
Step 102:, the entry condition of each bar link in the MFR interface is set at transmitting terminal.
Here, send the time that message spent, therefore, in this step, whether the entry condition of each bar link in the MFR interface must fixedly be set according to the length of transmitting terminal message because the length of message can influence a link.
One, fixes when message length, that is to say, when each need send the identical length etc. of message, in this step, can be according to the quantity of message in the formation of transmitting terminal buffer memory message or the length of message, and, the entry condition of each bar link in the multilink Frame Relay Interface is set in conjunction with the speed of each link in the multilink Frame Relay Interface, specifically comprise:
A) for article one link L1 after the ordering, because its speed maximum, therefore in order to guarantee to increase transmission bandwidth, the entry condition that this speed maximum link L1 is set is: in the formation of buffer memory message, the number of message greater than 0 or the total length of each message greater than 0, that is to say that as long as have message in the formation of buffer memory message, the link L1 that then starts MFR interface medium velocity maximum sends message.
B) for the second link L2 after the ordering, because its speed is littler but bigger than other link than L1, therefore, can't satisfy in the formation that transmission demand is the buffer memory message message at L1 constantly accumulates, guaranteeing that message arrives under the not out of order condition of receiving terminal, then should start L2 when only using L1 not use the transmission speed of L1 and L2 fast simultaneously.And the entry condition that this link L2 is set is: in the formation of buffer memory message, the number of message is greater than (V
1/ V
2) or the total length of each message greater than (V
1/ V
2) * S, wherein, S is the regular length of message.
Such as, the speed V of L1
1Be 44M, the speed V of L2
2Be 2M, because V
1/ V
2=22, that is to say that 22 times that message spent of link L1 transmission equate with link L2 1 time that message spent of transmission, therefore, have (V in the formation of buffer memory message
1/ V
2+ 1)=23 during message, if preceding 22 messages in the formation of use L1 link transmission buffer memory message, the 23rd message in the formation of use L2 link transmission buffer memory message, like this, after the L1 link sends preceding 21 messages, the 22nd message that the L1 link sends will almost arrive receiving terminal simultaneously with the 23rd message that sends by the L2 link, thereby guarantee to have guaranteed when improving transmission speed promptly increases transmission bandwidth that the message sequence that receiving terminal receives is correct.Therefore, as can be known, for the order that guarantees the message that receiving terminal receives is correct, the entry condition of L2 is: in the formation of buffer memory message, the number of message is greater than (V
1/ V
2).A kind of flexible entry condition according to this entry condition is: in the formation of buffer memory message, the total length of message is greater than (V
1/ V
2) * S.
C) for the 3rd link L3 after the ordering, because its speed is littler but bigger than other link than L1 and L2, at L1 and L2 message transmission but still can't satisfy in the formation that transmission demand is the buffer memory message accumulation message time simultaneously, owing to guaranteeing that message arrives under the not out of order condition of receiving terminal, only use L1 and L2 not to have the transmission speed of simultaneously L1, L2 and L3 fast, therefore should start L3.And the entry condition that this link L3 is set is: in the formation of buffer memory message, the number of message is greater than (V
1+ V
2)/V
3Or the total length of each message is greater than (V
1+ V
2)/V
3* S, wherein, S is the regular length of message.
Such as, the speed V of L1
1Be 44M, the speed V of L2
2Be 2M, the speed V of L3
3Be 1.5M,, therefore, should use the speed of L1 and L2 and compare with the speed of L3 because when starting L3, L1 and L2 have all started and sent message.Because (V
1+ V
2)/V
3=30.6, that is to say that 30.6 times that message spent of the common transmission of link L1 and L2 equate with link L3 1 time that message spent of transmission, therefore, have (V in the formation of buffer memory message
1+ V
2)/V
3Add 1 after rounding up promptly during 32 messages, if preceding 30.6 messages in the formation of use L1 and L2 link transmission buffer memory message, consider the integrality that sends message, in the present invention, when speed ratio is decimal, can round up to this fractional value, promptly get smallest natural number greater than this fractional value, 30.6 be 31 after rounding up, so the next message that uses the L3 link to send in the buffer memory message queue is the 32nd message, like this, after L1 and L2 send preceding 30 messages simultaneously, the 31st message that is sent will almost arrive receiving terminal simultaneously with the 32nd message that sends by the L3 link, thereby guarantee to have guaranteed that the order of the message that receiving terminal receives is correct when improving transmission speed promptly increases transmission bandwidth.Therefore, as can be known, for the order that guarantees the message that receiving terminal receives is correct, the entry condition of L3 is: in the formation of buffer memory message, the number of message is greater than (V
1+ V
2)/V
3A kind of flexible entry condition according to this entry condition is: in the formation of buffer memory message, the total length of message is greater than (V
1+ V
2)/V
3* S.
The rest may be inferred, can obtain for the last item link Ln after the ordering, because its speed minimum, therefore, at L1 to Ln-1 message transmission but still can't satisfy and start Ln still can have message in the formation that transmission demand is the buffer memory message time simultaneously, therefore, the entry condition that the link Ln of this speed minimum is set is: in the formation of buffer memory message, the number of message is greater than (V
1+ V
2+ ... V
N-1)/V
nOr the total length of each message is greater than (V
1+ V
2+ ... V
N-1/ V
n) * S.
In sum, when message length fixedly the time, in order to guarantee to guarantee that the order of the message that receiving terminal receives is correct when improving transmission speed promptly increases transmission bandwidth, for the fastest link L1 of MFR interface medium velocity, its entry condition is: in the formation of buffer memory message, the number of message greater than 0 or the total length of each message greater than 0; For any link Li except that fastest link in the MFR interface, i is any one natural number in 2 to n, and the entry condition of this Li is: in the formation of buffer memory message, the number of message is greater than (V
1+ V
2+ ... V
I-1)/V
iJust greater than
Perhaps, in the formation of buffer memory message, the total length of each message is greater than (V
1+ V
2+ ... V
I-1) * S/V
iJust greater than
Two, no matter whether message length is fixing, in this step, all can be according to the length of message in the formation of transmitting terminal buffer memory message, and in conjunction with the speed of each link in the multilink Frame Relay Interface, the entry condition of each bar link in the multilink Frame Relay Interface is set, specifically comprises:
1) for article one link L1 after the ordering, because its speed maximum, therefore in order to guarantee to increase transmission bandwidth, the entry condition that this speed maximum link L1 is set is: in the formation of buffer memory message, the total length of each message is greater than 0, that is to say that as long as have message in the formation of buffer memory message, the link L1 that then starts MFR interface medium velocity maximum sends message.
2) for the second link L2 after the ordering, because its speed is littler but bigger than other link than L1, therefore, can't satisfy in the formation that transmission demand is the buffer memory message message at L1 constantly accumulates, guaranteeing that message arrives under the not out of order condition of receiving terminal, then should start L2 when only using L1 not use the transmission speed of L1 and L2 fast simultaneously.And the entry condition that this link L2 is set is: in the formation of buffer memory message, have a message arrangement position, make each message length before this message arrangement position and with this message arrangement position on the ratio of message length more than or equal to V
1/ V
2
Such as, the speed V of L1
1Be 44M, the speed V of L2
2Be 2M, because V
1/ V
2=22, so the time that message spent of time that message spent and the every transmission of link L2 1 byte length of 22 byte lengths of the every transmission of link L1 equates, therefore, in this step, in the formation of current buffer memory message, if message arrangement position such as each message length before the 7th the message arrangement position and with the 7th message arrangement position on the ratio of message length less than V
1/ V
2If then expression sends the 7th message arrangement position each message before by L1, and sends the 7th message on the message arrangement position by L2, so, because V
1/ V
2=22, when L1 finished preceding 6 messages in transmission, the 7th message that this L2 sends be the no show receiving terminal also, like this, then greatly reduces transmission speed, and can cause the receiving terminal packet out-ordering.Otherwise, if exist a message arrangement position such as each message length before the 7th the message arrangement position and with the 7th message arrangement position on the ratio of message length more than or equal to V
1/ V
2If then expression is sent each message before the 7th the message arrangement position and is sent the 7th message on the message arrangement position by L2 by L1, so, because V
1/ V
2=22, when L1 finished preceding 6 messages in transmission, the 7th message that this L2 sends also can arrive receiving terminal, like this, can guarantee that then improving transmission speed greatly under the correct situation of receiving terminal message sequence promptly increases transmission bandwidth.Therefore, the entry condition that this link L2 is set is: in the formation of buffer memory message, have a message arrangement position, make each message length before this message arrangement position and with this message arrangement position on the ratio of message length more than or equal to V
1/ V
2
3) for the 3rd link L3 after the ordering, because its speed is littler but bigger than other link than L1 and L2, at L1 and L2 message transmission but still can't satisfy in the formation that transmission demand is the buffer memory message accumulation message time simultaneously, owing to guaranteeing that message arrives under the not out of order condition of receiving terminal, only use L1 and L2 not to have the transmission speed of simultaneously L1, L2 and L3 fast, therefore should start L3.And the entry condition that this link L3 is set is: in the formation of buffer memory message, have a message arrangement position, make each message length before this message arrangement position and with this message arrangement position on the ratio of message length more than or equal to (V
1+ V
2)/V
3
Such as, the speed V of L1
1Be 44M, the speed V of L2
2Be 2M, the speed V of L3
3Be 1.5M,, therefore, should determine whether to exist the condition that starts L3 according to the speed of L1 and L2 with the difference of the speed of L3 because when starting L3, L1 and L2 have all started and sent message.Such as, the speed V of L1
1Be 44M, the speed V of L2
2Be 2M, because (V
1+ V
2)/V
3=30.6, thus the time that message spent of time that message spent and the every transmission of link L3 1 byte length of 30.6 byte lengths of the every transmission of link L1 and L2 equate.Therefore, in this step, in the formation of current buffer memory message, if message arrangement position such as each message length before the 7th the message arrangement position and with the 7th message arrangement position on the ratio of message length less than (V
1+ V
2)/V
3If then expression is sent each message before the 7th the message arrangement position and is sent the 7th message on the message arrangement position by L3 by L1 and L2, so, when L1 and L2 finish preceding 6 messages in transmission, the 7th message that this L3 sends be the no show receiving terminal also, like this, then greatly reduce transmission speed, and can cause the receiving terminal packet out-ordering.Otherwise, if message arrangement position such as each message length before the 7th the message arrangement position and with the 7th message arrangement position on the ratio of message length more than or equal to (V
1+ V
2)/V
3If then expression is sent each message before the 7th the message arrangement position and is sent the 7th message on the message arrangement position by L3 by L1 and L2, so, send when finishing preceding 6 messages at L1 and L2, the 7th message that this L3 sends also can arrive receiving terminal, like this, then improved transmission speed greatly.Therefore, the entry condition that this link L3 is set is: in the formation of buffer memory message, have a message arrangement position, make each message length before this message arrangement position and with this message arrangement position on the ratio of message length more than or equal to (V
1+ V
2)/V
3
The rest may be inferred, can obtain for the last item link Ln after the ordering, because its speed minimum, therefore, at L1 to Ln-1 message transmission but still can't satisfy and start Ln still can have message in the formation that transmission demand is the buffer memory message time simultaneously, therefore, the entry condition that the link Ln of this speed minimum is set is: in the formation of buffer memory message, have a message arrangement position, make each message length before this message arrangement position and with this message arrangement position on the ratio of message length more than or equal to (V
1+ V
2+ ... V
N-1)/V
n
In sum, no matter whether message length is fixing, in order to guarantee to guarantee that the order of the message that receiving terminal receives is correct when improving transmission speed promptly increases transmission bandwidth, for the fastest link L1 of MFR interface medium velocity, its entry condition can for: in the formation of buffer memory message, the total length of each message is greater than 0.For any link Li except that fastest link L1 in the MFR interface, i is the natural number greater than 1, the entry condition of this Li is: in the formation of buffer memory message, have a message arrangement position, make each message length before this message arrangement position and with this message arrangement position on the ratio of message length more than or equal to (V
1+ V
2+ ... V
I-1)/V
i, just there is a message arrangement position X, make and satisfy inequality
Wherein, S
kAnd S
XThe message length of representing K position and X position in the formation of current cache message respectively.
Step 103:, the formation of buffer memory message is detected in real time at transmitting terminal.
Here, when message length fixedly the time,, then in this step, be that the number to message in the buffer memory message queue detects in real time if be the entry condition that the number according to message in the formation of buffer memory message is provided with in step 102; If be the entry condition that is provided with according to the length of message in the formation of buffer memory message in step 102, then in this step, be that the length to message in the buffer memory message queue detects in real time.When the length of message not fixedly the time, then in this step, be that the length to message in the buffer memory message queue detects in real time.
Step 104:, when detecting the entry condition that satisfies any link in the MFR interface, determine that the current pairing message of link that satisfies entry condition obtains the position at transmitting terminal.
Step 105:, take out current determined message in the formation of the stored messages of postponing and obtain the message of position, and send by the current link that starts at transmitting terminal.
In step 104 to step 105, for the fastest link L1 of MFR interface medium velocity, no matter whether message length is fixing, according to set entry condition, in the formation that detects the buffer memory message total length of each message greater than 0 or the number that detects message in the formation of buffer memory message greater than 0 o'clock, then think the entry condition that satisfies link L1, and, determine that the pairing message of link L1 obtains the position and is: be positioned at the message of first position in the current cache message queue, be positioned at the message of first position then in the formation by this fastest link L1 transmission current cache message.
In step 104 to step 105, for any link Li except that fastest link L1 in the MFR interface, wherein, i is the natural number greater than 1, no matter whether message length is fixing, according to set entry condition, in the formation that detects the buffer memory message, there is a message arrangement position X, makes this message arrangement position X satisfy inequality
S
kAnd S
XThe message length of representing K position and X position in the formation of current cache message is respectively then thought the current entry condition that satisfies link Li.And, in order to guarantee that message arrives receiving terminal and guarantees to improve transmission speed with correct order, in this step, get the minimum value X ' of X, and the pairing message of definite link Li to obtain the position be X ', send the message that is positioned at the individual position of X ' in the formation of current cache message by link Li then.
Need to prove, to step 105, not fixedly the time, in the formation of buffer memory message, may have a message arrangement position X, make this X satisfy a plurality of inequality that i is a different value simultaneously at message length in step 104
That is to say, in practical business is realized, may detect the entry condition that a message arrangement position X satisfies multilink in the MFR interface simultaneously, like this, in order to guarantee to improve transmission speed and avoid the receiving terminal packet out-ordering, promptly detect and have a message arrangement position X whenever detecting entry condition that a message arrangement position X satisfies multilink in the MFR interface simultaneously, make this message arrangement position X satisfy a plurality of inequality that i is a different value simultaneously
The time, the minimum value of getting i that is to say, gets the corresponding the fastest link of multilink medium velocity of described different i value and sends message as the current link that satisfies entry condition.
Also need to prove, when the length of message fixedly the time, for any link Li except that fastest link L1 in the MFR interface, wherein, i is the natural number greater than 1, according to set entry condition, in step 104 to step 105, can also be in the formation that detects the buffer memory message, the number of message is greater than (V
1+ V
2+ ... V
I-1)/V
iJust greater than
The time; Perhaps, whenever detecting in the formation of buffer memory message, the total length of each message is greater than (V
1+ V
2+ ... V
I-1) * S/V
iJust greater than
The time, then think the entry condition that satisfies link Li.And, improve transmission speed in order to guarantee that message arrives receiving terminal and guarantees with correct order, determine that the pairing message of link Li obtains the position and is: in the formation of buffer memory message, be positioned at the
The message of individual position sends by link Li then and is positioned at the in the formation of current cache message
The message of individual position.
As seen, by the process of above-mentioned steps 103 to step 105, each bar link in the MFR interface then can set entry condition start the transmission message according to the present invention, and, the position of the message that sends is to have taken all factors into consideration the speed that starts link to determine, when thereby the transmission bandwidth that can be implemented in assurance increase MFR interface promptly improves transmission speed, farthest avoid the problem of receiving terminal packet out-ordering.
In addition, in the present invention, when message length not fixedly the time, owing to be to be inequality according to message length ratio
Come definite message that starts the link correspondence to obtain the position, and from inequality
As can be seen, for a definite i value, i.e. definite link, this inequality the right is a definite value, and the value on this inequality left side changes along with the value of message arrangement position X, the increase of message arrangement position X might not make
Value increase because running into a message arrangement position X message length S of place
xUnder the very big situation,
Value will reduce, like this, for a definite link Li, in the formation of buffer memory message, even the total length of message is very big, also possibly can't find message arrangement position X make
Value more than or equal to
Promptly can't detect the entry condition of link Li, thereby can reduce the transmission speed of MFR interface greatly.For fear of the generation of situation herein, the present invention can adopt the method for burst, a burst length threshold promptly is set, the length that has a message when transmitting terminal is during greater than set burst length threshold, before this message is put into the formation of buffer memory message, this message is carried out burst, that is to say, this message from first to last is divided into equal-sized a plurality of message, so that the length of other message is more or less the same in each message told and the buffer memory message queue, the a plurality of messages that to tell are then put into the formation of buffer memory message successively according to cephalocaudal order, processing procedure afterwards then can adopt above-mentioned steps 103 to the process of step 105 to realize.
In the present invention, except that the MFR interface, described multilink interface can also be other interface that binding has multilink, and such as the NP interface, at this moment, the principle of implementation procedure of the present invention and the described process of the foregoing description is identical.
In a word, the above is preferred embodiment of the present invention only, is not to be used to limit protection scope of the present invention.Within the spirit and principles in the present invention all, any modification of being done, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (8)
1, a kind of transmission method of message is applied on the binding multiple linkage interface, and message length fixes, and it is characterized in that, according to speed order from big to small each the bar link in the multilink interface is sorted;
For the fastest link L1 of multilink interface medium velocity, the entry condition that this fastest link L1 is set is that in the formation of buffer memory message, the number of message is greater than 0;
For any link Li except that fastest link L1 in the multilink interface, i is the natural number greater than 1, and the entry condition of this Li is: in the formation of buffer memory message, the number of message greater than
Wherein, Vk, Vi are respectively the speed of k bar and i bar link in the multilink interface;
This method also comprises:
A, the number of message in the message queue of transmitting terminal buffer memory is detected in real time;
B, the number of message sent the message that is positioned at first position in the formation of current cache message greater than 0 o'clock by link L1 in the formation that detects the buffer memory message; In the formation that detects the buffer memory message number of message greater than
The time, send by link Li and be positioned at the in the formation of current cache message
The message of individual position;
When described
Value when being decimal, determined
The message of individual position is
Value round up the back position message.
2, method according to claim 1 is characterized in that, described multilink interface is multilink frame relay MFR interface or multilink point-to-point interface.
3, a kind of transmission method of message is applied on the binding multiple linkage interface, and message length fixes, and it is characterized in that, according to speed order from big to small each the bar link in the multilink interface is sorted;
For the fastest link L1 of multilink interface medium velocity, the entry condition that this fastest link L1 is set is that in the formation of buffer memory message, the total length of each message is greater than 0;
For any link Li except that fastest link L1 in the multilink interface, i is the natural number greater than 1, and the entry condition of this Li is: in the formation of buffer memory message, the total length of each message greater than
Wherein, Vk, Vi are respectively the speed of k bar and i bar link in the multilink interface, and S is the length of each message;
This method also comprises:
A, the total length of each message in the message queue of transmitting terminal buffer memory is detected in real time;
B, the total length of each message sent the message that is positioned at first position in the formation of current cache message greater than 0 o'clock by link L1 in the formation that detects the buffer memory message; In the formation that detects the buffer memory message total length of each message greater than
The time, send by link Li and be positioned at the in the formation of current cache message
The message of individual position;
When described
Value when being decimal, determined
The message of individual position is
Value round up the back position message.
4, method according to claim 3 is characterized in that, described multilink interface is multilink frame relay MFR interface or multilink point-to-point interface.
5, a kind of transmission method of message is applied to it is characterized in that on the binding multiple linkage interface, according to speed order from big to small each the bar link in the multilink interface is sorted;
For the fastest link L1 of multilink interface medium velocity, the entry condition that this fastest link L1 is set is that in the formation of buffer memory message, the total length of each message is greater than 0;
For any link Li except that fastest link L1 in the multilink interface, i is the natural number greater than 1, and the entry condition that this Li is set is: in the formation of buffer memory message, exist message arrangement position X to satisfy inequality
Wherein, Vk, Vi are respectively the speed of k bar and i bar link in the multilink interface, S
kAnd S
xThe message length of representing k position and X position in the formation of current cache message respectively;
This method also comprises:
A, the total length of each message in the message queue of transmitting terminal buffer memory is detected in real time;
B, the total length of each message sent the message that is positioned at first position in the formation of current cache message greater than 0 o'clock by link L1 in the formation that detects the buffer memory message; In the formation that detects the buffer memory message, exist message arrangement position X to satisfy inequality
The time, get the minimum value X ' of X, send the message that is positioned at the individual position of X ' in the formation of current cache message by link Li.
6, method according to claim 5 is characterized in that, is the different links of different value when detecting for i, and when having identical message arrangement position X ', the link Li of the individual position of described transmission X ' message is the fastest link of this difference link medium velocity.
7, according to claim 5 or 6 described methods, it is characterized in that, when transmitting terminal message length not fixedly the time, this method further comprises: the length that has a message when transmitting terminal is during greater than the burst length threshold that sets in advance, when this message is put into the formation of buffer memory message, this message from first to last is divided into equal-sized a plurality of message, and a plurality of messages that will tell are then put into the formation of buffer memory message successively according to cephalocaudal order.
8, method according to claim 5 is characterized in that, described multilink interface is multilink frame relay MFR interface or multilink point-to-point interface.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100073972A CN100563238C (en) | 2006-02-13 | 2006-02-13 | A kind of transmission method of message |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100073972A CN100563238C (en) | 2006-02-13 | 2006-02-13 | A kind of transmission method of message |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1809050A CN1809050A (en) | 2006-07-26 |
| CN100563238C true CN100563238C (en) | 2009-11-25 |
Family
ID=36840741
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2006100073972A Expired - Fee Related CN100563238C (en) | 2006-02-13 | 2006-02-13 | A kind of transmission method of message |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100563238C (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101309154B (en) * | 2007-05-18 | 2011-04-13 | 华为技术有限公司 | Datagram sending method, sending apparatus and transmission system |
| CN102223306B (en) * | 2011-06-17 | 2018-10-16 | 南京中兴软件有限责任公司 | A kind of message transmitting method and device |
| CN104168093B (en) * | 2013-05-20 | 2017-07-25 | 腾讯科技（深圳）有限公司 | A kind of document transmission method and device |
| CN109218217B (en) * | 2017-07-04 | 2022-01-28 | 中兴通讯股份有限公司 | Load sharing method, device, routing equipment and storage medium |
| CN108765847A (en) * | 2018-05-03 | 2018-11-06 | 南京铁道职业技术学院 | A kind of fire alarm system |
| CN108538034A (en) * | 2018-05-03 | 2018-09-14 | 南京铁道职业技术学院 | A kind of method of fire alarm system |
-
2006
- 2006-02-13 CN CNB2006100073972A patent/CN100563238C/en not_active Expired - Fee Related
Non-Patent Citations (1)
| Title |
|---|
| "多链路PPP的调度算法". 王炜等.北京邮电大学学报，第23卷第3期. 2000 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1809050A (en) | 2006-07-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100563238C (en) | A kind of transmission method of message | |
| JP4648833B2 (en) | Bandwidth management device | |
| CN110545241B (en) | Message processing method and device | |
| US20080247410A1 (en) | Creating A Low Bandwidth Channel Within A High Bandwidth Packet Stream | |
| EP2129045B1 (en) | Master slave communication system and master slave communication method | |
| US20080232253A1 (en) | Port rate smoothing in an avionics network | |
| CN102811176B (en) | A kind of data flow control method and device | |
| CN111147184B (en) | Ethernet transmitting and receiving equipment supporting multi-service preemption and continuous transmission and transmission device | |
| KR101119300B1 (en) | Residential Ethernet Node Apparatus For Strict Guarantee of Super Frame's Start and Its Frame Processing Method | |
| CN100420238C (en) | Method for partitioned recombining IP message | |
| JP4652314B2 (en) | Ether OAM switch device | |
| US20150023356A1 (en) | Alignment circuit and receiving apparatus | |
| CN111194056A (en) | Encapsulation of data packets | |
| CN101483637B (en) | Method for lowering buffered packet number in Multilink PPP waiting queue | |
| US20100014421A1 (en) | Speculative credit data flow control | |
| JP3416156B2 (en) | Method and circuit arrangement for transmitting information cells via virtual connections of different priorities | |
| CN110267310B (en) | Link capacity adjusting method, device, system, controller and network node | |
| US7324564B2 (en) | Transmitting odd-sized packets over a double data rate link | |
| KR101196823B1 (en) | Method and apparatus for constant-bit-rate traffic clock recovery | |
| CN105187949A (en) | Video transmission method and client | |
| Yang et al. | Research on delay and packet loss control mechanism in wireless mesh networks | |
| De Turck et al. | Delay analysis of the Go-Back-N ARQ protocol over a time-varying channel | |
| Bellalta | A queuing model for the non-continuous frame assembly scheme in finite buffers | |
| De Turck et al. | Delay and stability aspects of the go-back-N ARQ scheme for correlated error channels | |
| Chlebus et al. | Stability of the multiple-access channel under maximum broadcast loads |
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 | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 310052 Binjiang District Changhe Road, Zhejiang, China, No. 466, No. Patentee after: Xinhua three Technology Co., Ltd. Address before: 310053 Hangzhou hi tech Industrial Development Zone, Zhejiang province science and Technology Industrial Park, No. 310 and No. six road, HUAWEI, Hangzhou production base Patentee before: Huasan Communication Technology 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: 20091125 Termination date: 20210213 |