CN1549511A - Flow shaping processing method - Google Patents
Flow shaping processing method Download PDFInfo
- Publication number
- CN1549511A CN1549511A CNA03131337XA CN03131337A CN1549511A CN 1549511 A CN1549511 A CN 1549511A CN A03131337X A CNA03131337X A CN A03131337XA CN 03131337 A CN03131337 A CN 03131337A CN 1549511 A CN1549511 A CN 1549511A
- Authority
- CN
- China
- Prior art keywords
- burst
- data message
- message
- bandwidth
- traffic shaping
- 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
The method includes the following steps: a) data message is received, b)whether burst is carried on for data message or not is judged and step c) is applied if it is or step d) is applied if it is not; c) bandwidth required for bursting is calculated, d) whether FRTS token is satisfied to requirement of bandwidth or not is judged and step e) is applied if it is or step f) is applied if it is not, e) data message is bursted and transmitted, and f) data message is not transmitted.
Description
Technical field
The present invention relates to the communications field, relate in particular to when frame relay burst and frame relay traffic shaping applied in any combination, the processing method of burst occupied bandwidth.
Technical background
VoFR has expanded the application of frame relay, makes the terminal use to carry out different business such as data, voice and fax simultaneously on a lease line, improves bandwidth availability ratio greatly.When the speech data simultaneous interpretation, cause voice delay time to increase for fear of sending big data message, shake tangible phenomenon, introduced the method for data fragmentation.By being split as big data message a plurality of and the sizable small messages of voice, voice and fragment data send at interval, and voice are in time handled, and have avoided the generation of delay variation.
In order to send data traffic with uniform speed to network, frame relay has been introduced the method for traffic shaping (Traffic Shaping), controls the data flows by the mechanism of token bucket in addition.Token bucket algorithm employing token bucket comes the administration queue controller, allows bursty data to exist, and it can also guarantee a kind of long-time interior stable average transmission rate simultaneously.Token bucket algorithm produces token with constant speed, leaks bucket and can keep token.Input data data flow out token bucket after entering bucket with Louing and obtaining token, if do not obtain the just generation of the next token of wait in bucket of token.Each token can transmit a grouping, and data message flows out the corresponding token in bucket back and just disappeared.The speed that produces by the control token like this, the Mean Speed of data that just can the control flows dequeue.By the mechanism of this token bucket, can make Virtual network operator guarantee service for the user provides minimum committed information rate (MinCIR).When network was idle, the user can be with the speed multiplexing bandwidth greater than MinCIR, and when network congestion, operator can guarantee to provide for the user bandwidth of MinCIR.Its schematic diagram is as shown in Figure 1:
Fig. 1 is the processing procedure of traffic shaping, as can be seen from the figure, and during existing traffic shaping is handled, when data message arrives, at first the data message is classified, handle, then data message is sent into traffic shaping queue if data message need carry out traffic shaping.If formation is empty, then enters in the token bucket and handle.If there are enough tokens can be used for sending datagram in the token bucket, then data message can pass through, and can be continued to send.If the token in the token bucket does not satisfy the transmission condition of data message, then data message goes into to get back in the traffic shaping queue; If queue full then data message is dropped.When in the traffic shaping queue data message being arranged, traffic shaping takes out data message by certain cycle and sends from formation, when sending datagram, will all send the data volume of the representative of the token in the traffic shaping token bucket at every turn.
When frame relay burst and traffic shaping applied in any combination,, can take than original higher bandwidth because behind the data data fragmentation, fragment message will add extra heading.After for example the data message of one 1500 byte arrived the frame relay link layer, if do not dispose burst, according to protocol encapsulation form Q.922, the byte that needs to add comprises: 2 byte DLCI heads+2 byte protocol numbers+2 byte FCS, total length was 1506 bytes.If disposed frame relay FRF.12 burst, and the burst size is 30 words, the protocol number (NLPID) that the data message of 1500 bytes adds 2 bytes 1502 bytes altogether must be split as 51, the extra byte that each burst need add comprises: 2 byte DLCI heads+2 byte slotted protocols number+2 byte fragment message heads+2 byte FCS, just each burst will consume 8 bytes.The bandwidth that the data actual needs of such 1500 bytes takies is: the 51*8+1502=1910 byte has increased by 26.8% than 1506 original bytes.And during with the traffic shaping applied in any combination,, can cause flow and the serious deviation of expection after the shaping if do not consider the extra bandwidth that burst takies.
In the existing flow shape technique field, company of Cisco (CISCO) has adopted elder generation that the data message is carried out burst and then has done the method for traffic shaping.The token of the extra heading that all like this bursts take in yet can consumed flow shaping token bucket can not cause the flow rate calculation mistake.This first burst of Cisco is done the method for traffic shaping again, can avoid the inaccurate phenomenon of flow rate calculation, can't successfully send the problem that causes the opposite end to recombinate but occur fragment message easily.The data message of one 1500 byte for example recited above, split into 51 after, need take 1910 bytes.If the current size of traffic shaping token bucket this moment can only send 20 bursts, remaining 31 bursts can only be gone into formation earlier, if formation can't be deposited all 31 data messages, must abandon some fragment message, suppose to have abandoned 50 and 51 two bursts.All the other 49 bursts arrive behind opposite ends owing to can't receive the burst end message all the time like this, so burst can't be recombinated, can only abandon at last, and these 49 bursts can be thought " rubbish message ", have taken meaningless bandwidth.So must " complete send out full the receipts " at frame relay burst reorganization, any one burst abandon the disabled characteristics that all can cause whole group.
Summary of the invention
Purpose of the present invention is exactly to do at first burst in the above-mentioned prior art to occur fragment message easily in the method for traffic shaping and can't successfully send again, and the problem that causes the opposite end to recombinate proposes the method that a kind of first burst is done traffic shaping again.
A kind of traffic shaping processing method is characterized in that may further comprise the steps:
A, receiving data packets;
B, judge whether and need carry out burst the data message, if, enter step c, if not, enter steps d;
The bandwidth that c, calculating burst need enters steps d;
D, judge whether the FRTS token meets the demand of bandwidth, if, enter step e, if not, enter step f;
E, with the data message burst and send;
F, data message is not sent processing.
The step of the bandwidth that described calculating burst needs, calculate by following formula:
Total byte behind the burst=(forward rounds (fragmented message length/burst size)) * (fragment message head length degree+slotted protocol byte+frame relay heading+frame relay FCS byte)+fragmented message length.
Described step f is that data message is abandoned.
Described step f is that data message is gone into formation again, returns steps d.
Owing to adopted traffic shaping processing method of the present invention, can avoid fragment message all to send, the problem that causes receiving terminal correctly not recombinate.Simultaneously, because the bandwidth that accurate calculating burst need consume is avoided traffic statistics problem less than normal.
Description of drawings
Fig. 1 is a traffic shaping processing procedure schematic diagram in the prior art;
Fig. 2 is a sliced bandwidth computing flow chart of the present invention;
Fig. 3 is another sliced bandwidth computing flow chart of the present invention.
Embodiment
Below in conjunction with Figure of description the specific embodiment of the present invention is described.
The advantage that elder generation's traffic shaping is done the data message stripping strategy again is: processing represents that promptly transmitting terminal has enough bandwidth to guarantee the transmission of data message if data message has passed through shaping, data message is done burst at this moment again and is handled, just can guarantee that all bursts can both send at transmitting terminal, avoid the generation of " rubbish message ".Here the problem that will solve is: owing to will consume more bandwidth behind the burst, how during at burst not this part extra bandwidth is counted accurately, avoid surpassing target flow.
Shown in Fig. 2 and 3, it is the flow chart that adopts traffic shaping processing method of the present invention, as can be seen from the figure, the method that the present invention adopts is, when data message process traffic shaping, when calculating the token number that needs consumption, judge that according to the slotted protocol type, the extra bandwidth that need consume burst counts in advance if need the data message of burst, if token enough data message can send, otherwise go into formation or abandon.The bandwidth calculation of data message, traffic shaping, burst unification are all finished by router.
In view of the above, traffic shaping processing method of the present invention may further comprise the steps:
A, receiving data packets;
Receive data message earlier from the upper strata.
B, judge whether and need carry out burst the data message, if, enter step c, if not, enter steps d;
This step will judge earlier whether the data message that receives among this step a needs to carry out burst, according to design philosophy of the present invention, carries out burst if desired, the step that then enters the computation bandwidth of back, otherwise directly this data message is entered token bucket.
The bandwidth that c, calculating burst need enters steps d;
This is a key point place of the present invention, and promptly the extra bandwidth that need consume burst in advance counts.In the present invention, adopt following bandwidth calculation method:
Total byte behind the burst=(forward rounds (fragmented message length/burst size)) * (fragment message head length degree+slotted protocol byte+frame relay heading+frame relay FCS byte)+fragmented message length.
Like this, just can guarantee that all bursts can both send at transmitting terminal, avoid the generation of " rubbish message ".
D, judge whether the FRTS token meets the demand of bandwidth, if, enter step e, if not, enter step f;
The FRTS frame relay traffic shaping is to control the data flows by the method for token bucket, and token can be understood as the ability that data send.The present invention is not specifically limited the token mechanism that adopts.
E, with the data message burst and send;
F, data message is not sent processing.
Wherein, do not send processing among the step f, can comprise following situation:
1, data message is abandoned;
This is actually the transmission flow of the data message that is through with, and this is comparatively common a kind of processing mode during message sends, and usually adopts when the data message generation is congested.
2, data message is gone into formation again, return steps d.
This mode is that a postponement that sends message is sent, and the data message that can't normally send at present is included into formation again, waits for when Internet resources allow that resend, this also is a kind of comparatively common mode again.
Below we see an object lesson that adopts traffic shaping processing method of the present invention:
If disposed the FRF.12 burst, the burst size is 50 bytes, and then the total bandwidth of the fragmented message needs of 800 bytes is,
Total byte behind the burst=
Forward rounds (800/50) * (2+2+2+2)+800=16*8+800=928
The concrete length of above fragment message head length degree and slotted protocol byte is with reference to the definition of each agreement.
The present invention can avoid fragment message all to send, the problem that causes receiving terminal correctly not recombinate in the process that data message sends by the strength reshaping treatment method of above narration.Simultaneously, because the bandwidth that accurate calculating burst need consume can avoid traffic statistics less than normal.Because the method for frame relay bandwidth accurate Calculation is used to control local terminal router transmission rate, do not have the inter-communicating problem between the different vendor, so other routers adopts this algorithm also can realize corresponding function.
The above; only for the preferable embodiment of the present invention, but protection scope of the present invention is not limited thereto, and anyly is familiar with those skilled in the art in the technical scope that the present invention discloses; the variation that can expect easily or replacement all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection range of claims.
Claims (4)
1, a kind of traffic shaping processing method is characterized in that may further comprise the steps:
A, receiving data packets;
B, judge whether and need carry out burst the data message, if, enter step c, if not, advance
Go into steps d;
The bandwidth that c, calculating burst need enters steps d;
D, judge whether the FRTS token meets the demand of bandwidth, if, enter step e, if not, advance
Go into step f;
E, with the data message burst and send;
F, data message is not sent processing.
2, traffic shaping processing method as claimed in claim 1 is characterized in that the step of the bandwidth of described calculating burst needs, calculates by following formula:
Total byte behind the burst=(forward rounds (fragmented message length/burst size)) * (fragment message head length degree+slotted protocol byte+frame relay heading+frame relay frame verification and byte)+fragmented message length.
3, traffic shaping processing method as claimed in claim 2 is characterized in that described step f, is that data message is abandoned.
4, traffic shaping processing method as claimed in claim 2 is characterized in that described step f, is that data message is gone into formation again, returns steps d.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB03131337XA CN100499477C (en) | 2003-05-13 | 2003-05-13 | Flow shaping processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB03131337XA CN100499477C (en) | 2003-05-13 | 2003-05-13 | Flow shaping processing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1549511A true CN1549511A (en) | 2004-11-24 |
| CN100499477C CN100499477C (en) | 2009-06-10 |
Family
ID=34322840
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB03131337XA Expired - Fee Related CN100499477C (en) | 2003-05-13 | 2003-05-13 | Flow shaping processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100499477C (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101119492B (en) * | 2007-08-20 | 2011-03-16 | 武汉烽火网络有限责任公司 | Jitter restriction based data cache reshaping method |
| CN102271086A (en) * | 2011-07-25 | 2011-12-07 | 华为技术有限公司 | Data transmission method and device |
| CN103780907A (en) * | 2014-01-27 | 2014-05-07 | 浙江宇视科技有限公司 | Method and device for video data flow shaping |
-
2003
- 2003-05-13 CN CNB03131337XA patent/CN100499477C/en not_active Expired - Fee Related
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101119492B (en) * | 2007-08-20 | 2011-03-16 | 武汉烽火网络有限责任公司 | Jitter restriction based data cache reshaping method |
| CN102271086A (en) * | 2011-07-25 | 2011-12-07 | 华为技术有限公司 | Data transmission method and device |
| US9225654B2 (en) | 2011-07-25 | 2015-12-29 | Huawei Technologies Co., Ltd. | Method and apparatus for sending packet |
| CN103780907A (en) * | 2014-01-27 | 2014-05-07 | 浙江宇视科技有限公司 | Method and device for video data flow shaping |
| CN103780907B (en) * | 2014-01-27 | 2018-01-16 | 浙江宇视科技有限公司 | A kind of method and apparatus of video data flow shaping |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100499477C (en) | 2009-06-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6694471B1 (en) | System and method for periodic retransmission of messages | |
| US8027344B2 (en) | Transmission of data packets of different priority levels using pre-emption | |
| CN1778079B (en) | Methods and devices for the coordination of flow control between a TCP/IP network and other networks | |
| JP3321043B2 (en) | Data terminal in TCP network | |
| CN107979449B (en) | Data transmission method and device | |
| US8509080B2 (en) | Network traffic accelerator | |
| CN100534069C (en) | Accelerating method for asymmetric and multi-concurrency network | |
| CN1926816A (en) | Communication quality management method and apparatus | |
| CN101080914A (en) | Method, system and article for improved network performance by avoiding data corruption | |
| CN1842052A (en) | Data transmission method for wireless link control layer | |
| US7321557B1 (en) | Dynamic latency assignment methodology for bandwidth optimization of packet flows | |
| CN1248465C (en) | Management method of data fransmission/receiving butter region in network communication | |
| CN1633796A (en) | Modifications to TCP/IP for broadcast or wireless networks | |
| CN1152313C (en) | Section-by-section congestion control method based on random early detection | |
| CN1194509C (en) | Method for controlling message transmitting spped rate in router interface | |
| CN1549511A (en) | Flow shaping processing method | |
| EP2002609B1 (en) | Method and termination node for bundling multiple messages into a packet | |
| US7633947B2 (en) | Method and apparatus for performing active packet bundling in a Voice over-IP communications system based on source location in talk spurts | |
| CN115695582A (en) | Low-delay transmission control method, receiving terminal and transmitting terminal | |
| CN1719801A (en) | Method for improving connecting performance of multi-non-connecting layer in wireless network | |
| CN1901519A (en) | Optimizing system and method for medium transmission on different transmission channels | |
| CN113038530A (en) | High-efficiency transmission method for packet service of QoS guarantee of satellite mobile communication system | |
| US7088738B1 (en) | Dynamic fragmentation of information | |
| KR101040780B1 (en) | ??????? based Real-time Multimedia Data Transmission for QoS enhancement | |
| CN1630270A (en) | Method for dynamically adjusting data packet length based on network service quality |
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 | ||
| 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: 20090610 Termination date: 20160513 |