CN102404214A - Bandwidth distribution method, device and network equipment - Google Patents
Bandwidth distribution method, device and network equipment Download PDFInfo
- Publication number
- CN102404214A CN102404214A CN2011103692941A CN201110369294A CN102404214A CN 102404214 A CN102404214 A CN 102404214A CN 2011103692941 A CN2011103692941 A CN 2011103692941A CN 201110369294 A CN201110369294 A CN 201110369294A CN 102404214 A CN102404214 A CN 102404214A
- Authority
- CN
- China
- Prior art keywords
- user
- ethernet frame
- data volume
- sent
- output queue
- 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)
- Small-Scale Networks (AREA)
Abstract
The invention provides a bandwidth distribution method, a device and a piece of network equipment. The method comprises the following steps: classifying the Ethernet frames in the output queues on an Ethernet port according to user characteristics, and generating the output queues corresponding to each user; respectively determining the Ethernet frames of each user to be transmitted according to the distributed bandwidth of each user and the length of the Ethernet frame in the output queue corresponding to the user; and adding the Ethernet frames of each user to be transmitted to a hardware transmitting queue and transmitting to Ethernet lines. The device comprises a classifying module, a calculating module and a transmitting module. The network equipment comprises the bandwidth distribution device. The bandwidth distribution mechanism of the invention is simple, so the calculation resource consumption of the system is little.
Description
Technical field
The present invention relates to the communication technology, relate in particular to a kind of bandwidth allocation methods, device and the network equipment.
Background technology
Ethernet (Ethernet) is as a kind of LAN networking technology; It is the most general local area network technology of current application; The IEEE802.3 standard has provided the technical standard of Ethernet, and it has stipulated to comprise the content of line, the signal of telecommunication and the medium access layer protocol of physical layer.The standard topology structure of Ethernet is the bus-type topology; But present Fast Ethernet is in order farthest to reduce conflict; Farthest improve network speed and service efficiency; Use switch (switch hub) to carry out network connection and tissue, so just formed star-like Ethernet topological structure, but logically still used bus-type topological sum csma/cd (Carrier Sense Multiple Access/Collision Detect; Hereinafter to be referred as: bus conflict technology CSMA/CD).Allocated bandwidth is a kind of technology commonly used on the current various communication line, is used to control the bandwidth of the shared circuit of certain Business Stream.
Be used for the most frequently used allocated bandwidth mechanism of Ethernet in the prior art and be configuration service quality (Quality of Service on Ethernet interface; Hereinafter to be referred as QoS) business, flow the priority of the IP load of formulating each ethernet frame through identification services, get into different formations according to different priorities again, the bandwidth that takies according to each formation is calculated the quantity of the frame that send to circuit.
Yet, the allocated bandwidth mechanism more complicated of prior art is bigger to the calculation resources consumption of system.
Summary of the invention
First aspect of the present invention provides a kind of bandwidth allocation methods, comprising:
According to user characteristics the ethernet frame in the output queue on the Ethernet interface is classified, and generate the corresponding output queue of each user;
The length of each ethernet frame in the corresponding output queue of the bandwidth that is assigned with according to said each user and said each user is confirmed said each user's ethernet frame to be sent respectively;
Successively said each user's ethernet frame to be sent is joined in the hardware transmit queue and send to the ether circuit.
Another aspect of the present invention provides a kind of allocated bandwidth device, comprising:
Sort module is used for according to user characteristics the ethernet frame of output queue on the Ethernet interface being classified, and generates the corresponding output queue of each user;
Computing module, the length of each ethernet frame of output queue that the bandwidth that is used for being assigned with according to said each user and said each user are corresponding is confirmed said each user's ethernet frame to be sent respectively;
Sending module is used for successively ethernet frame to be sent with said each user and joins the hardware transmit queue and send to the ether circuit.
Another aspect of the present invention provides a kind of network equipment, comprises above-mentioned allocated bandwidth device.
Technique effect of the present invention is: according to user characteristics the ethernet frame in the output queue on the Ethernet interface is divided in the corresponding output queue of each user; The length of each ethernet frame in the bandwidth that is assigned with according to the user and the corresponding output queue; Confirm user's ethernet frame to be sent, and each user's ethernet frame to be sent joined in the hardware transmit queue send to the ether circuit; The present invention carries out simple analysis through the length of ethernet frame in bandwidth that the user is assigned with and the output queue and calculates; Just can realize the accurate control of the band width in physical that the data to different user take on the ether circuit; Allocated bandwidth mechanism of the present invention is fairly simple, and is less to the calculation resources consumption of system.
Description of drawings
Fig. 1 is the flow chart of bandwidth allocation methods embodiment one of the present invention;
Fig. 2 is the flow chart of bandwidth allocation methods embodiment two of the present invention;
Fig. 3 is bandwidth allocation sketch map one among the bandwidth allocation methods embodiment two of the present invention;
Fig. 4 is bandwidth allocation sketch map two among the bandwidth allocation methods embodiment two of the present invention;
Fig. 5 is the transformat sketch map of ethernet frame among the bandwidth allocation methods embodiment two of the present invention;
Fig. 6 is bandwidth allocation sketch map three among the bandwidth allocation methods embodiment two of the present invention;
Fig. 7 is bandwidth allocation sketch map four among the bandwidth allocation methods embodiment two of the present invention;
Fig. 8 is the structural representation of allocated bandwidth device embodiment one of the present invention;
Fig. 9 is the structural representation of allocated bandwidth device embodiment two of the present invention.
Embodiment
Fig. 1 is the flow chart of bandwidth allocation methods embodiment one of the present invention, and as shown in Figure 1, present embodiment provides a kind of bandwidth allocation methods, can specifically comprise the steps:
Present embodiment provides a kind of allocated bandwidth mechanism at the Ethernet transmitting data, and the Ethernet here can be specially two Ethernets that website connects in the system, does not comprise the Ethernet that connects a plurality of autonomous systems.Transmission need be controlled the bandwidth that every service traffics take during data between any two ethernet devices, and is concrete through the scheduling controlling to each ethernet frame, and the bandwidth of each CU is limited in respectively within the bandwidth that is assigned with separately.This step describes with the example that is scheduling to the ethernet frame in the output queue on certain Ethernet interface; According to user characteristics the ethernet frame in the output queue on the Ethernet interface is classified, specifically can be according to VLAN (the Virtual Local Area Network of ethernet frame in the output queue; Hereinafter to be referred as: VLAN) sign (Identifier; Hereinafter to be referred as: ID) classify, also can be according to medium access control (the Media Access Control of ethernet frame; Hereinafter to be referred as: MAC) classify in the address, and generate the corresponding output queue of each user, and it is temporary respectively in different output queues to be about to classification ethernet frame afterwards, the corresponding output queue of each user.
After ethernet frame is carried out classification processing,, come to confirm respectively each user's ethernet frame to be sent according to the length and the bandwidth that each user is assigned with of each ethernet frame in the corresponding output queue of each user of classification back.In the corresponding output queue of each user; The ethernet frame that possibly comprise varying number; The length of ethernet frame has nothing in common with each other in the output queue, and shared bandwidth had nothing in common with each other when promptly each ethernet frame transmitted on physical circuit, and system is also inequality for the bandwidth that each user distributes.This step is through the length of each ethernet frame in the corresponding output queue of analysis user; The bandwidth that is assigned with in conjunction with the user; Confirm this user's ethernet frame to be sent, the data volume summation of the feasible ethernet frame of confirming to be sent is within the restriction of the bandwidth that this user is assigned with.Present embodiment is a unit with each dispatching cycle, analyzes the ethernet frame of confirming that each user can send in each dispatching cycle.
Behind the ethernet frame to be sent of confirming each user; Each user's that will confirm successively ethernet frame to be sent joins in the hardware transmit queue; Completion is to the once scheduling of ethernet frame in the output queue, and follow-up sequencing according to each ethernet frame entering formation to be sent in the hardware transmit queue sends to the ether circuit with ethernet frame to be sent.
Present embodiment provides a kind of bandwidth allocation methods; According to user characteristics the ethernet frame in the output queue on the Ethernet interface is divided in the corresponding output queue of each user; The length of each ethernet frame in the bandwidth that is assigned with according to the user and the corresponding output queue; Confirm user's ethernet frame to be sent, and each user's ethernet frame to be sent joined in the hardware transmit queue send to the ether circuit; Present embodiment carries out simple analysis through the length of ethernet frame in bandwidth that the user is assigned with and the output queue and calculates; Just can realize the accurate control of the band width in physical that the data to different user take on the ether circuit; The allocated bandwidth mechanism of present embodiment is fairly simple, and is less to the calculation resources consumption of system.
Fig. 2 is the flow chart of bandwidth allocation methods embodiment two of the present invention, and as shown in Figure 2, present embodiment provides a kind of bandwidth allocation methods, can specifically comprise the steps:
This step describes with the example that is scheduling to the ethernet frame in the output queue on certain Ethernet interface; According to user characteristics the ethernet frame in the output queue on the Ethernet interface is classified; And generate the corresponding output queue of each user; It is temporary respectively in different output queues to be about to classification ethernet frame afterwards, the corresponding output queue of each user.Fig. 3 is bandwidth allocation sketch map one among the bandwidth allocation methods embodiment two of the present invention; As shown in Figure 3; After ethernet frame 1-22 in the output queue on certain ether interface classified; Be divided into respectively in the corresponding output queue of each user, be about to ethernet frame 1,2,5,6,7 and be divided into user's 1 output queue, ethernet frame 3,4,8,9,10 is divided into user's 2 output queues; Ethernet frame 11,12,13,14,15,17 is divided into user's 3 output queues, ethernet frame 16,18,19,20,21,22 is divided into user's 4 output queues.
In the present embodiment; With the dispatching cycle is that unit sends control to the ethernet frame in the corresponding output queue of each user; Each user's dispatching cycle is identical in the system; But the time of each dispatching cycle maybe be different, specifically adjusted according to the actual schedule situation by system, are generally about 100us.In first dispatching cycle, current system dispatched users output queue for the first time sends, and supposes the dispatching cycle of 150us not.In first dispatching cycle, only write down the system time of this scheduling, and local scheduling does not carry out any processing to user's data, will or not join in the hardware transmit queue user's data.When dispatching for the second time; Still write down the system time of this scheduling; Calculate this scheduling time at interval according to the system time of this scheduling of writing down and the system time of last scheduled, two system times that are about to record are got difference, can suppose that here difference is 160us.
Calculate this scheduling time at interval after, this step is for according to the local scheduling time interval and the bandwidth that each user is assigned with, the data volume of coming concrete each user of calculating in this dispatching cycle, can send.Fig. 4 is bandwidth allocation sketch map two among the bandwidth allocation methods embodiment two of the present invention; As shown in Figure 4; The bandwidth that user 1, user 2, user 3 and user 4 are assigned with is respectively 16Mbps, 10Mbps, 4Mbps and 30Mbps; Because this scheduling time that above-mentioned steps 202 is calculated is spaced apart 160us, then the data volume that can send of user 1 is 2560bit, and the data volume that user 2 can send is 1600bit; The data volume that user 3 can send is 640bit, and the data volume that user 4 can send is 4800bit.
In the present embodiment; The bandwidth that the user's data amount of sending in each dispatching cycle is assigned with the user is a limit; But because the length of each ethernet frame has nothing in common with each other in the output queue of user's correspondence; The data volume that above-mentioned steps 203 definite users can send can is a bit larger tham or be slightly smaller than to the data volume of therefore sending in each dispatching cycle in the present embodiment, promptly exists a correction value that the data volume that the user can send is revised.This step is for obtaining the corresponding data volume correction value of each user respectively, and this data volume correction value had nothing in common with each other for each dispatching cycle, need in each dispatching cycle, calculate respectively.The corresponding data volume correction value of each user is respectively the difference of the data volume that data volume that each user sends in the cycle in last scheduled and user can send in the last scheduled cycle in the present embodiment.In the present embodiment with t represent with us be unit this scheduling time at interval; Represent with Mbps to be the bandwidth that the user of unit is assigned with w; With s
1Represent that the ethernet frame that sends this dispatching cycle surpasses the quantity of the data volume that can send in this dispatching cycle, unit is bit; With s
2Represent that the ethernet frame that sends this dispatching cycle is less than the quantity of the data volume that can send in this dispatching cycle, unit is bit; With s
3The ethernet frame that sends of expression last scheduled cycle surpasses the quantity of the data volume that last scheduled can send in the cycle, and unit is bit; With s
4The ethernet frame that sends of expression last scheduled cycle is less than the quantity of the data volume that last scheduled can send in the cycle, and unit is bit.The corresponding data volume correction value of the user who obtains in this step promptly can be expressed as (s
3-s
4).
Step 205; The length of each ethernet frame in the corresponding output queue of the corresponding data volume correction value of the data volume that in this dispatching cycle, can send according to said each user, said each user, preset data volume preset value and each user is confirmed the quantity of the to be sent ethernet frame of said each user in this dispatching cycle respectively.
Get access to data volume that each user can send and the corresponding data volume correction value of each user in this dispatching cycle through above-mentioned steps after; The length of each ethernet frame in the corresponding output queue of the corresponding data volume correction value of the data volume that this step can be sent in this dispatching cycle according to the user, each user, preset data volume preset value and each user, the quantity of coming to confirm respectively the to be sent ethernet frame of each user in this dispatching cycle.Wherein, preset data volume preset value can be used s
5Represent that unit is bit, represent that the band width in physical that the transmission ethernet frame takies in predefined each dispatching cycle can be above the quantity of bandwidth.
Fig. 5 is the transformat sketch map of ethernet frame among the bandwidth allocation methods embodiment two of the present invention; As shown in Figure 5; Ethernet frame transmits on physical circuit according to the form shown in the figure; When an ethernet frame transmits on physical circuit; Except the fields of in system, seeing: destination address (Destination Address), source address (Source Address), type length (Type Length) and the data (Data), also comprise CRC (the Cyclic Redundancy Check of postamble; Hereinafter to be referred as: CRC) field, anterior frame gap (the InterFrame Gap of frame; Hereinafter to be referred as: IFG) field, leading (Preamble) field and starting-frame delimiter (Start Frame Delimiter; Hereinafter to be referred as: it is that an ethernet frame transmits shared band width in physical on the line that SFD) field, these fields add together.
Continuation is referring to above-mentioned Fig. 4, and in the transmit queue of user's 1 correspondence, the length of supposing frame 1 is 64 bytes, and the length of frame 2 is 70 bytes, and the length of frame 5 is 120 bytes, and the length of frame 6 is 110 bytes, and the length of frame 7 is 220 bytes; In the transmit queue of user's 2 correspondences, the length of supposing frame 3 is 80 bytes, and the length of frame 4 is 75 bytes, and the length of frame 8 is 160 bytes, and the length of frame 9 is 220 bytes, and the length of frame 10 is 260 bytes; In the transmit queue of user's 3 correspondences, the length of supposing frame 11 is 90 bytes, and the length of frame 12 is 100 bytes, and the length of frame 13 is 190 bytes, and the length of frame 14 is 700 bytes, and the length of frame 15 is 210 bytes, and the length of frame 17 is 90 bytes; In the transmit queue of user's 4 correspondences, the length of supposing frame 16 is 65 bytes, and the length of frame 18 is 90 bytes, and the length of frame 19 is 100 bytes, and the length of frame 20 is 110 bytes, and the length of frame 21 is 111 bytes, and the length of frame 22 is 590 bytes.
In this step, can specifically adopt following formula (1) to calculate the quantity of the to be sent ethernet frame of each user in this dispatching cycle:
Wherein, F
iRepresent i ethernet frame in the output queue, n is a positive integer.Through the length of each ethernet frame in the corresponding output queue of each user is carried out analytical calculation, can get access to the n value that satisfies condition shown in the above-mentioned formula (1).The difference and the data volume preset value sum of the data volume correction value that the data volume that the data volume of the user's who satisfies formula (1) condition who promptly obtains n ethernet frame to be sent can be sent in this dispatching cycle less than this user is corresponding with this user.Be example still with above-mentioned Fig. 4; Obtain through analytical calculation; The quantity n of the to be sent ethernet frame of user 1 in this dispatching cycle is 3; The quantity n of the to be sent ethernet frame of user 2 in this dispatching cycle is 2, and the quantity n of the to be sent ethernet frame of user 3 in this dispatching cycle is 1, and the quantity n of the to be sent ethernet frame of user 4 in this dispatching cycle is 5.
Step 206 according to the quantity of the said to be sent ethernet frame of said each user in this dispatching cycle, is confirmed said each user's ethernet frame to be sent from the corresponding output queue of said each user.
Behind the quantity n that gets access to the to be sent ethernet frame of each user in this dispatching cycle, according to this quantity definite each user's from the corresponding output queue of each user ethernet frame to be sent, promptly putting in order according to ethernet frame in the output queue confirmed.For user 1, the ethernet frame of confirming to be sent is to come preceding 3 frame of output queue 1,2,5; For user 2, the ethernet frame of confirming to be sent is to come preceding 2 frame of output queue 3,4; For user 3, the ethernet frame of confirming to be sent is to come preceding 1 frame of output queue 11; For user 4, the ethernet frame of confirming to be sent is to come preceding 5 frame of output queue 16,18,19,29,21.
Behind the ethernet frame to be sent of confirming each user; Successively each user's ethernet frame to be sent is joined in the hardware transmit queue according to user's order; As shown in Figure 4, and ethernet frame to be sent is sent to the ether circuit according to the front and back order that each frame in the hardware transmit queue is arranged.
Further; The bandwidth allocation methods that present embodiment provides can also comprise: when the quantity of the to be sent ethernet frame of a user in this dispatching cycle of confirming is zero, in said this dispatching cycle, the ethernet frame in the corresponding output queue of said user is not dispatched.The i.e. analytical calculation of the length of ethernet frame in the output queue corresponding through bandwidth that the user is assigned with and this user; Can not find the n value that satisfies condition shown in the above-mentioned formula (1); The data volume of the ethernet frame that represent will to send in this dispatching cycle this moment is considerably beyond this user operable data volume in this dispatching cycle; Redispatch when having only dispatch the next time of wait, and in next dispatching cycle, each parameter in the formula (1) is respectively as shown in the formula shown in (2) and (3):
It is to be noted; The step 201-207 of foregoing description is the data dispatch process of transmitting in dispatching cycle in the present embodiment; When user data sends on to whole ether circuit; Need repeat above-mentioned steps 201-207, to accomplish the transmission of all data through a plurality of dispatching cycles.After the scheduling of the ethernet frame of accomplishing this dispatching cycle is sent, can get access to parameter s
3, s
4Value, use when being used for calculating n next dispatching cycle.Wherein, The ethernet frame that user 1 can send be numbered 1,2,5; Obtain on physical circuit, taking the actual physical bandwidth through its frame length calculations and be ((64+24)+(70+24)+(120+24)) * 8=2608bit; Surpass the user 1 actual 48bit that allows the data volume 2560bit of transmission, i.e. s in this dispatching cycle
3=48bit, s
4=0.The numbering of the ethernet frame that user 2 can send is 3,4, on physical circuit, takies the actual physical bandwidth and is ((80+24)+(75+24)) * 8=1624bit, surpasses the user 2 actual 24bit that allows the data volume 1600bit of transmission, i.e. s in this dispatching cycle
3=24bit, s
4=0.The numbering of the ethernet frame that user 3 can send is 11, on physical circuit, takies actual physical bandwidth ((90+24)) * 8=912bit, surpasses the user 3 actual 272bit that allows the data volume 640bit of transmission, i.e. s in this dispatching cycle
3=272bit, s
4=0.The numbering of the ethernet frame that user 4 can send is 16,18,19,29,21; On physical circuit, take actual physical bandwidth ((65+24)+(90+24)+(100+24)+(110+24)+(111+24)) * 8=4768bit; Lack 32bit than user 4 actual data volume 4800bit that sends that allows in this dispatching cycle, i.e. s
3=0, s
4=32bit.
Fig. 6 is for bandwidth allocation sketch map three among the bandwidth allocation methods embodiment two of the present invention, and is as shown in Figure 6, and when dispatching for the third time, the record current system time is divided into the frame of output queue respectively in the corresponding output queue of each user.Time interval of system time of supposing system time and a preceding dispatching record of current record is 155us; Then can calculate the data volume that user 1 can send in this dispatching cycle is 16*155-48=2480-48=2432bit; The data volume that user 2 can send in this dispatching cycle is 10*155-24=1550-24=1526bit; The data volume that user 3 can send in this dispatching cycle is 4*155-272=620-272=348bit, and the data volume that user 4 can send in this dispatching cycle is 40*155+32=4650+32=4682bit.
Fig. 7 is bandwidth allocation sketch map four among the bandwidth allocation methods embodiment two of the present invention; As shown in Figure 7, in the output queue of user's 1 correspondence, suppose that frame 6 length are 90 bytes; The length of frame 7 is 80 bytes; The length of frame 25 is 120 bytes, and frame 28 length are 110 bytes, and the length of frame 31 is 230 bytes.In the output queue of user's 2 correspondences, suppose that frame 8 length are 80 bytes, frame 9 length are 85 bytes, and frame 10 length are 170 bytes, and frame 32 length are 120 bytes, and frame 33 length are 710 bytes.In the output queue of user's 3 correspondences, suppose that frame 12 is 100 bytes, frame 13 is 190 bytes, and frame 14 is 700 bytes, and frame 15 length are 210 bytes, and frame 17 length are 90 bytes, frame 17 length are 76 bytes.In the output queue of user's 4 correspondences, suppose that frame 22 length are 590 bytes, frame 23 length are 190 bytes; Frame 24 length are 180 bytes, and frame 27 length are 220 bytes, and frame 29 length are 124 bytes; Frame 30 length are 106 bytes, and frame 35 length are 170 bytes, and frame 36 length are 490 bytes; Frame 37 length are 690 bytes, and frame 38 length are 890 bytes.Suppose s in this dispatching cycle
5Can value be 500bit; Obtain through analytical calculation; The frame number that user 1 can send is 1,2,5; It takies the actual physical bandwidth and is ((90+24)+(80+24)+(120+24)) * 8=2896bit on physical circuit, surpass user 1 actual 464bit that allows the data volume of transmission in this dispatching cycle.The frame number that user 2 can send is 8,9, on physical circuit, takies the actual physical bandwidth and is ((80+24)+(85+24)) * 8=1704bit, surpasses user 1 actual 178bit that allows the data volume of transmission in this dispatching cycle.User's 3 these scheduling can not transmit frame; Because this frame 12 that sends among the user 3 needs bandwidth 992bit; And the bandwidth that user 3 allowed to use in this dispatching cycle has only 348bit, can know 992＞(348+500) according to aforementioned calculation formula (1), the n value that does not then satisfy condition; Therefore when having only scheduling such as the next time of grade, ability transmit frame 12; Because this scheduling do not have transmit frame, so this scheduling saved the bandwidth of 348bit, can in scheduling next time, use.The frame number that user 4 can send is 22, on physical circuit, takies the actual physical bandwidth and is ((590+24)) * 8=4912bit, surpasses actual bandwidth 230bit.
Present embodiment provides a kind of bandwidth allocation methods; According to user characteristics the ethernet frame in the output queue on the Ethernet interface is divided in the corresponding output queue of each user; The length of each ethernet frame in the bandwidth that is assigned with according to the user and the corresponding output queue; Confirm user's ethernet frame to be sent, and each user's ethernet frame to be sent joined in the hardware transmit queue send to the ether circuit; Present embodiment carries out simple analysis through the length of ethernet frame in bandwidth that the user is assigned with and the output queue and calculates; Just can realize the accurate control of the band width in physical that the data to different user take on the ether circuit; The allocated bandwidth mechanism of present embodiment is fairly simple, and is less to the calculation resources consumption of system.
One of ordinary skill in the art will appreciate that: all or part of step that realizes above-mentioned each method embodiment can be accomplished through the relevant hardware of program command.Aforesaid program can be stored in the computer read/write memory medium.This program the step that comprises above-mentioned each method embodiment when carrying out; And aforesaid storage medium comprises: various media that can be program code stored such as ROM, RAM, magnetic disc or CD.
Fig. 8 is the structural representation of allocated bandwidth device embodiment one of the present invention, and as shown in Figure 8, present embodiment provides a kind of allocated bandwidth device, can specifically carry out each step among the said method embodiment one, repeats no more here.The allocated bandwidth device that present embodiment provides can specifically comprise sort module 801, computing module 802 and sending module 803.Wherein, sort module 801 is used for according to user characteristics the ethernet frame of output queue on the Ethernet interface being classified, and generates the corresponding output queue of each user.The length of each ethernet frame of output queue that the bandwidth that computing module 802 is used for being assigned with according to said each user and said each user are corresponding is confirmed said each user's ethernet frame to be sent respectively.Sending module 803 is used for successively ethernet frame to be sent with said each user and joins the hardware transmit queue and send to the ether circuit.
Fig. 9 is the structural representation of allocated bandwidth device embodiment two of the present invention, and as shown in Figure 9, present embodiment provides a kind of allocated bandwidth device, can specifically carry out each step among the said method embodiment two, repeats no more here.The allocated bandwidth device that present embodiment provides is on above-mentioned basis shown in Figure 8, and computing module 802 can comprise specifically that first computing unit 812, second computing unit 822 and frame confirm unit 832.Wherein, first computing unit 812 is used for calculating this scheduling time at interval according to the system time of this scheduling of writing down and the system time of last scheduled.Second computing unit 822 be used for according to said this scheduling time at interval and the bandwidth that is assigned with of said each user calculate the data volume that said each user can send in this dispatching cycle respectively.Frame is confirmed the length of each ethernet frame in the corresponding output queue of data volume that unit 832 is used for sending in this dispatching cycle according to said each user and said each user, confirms said each user's ethernet frame to be sent respectively.
More specifically, the frame in the present embodiment confirms that unit 832 can comprise specifically that obtaining subelement 8321, number of frames confirms that subelement 8322 and frame confirm subelement 8323.Wherein, Obtain subelement 8321 and be used for obtaining respectively the corresponding data volume correction value of said each user, the corresponding data volume correction value of said each user is respectively the difference of the data volume that data volume that said each user sends in the cycle in last scheduled and said user can send in the last scheduled cycle.Number of frames is confirmed the length of each ethernet frame in the corresponding output queue of the corresponding data volume correction value of data volume, said each user that subelement 8322 is used for sending in this dispatching cycle according to said each user, preset data volume preset value and each user, confirms the quantity of the to be sent ethernet frame of said each user in this dispatching cycle respectively.Frame confirms that subelement 8323 is used for the ethernet frame to be sent quantity definite said each user from the corresponding output queue of said each user of the said ethernet frame to be sent of this dispatching cycle according to said each user.Wherein, the difference and the said data volume preset value sum of the data volume that in this dispatching cycle, can send less than the user of the data volume of a user's the ethernet frame to be sent data volume correction value corresponding with said user.
Further; The allocated bandwidth device that present embodiment provides can also comprise dispatching control module 804; Dispatching control module 804 be used for when a user who confirms when the quantity of the ethernet frame to be sent of this dispatching cycle is zero, in said this dispatching cycle, the ethernet frame in the corresponding output queue of said user is not dispatched.
Present embodiment provides a kind of allocated bandwidth device; According to user characteristics the ethernet frame in the output queue on the Ethernet interface is divided in the corresponding output queue of each user; The length of each ethernet frame in the bandwidth that is assigned with according to the user and the corresponding output queue; Confirm user's ethernet frame to be sent, and each user's ethernet frame to be sent joined in the hardware transmit queue send to the ether circuit; Present embodiment carries out simple analysis through the length of ethernet frame in bandwidth that the user is assigned with and the output queue and calculates; Just can realize the accurate control of the band width in physical that the data to different user take on the ether circuit; The allocated bandwidth mechanism of present embodiment is fairly simple, and is less to the calculation resources consumption of system.
Present embodiment also provides a kind of network equipment, for example switch, router etc., and this network equipment can specifically comprise above-mentioned Fig. 8 or allocated bandwidth device shown in Figure 9.
What should explain at last is: above each embodiment is only in order to explaining technical scheme of the present invention, but not to its restriction; Although the present invention has been carried out detailed explanation with reference to aforementioned each embodiment; Those of ordinary skill in the art is to be understood that: it still can be made amendment to the technical scheme that aforementioned each embodiment put down in writing, perhaps to wherein part or all technical characteristic are equal to replacement; And these are revised or replacement, do not make the scope of the essence disengaging various embodiments of the present invention technical scheme of relevant art scheme.
Claims (9)
1. a bandwidth allocation methods is characterized in that, comprising:
According to user characteristics the ethernet frame in the output queue on the Ethernet interface is classified, and generate the corresponding output queue of each user;
The length of each ethernet frame in the corresponding output queue of the bandwidth that is assigned with according to said each user and said each user is confirmed said each user's ethernet frame to be sent respectively;
Successively said each user's ethernet frame to be sent is joined in the hardware transmit queue and send to the ether circuit.
2. method according to claim 1 is characterized in that, the length of each ethernet frame in the corresponding output queue of the said bandwidth that is assigned with according to said each user and said each user confirms that respectively said each user's ethernet frame to be sent comprises:
Calculate this scheduling time at interval according to the system time of this scheduling of writing down and the system time of last scheduled;
According to said this scheduling time at interval and the bandwidth that is assigned with of said each user calculate the data volume that said each user can send respectively in this dispatching cycle;
The length of each ethernet frame in the corresponding output queue of the data volume that in this dispatching cycle, can send according to said each user and said each user is confirmed said each user's ethernet frame to be sent respectively.
3. method according to claim 2; It is characterized in that; The length of each ethernet frame in the corresponding output queue of the said data volume that in this dispatching cycle, can send according to said each user and said each user, confirm that respectively said each user's ethernet frame to be sent comprises:
Obtain the corresponding data volume correction value of said each user respectively, the corresponding data volume correction value of said each user is respectively the difference of the data volume that data volume that said each user sends in the cycle in last scheduled and said user can send in the last scheduled cycle;
The length of each ethernet frame in the corresponding output queue of the corresponding data volume correction value of the data volume that in this dispatching cycle, can send according to said each user, said each user, preset data volume preset value and said each user is confirmed the quantity of the to be sent ethernet frame of said each user in this dispatching cycle respectively;
According to the quantity of the said to be sent ethernet frame of said each user in this dispatching cycle, from the corresponding output queue of said each user, confirm said each user's ethernet frame to be sent;
Wherein, the difference and the said data volume preset value sum of the data volume that in this dispatching cycle, can send less than said user of the data volume of a user's the ethernet frame to be sent data volume correction value corresponding with said user.
4. method according to claim 3; It is characterized in that; Also comprise: when the quantity of the to be sent ethernet frame of a user in this dispatching cycle of confirming is zero, in said this dispatching cycle, the ethernet frame in the corresponding output queue of said user is not dispatched.
5. an allocated bandwidth device is characterized in that, comprising:
Sort module is used for according to user characteristics the ethernet frame of output queue on the Ethernet interface being classified, and generates the corresponding output queue of each user;
Computing module, the length of each ethernet frame of output queue that the bandwidth that is used for being assigned with according to said each user and said each user are corresponding is confirmed said each user's ethernet frame to be sent respectively;
Sending module is used for successively ethernet frame to be sent with said each user and joins the hardware transmit queue and send to the ether circuit.
6. device according to claim 5 is characterized in that, said computing module comprises:
First computing unit is used for calculating this scheduling time at interval according to the system time of this scheduling of writing down and the system time of last scheduled;
Second computing unit, be used for according to said this scheduling time at interval and the bandwidth that is assigned with of said each user calculate the data volume that said each user can send in this dispatching cycle respectively;
Frame is confirmed the unit, and the length of each ethernet frame in the corresponding output queue of the data volume that is used for sending in this dispatching cycle according to said each user and said each user is confirmed said each user's ethernet frame to be sent respectively.
7. device according to claim 6 is characterized in that, said frame confirms that the unit comprises:
Obtain subelement; Be used for obtaining respectively the corresponding data volume correction value of said each user, the corresponding data volume correction value of said each user is respectively the difference of the data volume that data volume that said each user sends in the cycle in last scheduled and said user can send in the last scheduled cycle;
Number of frames is confirmed subelement; The length of each ethernet frame in the corresponding output queue of the corresponding data volume correction value of the data volume that is used for sending in this dispatching cycle according to said each user, said each user, preset data volume preset value and said each user is confirmed the quantity of the to be sent ethernet frame of said each user in this dispatching cycle respectively;
Frame is confirmed subelement, is used for the quantity at the said ethernet frame to be sent of this dispatching cycle according to said each user, from the corresponding output queue of said each user, confirms said each user's ethernet frame to be sent;
Wherein, the difference and the said data volume preset value sum of the data volume that in this dispatching cycle, can send less than said user of the data volume of a user's the ethernet frame to be sent data volume correction value corresponding with said user.
8. device according to claim 7 is characterized in that, also comprises:
Dispatching control module, be used for when a user who confirms when the quantity of the ethernet frame to be sent of this dispatching cycle is zero, in said this dispatching cycle, the ethernet frame in the corresponding output queue of said user is not dispatched.
9. a network equipment is characterized in that, comprises each described allocated bandwidth device among the claim 5-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110369294.1A CN102404214B (en) | 2011-11-18 | 2011-11-18 | Bandwidth distribution method, device and network equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110369294.1A CN102404214B (en) | 2011-11-18 | 2011-11-18 | Bandwidth distribution method, device and network equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102404214A true CN102404214A (en) | 2012-04-04 |
| CN102404214B CN102404214B (en) | 2014-06-04 |
Family
ID=45886017
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110369294.1A Active CN102404214B (en) | 2011-11-18 | 2011-11-18 | Bandwidth distribution method, device and network equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102404214B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103051567A (en) * | 2012-12-28 | 2013-04-17 | 武汉烽火网络有限责任公司 | Method for realizing non-loss interconnection between exchanged chips of different factories |
| CN112311695A (en) * | 2020-10-21 | 2021-02-02 | 中国科学院计算技术研究所 | On-chip bandwidth dynamic allocation method and system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1433192A (en) * | 2002-01-17 | 2003-07-30 | 三星电子株式会社 | Method and frame structure for realizing several functions in billion ethernet passive light network |
| US20070230492A1 (en) * | 2006-03-28 | 2007-10-04 | Fujitsu Limited | Frame multiplexing device |
| US20080062988A1 (en) * | 2006-09-12 | 2008-03-13 | Brian Daigle | Methods, computer program products, and modules for dynamically allocating bandwidth of a subscriber line |
| CN101465789A (en) * | 2007-12-17 | 2009-06-24 | 大唐移动通信设备有限公司 | Bandwidth distribution control method and device |
| CN100518083C (en) * | 2005-11-08 | 2009-07-22 | 中兴通讯股份有限公司 | A realization method to dynamically change maximum length of bandwidth request |
| CN101888342A (en) * | 2010-07-23 | 2010-11-17 | 中兴通讯股份有限公司 | Bandwidth distribution method and device |
-
2011
- 2011-11-18 CN CN201110369294.1A patent/CN102404214B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1433192A (en) * | 2002-01-17 | 2003-07-30 | 三星电子株式会社 | Method and frame structure for realizing several functions in billion ethernet passive light network |
| CN100518083C (en) * | 2005-11-08 | 2009-07-22 | 中兴通讯股份有限公司 | A realization method to dynamically change maximum length of bandwidth request |
| US20070230492A1 (en) * | 2006-03-28 | 2007-10-04 | Fujitsu Limited | Frame multiplexing device |
| US20080062988A1 (en) * | 2006-09-12 | 2008-03-13 | Brian Daigle | Methods, computer program products, and modules for dynamically allocating bandwidth of a subscriber line |
| CN101465789A (en) * | 2007-12-17 | 2009-06-24 | 大唐移动通信设备有限公司 | Bandwidth distribution control method and device |
| CN101888342A (en) * | 2010-07-23 | 2010-11-17 | 中兴通讯股份有限公司 | Bandwidth distribution method and device |
Non-Patent Citations (1)
| Title |
|---|
| 钟瑾，张家树: "《一种基于固定长度的EPON动态带宽分配方案》", 《中国数据通信》, vol. 5, no. 7, 29 August 2003 (2003-08-29) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103051567A (en) * | 2012-12-28 | 2013-04-17 | 武汉烽火网络有限责任公司 | Method for realizing non-loss interconnection between exchanged chips of different factories |
| CN112311695A (en) * | 2020-10-21 | 2021-02-02 | 中国科学院计算技术研究所 | On-chip bandwidth dynamic allocation method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102404214B (en) | 2014-06-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108965171B (en) | Industrial wireless WIA-PA network and time sensitive network conversion method and device | |
| CN104982006B (en) | For providing the system and method for software definition protocol stack | |
| CN109618375B (en) | UAV ad hoc network time slot scheduling method based on service priority and channel interruption probability | |
| CN102802266A (en) | Implementation method of high dynamic self-organization network high efficiency TDMA (time division multiple address) protocol | |
| Wang et al. | Low complexity multi-resource fair queueing with bounded delay | |
| CN103023822A (en) | Fixed-length time slot based dynamic channel allocation method for time division duplex/time division multiple access | |
| CN113950104A (en) | Satellite network service deterministic scheduling method based on dynamic period mapping | |
| EP2874361B1 (en) | Method and device for transmitting enhanced transmission selection standard configuration information | |
| CN102404214B (en) | Bandwidth distribution method, device and network equipment | |
| WO2023109445A1 (en) | Service scheduling method based on time trigger ethernet | |
| Nikolova et al. | Bonded deficit round robin scheduling for multi-channel networks | |
| Xia et al. | Heterogeneous slot scheduling for real-time industrial wireless sensor networks | |
| CN114285541A (en) | Time-triggered Ethernet schedule generation method based on time delay error | |
| CN109873723B (en) | On-demand bandwidth allocation method based on node service priority | |
| CN102377665B (en) | IPv6-based radio frequency identification (RFID) router quality of service (QoS) method | |
| EP4005293A1 (en) | Systems and methods for allocating timeslots for unicast transmissions | |
| CN103973590A (en) | QoS (Quality of Service) scheduling method and QoS scheduling device | |
| CN103023811B (en) | The switching method and device of queue scheduling strategy | |
| CN103053183B (en) | video multicast method and apparatus | |
| CN116723551A (en) | Time delay guaranteeing method for cloud PLC service under 5G-TSN architecture | |
| Ng et al. | Wireless Controller Area Network Using Token Frame Scheme | |
| Yang et al. | An Interruption-Resilient Service Priority-Oriented TDMA Scheme for NAANET | |
| Mary et al. | Real time analysis of wireless controller area network | |
| Iwasawa et al. | Flexible Time-Aware Shaper Scheduling System for a Multitenancy Environment | |
| Ashjaei | Extending FTT-SE Protocol for Multi-Master/Multi-Slave Networks |
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 |