CN1450754A - Scheduling method for data pocket - Google Patents

Scheduling method for data pocket Download PDF

Info

Publication number
CN1450754A
CN1450754A CN03118918A CN03118918A CN1450754A CN 1450754 A CN1450754 A CN 1450754A CN 03118918 A CN03118918 A CN 03118918A CN 03118918 A CN03118918 A CN 03118918A CN 1450754 A CN1450754 A CN 1450754A
Authority
CN
China
Prior art keywords
packet
data flow
data
service time
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
Application number
CN03118918A
Other languages
Chinese (zh)
Other versions
CN1206835C (en
Inventor
杨宗凯
刘彦
王玉明
范兵
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huazhong University of Science and Technology
Original Assignee
Huazhong University of Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Huazhong University of Science and Technology filed Critical Huazhong University of Science and Technology
Priority to CNB031189180A priority Critical patent/CN1206835C/en
Publication of CN1450754A publication Critical patent/CN1450754A/en
Priority to KR20040019627A priority patent/KR101037929B1/en
Priority to US10/822,145 priority patent/US20050147103A1/en
Application granted granted Critical
Publication of CN1206835C publication Critical patent/CN1206835C/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/50—Queue scheduling
    • H04L47/62—Queue scheduling characterised by scheduling criteria
    • H04L47/629—Ensuring fair share of resources, e.g. weighted fair queuing [WFQ]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
    • H04L47/2441—Traffic characterised by specific attributes, e.g. priority or QoS relying on flow classification, e.g. using integrated services [IntServ]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/28—Flow control; Congestion control in relation to timing considerations

Abstract

A dispatching method quantizes data stream speed rate and data packet length and divides data stream queue to be dispatched into two parts of new coming stream queue and backlog stream queue. Two parts of queue are dispatched uniformly by dispatcher with SEFF strategy. The new coming stream queue part can be simplified if the hardware resource is not enough as a data packet selected from the stream queue part of new coming and backlog is transmitted by dispatcher according to whether it is the first data packet it its belonging data stream or not then to be arranged at the new coming stream queue part or backlog stream queue aprt when a data packet reaches the dispatching node.

Description

A kind of dispatching method of packet
Technical field
The invention belongs to network communicating system and applied technical field, be specifically related to a kind of dispatching method of packet, it is particularly useful for the scheduling of packet in the router.
Background technology
Along with the continuous growth of Internet, multimedia service also is more and more widely used.Because multimedia service such as voice, visual very high to bandwidth and delay requirement, this just needs router to have efficient and the fast data bag dispatching method comes to guarantee for the user provides reliable peer-to-peer services quality (QOS).
A present class is based on dispatching algorithm such as WFQ, the WF of GPS model 2Q, WF 2Q+ etc. have obtained extensive studies.GPS (Generalized Processor Sharing) model is a kind of Utopian fluid model, and it is based on following two hypothesis: one, data packet length limitlessly detachable; Its two, all stream can be accepted service simultaneously.And in real system, the minimum unit of a dispatcher services is a packet, and once can only be a stream service, so the GPS model can not be realized.So Jon C.R.Bennett and Hui Zhang (see document J.Bennett and H.Zhang, Hierarchical packet fair queuing algorithms, In Proceedings of theACM-SIGCOMM ' 96, pages 143-156, Palo Alto, CA, August 1996) WF proposed 2The Q+ dispatching algorithm is used at real system simulating GPS model.Its basic thought is, each packet in the convection current is safeguarded a beginning service time and is finished service time, when scheduler is ready for sending a packet, at first need reciprocity packet to be scheduled to carry out qualification test, have only the beginning service time just can be by test less than the packet of the virtual time of system, select to have the minimum packet that finishes service time then and send in by testing data packet, this strategy is called SEFF (Smallest Eligible virtual Finish time First) selection strategy.
Because WF 2Q+ have good fairness and the time ductility, and algorithm complex is lower, has therefore obtained the extensive concern of industry.But, WF 2There are the following problems on using for the Q+ algorithm: 1) complexity of algorithm increases and increases along with the quantity that is subjected to Scheduling Flow, particularly on the high-speed core router, when the quantity of data flow was very big, the application of algorithm will bring huge burden, 2 to system) on hardware, realize very difficult.
Summary of the invention
The object of the present invention is to provide a kind of dispatching method that can overcome the packet of above-mentioned defective, this method is simple, efficient, is convenient to hardware and realizes, and guaranteed WF substantially 2The performance of Q+ algorithm.
For achieving the above object, a kind of dispatching method of packet is divided into new flow queue and overstocked flow queue two parts of arriving with packet to be scheduled, is dispatched by scheduler, and its scheduling step is:
(1) initialization scheduling node, the initial value of initialization system virtual time;
(2) when certain packet arrives scheduling node, check whether this packet is number under it
According to first packet of stream, if be suspended to newly according to its speed and/or length and arrive
The corresponding Q of flow queue part 1Rear of queue, and, calculate this number by formula (3)
According to the empty initial service time of bag, this void initial service time i.e. this data flow empty initial
Service time; If it is this packet is not first packet of this data flow, straight
Connect the afterbody that this packet is hung over affiliated data flow;
When (3) dispatching, scheduler can scan a packet of first data flow in all formations
Empty initial service time, therefrom find out empty initial service time less than the system virtualization time
Legal data packet, again by formula (4), the void of calculating these legal data packet finishes
Service time, find out and have the minimum empty packet that finishes service time and send;
(4) process that a selected packet is sent is: at first with this packet from affiliated
Take off transmission among the data flow F, data flow F is again according to the speed of its new packet
Rate and length hang over and overstock the corresponding Q of flow queue part 2(R x, L y) rear of queue,
By formula (2) empty initial service time of new data stream F more, i.e. empty initial service time of the packet that data flow F is new, and by formula (1) update system virtual time; V ( t + τ ) = max ( V ( t ) + τ , mi n i ∈ B ( t ) ( S i h i ( t ) ) ) . . . . ( 1 ) F i k = S i k + L i k R i ( t ) . . . . . . . ( 4 )
(5) repeating step (2) is finished until work to (4).
In order to verify the performance of this dispatching method, in Network Simulator simulated environment, realized this dispatching method, and checked following performance index:
Figure A0311891800054
Bandwidth: the actual bandwidth that each data flow obtains (Mbps)
Figure A0311891800055
Bandwidth shake: poor (Mbps) of the average actual bandwidth that each data flow obtained in the adjacent time period
Time delay: the time difference that the packet of each data flow leaves and arrives (ms)
Figure A0311891800057
Delay variation: the average delay of each former and later two packet of data flow poor (ms)
It the results are shown in the embodiment part.In a word, this dispatching method is simply efficient, is easy to hardware and realizes, and guaranteed WF substantially 2The original premium properties of Q+ algorithm.
Description of drawings
Fig. 1 is the schematic diagram of dispatching method of the present invention;
Fig. 2 is the implementation framework of dispatching method;
Fig. 3 is for simplifying the implementation framework of dispatching method;
Fig. 4 is the emulation topological diagram;
Fig. 5 is the bandwidth characteristic of data flow 1 when using dispatching method;
Fig. 6 is the bandwidth jittering characteristic of data flow 1 when using dispatching method;
Fig. 7 is the delay character of data flow 1 when using dispatching method;
Fig. 8 is the delay variation characteristic of data flow 1 when using dispatching method;
The bandwidth characteristic of data flow 1 when Fig. 9 simplifies dispatching method for using;
The bandwidth jittering characteristic of data flow 1 when Figure 10 simplifies dispatching method for using;
The delay character of data flow 1 when Figure 11 simplifies dispatching method for using;
The delay variation characteristic of data flow 1 when Figure 12 simplifies dispatching method for using.
Embodiment
For the ease of following description, define some symbol descriptions and formula earlier:
Symbol Definition
??V(t) The system virtualization function of time
??S i k The empty initial service time of k packet of i data flow
??F i k The void of k packet of i data flow finishes service time
??τ The time interval that the system virtualization time upgrades
??B(t) The set of medium all streams to be scheduled of etching system during t
??h i(t) T is a packet of i data flow constantly
??Q i I the current packet number to be scheduled of data flow
??a i k The time of advent of k packet of i data flow
??L i k K length of data package of i data flow
??R i(t) T is the speed of i data flow constantly
V ( t + τ ) = max ( V ( t ) + τ , min i ∈ B ( t ) ( S i h i ( t ) ) ) . . . . ( 1 ) F i k = S i k + L i k R i ( t ) . . . . . . . . . ( 4 )
Below in conjunction with accompanying drawing the present invention is described in detail:
In dispatching method of the present invention, at different data flow, adopt speed that it is quantized, its quantification gradation number is M, is designated as R successively 1..., R MAt the different pieces of information bag in the data flow, adopt length that it is quantized, its quantification gradation number is N, is designated as L successively 1..., L N(for different data flow, its length quantification gradation number can be different, also can be identical, in order to simplify realization, adopt identical length quantification gradation number).R can obtain different formations with the combination of L, and each formation is designated as Q (R m, L n), so just total M * N formation.
Each data flow according to the speed R of its reservation (as R=R m) and a data packet length L (as L N-1＜L≤L n), hang over corresponding formation Q (R m, L n) on.Packet is meant the current foremost packet to be scheduled that comes in the data flow, and its length is quantified as than its big nearest level length that waits; Greater than highest ranking length, be quantified as highest ranking length.Data flow in the formation is designated as F successively 1, F 2..., last data flow is designated as F TailData flow F iIn packet be designated as P successively I1, P I2..., last data packet recording is P I, Tail
This data flow that is meant first packet of data flow take place for the first time or stop a period of time after processed at first packet when taking place again.Confirm one newly to packet do not have packet waiting for scheduling for the foundation of first packet of data flow under it is that data flow is current under this packet, if data flow data with existing bag is being waited for scheduling under this packet, then claim the follow-up data bag of this packet for data flow under it.Because the calculating of the empty initial service time of first packet of data flow and its follow-up packet is inconsistent, need be to first packet individual processing of this data flow.So we are divided into two parts with formation, newly arrive flow queue part 1. with overstocked flow queue part 2., as shown in Figure 1.New 1. first packet of responsible data streams of flow queue part that arrives overstocks the 2. follow-up data bag of responsible data streams of flow queue part, and two parts need 2 * M * N formation altogether.Scheduler effect 3. is to adopt the SEFF strategy that scheduling is unified in this two parts formation.
In conjunction with Fig. 2, the concrete scheduling step of this dispatching method is as follows:
(1) initialization scheduling node, the initial value of initialization system virtual time is such as 0;
(2) when certain packet arrives scheduling node, check whether this packet is number under it
According to first packet of stream, if be suspended to new to stream team according to its speed and length
The 1. corresponding Q of row part 1(R m, L n) rear of queue, and, calculate this by formula (3)
The empty initial service time of individual packet, this void initial service time also is this data flow
Empty initial service time.If this packet is not first data of this data flow
Bag, the afterbody of data flow under directly this packet being hung over;
When (3) dispatching, 3. scheduler can scan all formations and (comprise M * N the new stream team that arrives
Row Q 1(R, L) and M * N overstocked flow queue Q 2First data flow in (R, L))
The empty initial service time of packet, therefrom find out empty initial service time less than system
The legal data packet of virtual time by formula (4), is calculated these legal data packet again
Void finish service time, find out and have the minimum empty packet that finishes service time and carry out
Send;
(4) no matter be selected new to flow queue Q 1(R m, L n) a packet of first data flow advance
Row sends, the still selected flow queue Q that overstocks 2(R m, L n) a number of first data flow
Send according to bag, all make such processing: at first with this packet from affiliated number
Take off transmission according to stream among the F, data flow F is again according to the speed of its new packet
And length, hang over and overstock the 2. corresponding Q of flow queue part 2(R x, L y) rear of queue,
By formula (2) empty initial service time of new data stream F more, promptly data flow F is new
The empty initial service time of packet, and by formula (1) update system virtual time;
(5) repeating step (2) is finished until work to (4).
The scheduling node of indication of the present invention has promptly been used the equipment of this dispatching method, as router.
In order to reduce number of queues, save hardware resource, we have proposed the simplification dispatching method simultaneously.As shown in Figure 3, for newly 1. only adopting length (or speed) to quantize, be designated as Q to the flow queue part 1(L n) (or Q 1(R m)), need M * N+N (or M * N+M) individual formation so altogether.Below will be quantified as example with length.
In conjunction with Fig. 3, the concrete scheduling step of simplifying dispatching method is as follows:
(1) initialization scheduling node, the initial value of initialization system virtual time is such as 0;
(2) when certain packet arrives scheduling node, check whether this packet is number under it
According to first packet of stream, if be suspended to new according to its length to the flow queue part
1. corresponding Q 1(L n) rear of queue, and, calculate this packet by formula (3)
Empty initial service time, this void initial service time also is the empty initial clothes of this data flow
The affair time.If this packet is not first packet of this data flow, directly
This packet is hung over the afterbody of affiliated data flow.
When (3) dispatching, 3. scheduler can scan all formations and (comprise N the new flow queue that arrives
Q 1(L) and M * N overstocked flow queue Q 2A data of first data flow in (R, L))
The empty initial service time of bag, therefrom find out empty initial service time during less than system virtualization
Between legal data packet, again by formula (4), calculate the void knot of these legal data packet
Bundle service time, find out and have the minimum empty packet that finishes service time and send.
(4) no matter be selected new to flow queue Q 1(L n) a packet of first data flow carry out
Send the still selected flow queue Q that overstocks 2(R m, L n) a data of first data flow
The bag send, all make such processing: at first with this packet from affiliated data
Stream is taken off transmission among the F, data flow F again according to the speed of its new packet and
Length hangs over and overstocks the 2. corresponding Q of flow queue part 2(R x, L y) rear of queue, logical
Cross formula (2) the more empty initial service time of new data stream F, the i.e. new head of data flow F
The empty initial service time of packet, and by formula (1) update system virtual time;
(5) repeating step (2) is finished until work to (4).
Concrete emulation topological structure as shown in Figure 4.
In this emulation topological structure, each input link and output link bandwidth are 10M, and all data flow all are subjected to the scheduling of scheduling node and export from output link.At the scheduling node of Fig. 4, the speed and the length of dispatching method and simplification dispatching method all are quantized into 5 grades.
To 5 speed grade of stream data definition:
Figure A0311891800101
?0.1Mbps.
Figure A0311891800102
?0.3Mbps.
Figure A0311891800103
?1?Mbps.
Figure A0311891800104
?2?Mbps.
?5?Mbps.
To 5 length scale of data package definition:
Figure A0311891800106
?200?bytes.
Figure A0311891800107
?400?bytes.
Figure A0311891800108
?800?bytes.
?1000?bytes.
Figure A03118918001010
?1600?bytes.
To two kinds of transmission meanss of stream data definition:
CBR (Const Bit Rate): show that data flow is with constant speed rates;
On/Off: show that data flow is with certain speed intermittence ground transmission.
The data packet length is defined three kinds of distribution modes:
Equal length: all data packet lengths in the data flow are identical;
Be evenly distributed: data packets in data flows length evenly distributes in certain length range;
Figure A03118918001015
Normal distribution: data packets in data flows length is the center normal distribution with certain length.
Two kinds of dispatching methods have been carried out substantive test, chosen typical test configurations with the explanation test result.The exemplary simulation parameter of dispatching method is as shown in the table:
Stream identification Link bandwidth Reserved bandwidth Data flow transmission mode and speed Data packet length
????1 ??10Mbps ?5.0Mbps ???50％ ?On/Off ???5.0Mbps Normal distribution
????2 ??10Mbps ?2.0Mbps ???20％ ?On/Off ???2.0Mbps Normal distribution
????3 ??10Mbps ?1.0Mbps ???10％ ?On/Off ???1.0Mbps Normal distribution
????4 ??10Mbps ?1.0Mbps ???10％ ?On/Off ???1.0Mbps Normal distribution
????5 ??10Mbps ?0.3Mbps ???3％ ?On/Off ???0.3Mbps Normal distribution
????6 ??10Mbps ?0.3Mbps ???3％ ?On/Off ???0.3Mbps Normal distribution
????7 ??10Mbps ?0.1Mbps ???1％ ?On/Off ???0.1Mbps Normal distribution
????8 ??10Mbps ?0.1Mbps ???1％ ?On/Off ???0.1Mbps Normal distribution
????9 ??10Mbps ?0.1Mbps ???1％ ?On/Off ???0.1Mbps Normal distribution
????10 ??10Mbps ?0.1Mbps ???1％ ?On/Off ???0.1Mbps Normal distribution
The scheduling outlet ???????????????????????????10Mbps
Each data flow all adopts the On/Off model in the above-mentioned configuration, and data packets in data flows length adopts normal distribution (average is 1000 bytes, variance 400 bytes).Dispatching method and simplification dispatching method are carried out emulation experiment, obtain the performance index of data flow 1 in two kinds of methods: bandwidth, bandwidth shake, time delay and delay variation.As Fig. 5～shown in Figure 12.
According to Fig. 5～simulation result shown in Figure 12, as can be seen: these two kinds of dispatching methods all can guarantee the performances such as bandwidth, bandwidth shake, time delay and delay variation of data flow 1 preferably, thereby can guarantee quality of services for users preferably.
On the basis of NS emulation experiment, the FPGA of employing Xilinx company has realized supporting to simplify the scheduling chip of dispatching method.Maximum 128k data flow, 5 speed grade and 5 length scale, but and the value of each speed of dynamic-configuration and length scale supported of this chip.Through actual motion test, this chip can guarantee well each data flow reserved bandwidth, the time ductility and fairness.

Claims (1)

1. the dispatching method of a packet, this method are divided into packet to be scheduled newly to be dispatched by scheduler to flow queue and overstocked flow queue two parts, and its scheduling step is:
(1) initialization scheduling node, the initial value of initialization system virtual time;
(2) when certain packet arrives scheduling node, check whether this packet is first packet of data flow under it, if be suspended to new to the corresponding Q of flow queue part according to its speed and/or length 1Rear of queue, and by formula (3), the empty initial service time of calculating this packet, this void initial service time i.e. the empty initial service time of this data flow; If this packet is not first packet of this data flow, the afterbody of data flow under directly this packet being hung over;
When (3) dispatching, scheduler can scan the empty initial service time of a packet of first data flow in all formations, therefrom find out the legal data packet of empty initial service time less than the system virtualization time, again by formula (4), the void of calculating these legal data packet finishes service time, finds out to have the minimum empty packet that finishes service time and send;
(4) process that a selected packet is sent is: at first this packet is taken off transmission from affiliated data flow F, data flow F hangs over and overstocks the corresponding Q of flow queue part again according to the speed and the length of its new packet 2(R x, L y) rear of queue, by formula (2) empty initial service time of new data stream F more, i.e. empty initial service time of the packet that data flow F is new, and by formula (1) update system virtual time; V ( t + τ ) = max ( V ( t ) + τ , min i ∈ B ( t ) ( S i h i ( t ) ) ) . . . . ( 1 )
Figure A0311891800022
F i k = S i k + L i k R i ( t ) . . . . . . . . . . ( 4 )
(5) repeating step (2) is finished until work to (4).
CNB031189180A 2003-04-11 2003-04-11 Scheduling method for data pocket Expired - Fee Related CN1206835C (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CNB031189180A CN1206835C (en) 2003-04-11 2003-04-11 Scheduling method for data pocket
KR20040019627A KR101037929B1 (en) 2003-04-11 2004-03-23 Method and Apparatus for packet scheduling
US10/822,145 US20050147103A1 (en) 2003-04-11 2004-04-12 Packet scheduling method and apparatus

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CNB031189180A CN1206835C (en) 2003-04-11 2003-04-11 Scheduling method for data pocket

Publications (2)

Publication Number Publication Date
CN1450754A true CN1450754A (en) 2003-10-22
CN1206835C CN1206835C (en) 2005-06-15

Family

ID=28684369

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB031189180A Expired - Fee Related CN1206835C (en) 2003-04-11 2003-04-11 Scheduling method for data pocket

Country Status (2)

Country Link
KR (1) KR101037929B1 (en)
CN (1) CN1206835C (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1331336C (en) * 2004-05-25 2007-08-08 华中科技大学 Rapid analyzing method for data pack
CN100418313C (en) * 2005-12-21 2008-09-10 中国科学院计算技术研究所 Band wide redistribution method for band width changing link layering sharing and management zone
CN101674482B (en) * 2009-09-25 2011-05-11 上海大学 Method for optimized dispatching of extension type video flow in partially observational Markovian decision process
CN105915425A (en) * 2016-06-16 2016-08-31 厦门理工学院 Vehicular CAN FD network-oriented signal grouping and packaging method

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100333475B1 (en) * 1999-07-23 2002-04-25 안병엽 Rate proportional self-clocked fair queueing apparatus and method for high-speed packet-switched networks
DE10085287T1 (en) 1999-12-08 2003-03-27 Univ British Columbia Scheduler for a weighted fair queue operation

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1331336C (en) * 2004-05-25 2007-08-08 华中科技大学 Rapid analyzing method for data pack
CN100418313C (en) * 2005-12-21 2008-09-10 中国科学院计算技术研究所 Band wide redistribution method for band width changing link layering sharing and management zone
CN101674482B (en) * 2009-09-25 2011-05-11 上海大学 Method for optimized dispatching of extension type video flow in partially observational Markovian decision process
CN105915425A (en) * 2016-06-16 2016-08-31 厦门理工学院 Vehicular CAN FD network-oriented signal grouping and packaging method
CN105915425B (en) * 2016-06-16 2017-05-03 厦门理工学院 Vehicular CAN FD network-oriented signal grouping and packaging method

Also Published As

Publication number Publication date
KR20040089475A (en) 2004-10-21
CN1206835C (en) 2005-06-15
KR101037929B1 (en) 2011-05-31

Similar Documents

Publication Publication Date Title
CN1509094A (en) Downstream data-pack dispatching system and method in mobile communication system
CN1921444A (en) Method for classified package dispatching and resource distributing based on service quality
CN101035078A (en) Token-based Internet traffic control method
CN1302643C (en) Apparatus for testing network equipment transmission performance and method thereof
US20100085923A1 (en) Scheduling of users on a shared radio resource using a combination of link quality and traffic information
CN101075963A (en) Method and device for controlling dynamically based on network QoS
CN1917708A (en) Reception contro method based on measurement and QoS in broadband radio access system
CN1885824A (en) Sorter realizing method for active queue management
CN1503474A (en) Wireless insertion network system, method, and control equipment
CN1777147A (en) Sharing cache dynamic threshold early drop device for supporting multi queue
CN104619034A (en) Real-time service-oriented packet scheduling method for LTE communication system
CN1933653A (en) Up bandwidth distributing method for wide band wireless access system
CN100338911C (en) Method and apparatus for dynamic bandwidth allocation in IP telecom network
CN107204930B (en) Token adding method and device
CN1450754A (en) Scheduling method for data pocket
CN1728677A (en) Method and system for carrying video stream media through network in multiple velocities
CN1642145A (en) IP transmitting method for supporting service quality expanding of crossover service zone
CN1773955A (en) Queue dispatching method and apparatus in data network
CN1889521A (en) Data dispatching method and system
CN1607767A (en) Flow-control method based on network processor
CN1688136A (en) Parallel flow dispatching method for 10G network performance testing system
US9128755B2 (en) Method and apparatus for scheduling resources in system architecture
CN1283121C (en) Resource allocation method and apparatus in wireless communication system
CN1956412A (en) Method for admitting controlling integral service model
CN1929452A (en) Method and device for flux plastic

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
C17 Cessation of patent right
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20050615

Termination date: 20110411