CN110098893A - Difference explicit feedback transfer control method towards IPv6 - Google Patents
Difference explicit feedback transfer control method towards IPv6 Download PDFInfo
- Publication number
- CN110098893A CN110098893A CN201811568103.2A CN201811568103A CN110098893A CN 110098893 A CN110098893 A CN 110098893A CN 201811568103 A CN201811568103 A CN 201811568103A CN 110098893 A CN110098893 A CN 110098893A
- Authority
- CN
- China
- Prior art keywords
- rate
- value
- increment
- flow volume
- tidal flow
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
- H04L1/0033—Systems modifying transmission characteristics according to link quality, e.g. power backoff arrangements specific to the transmitter
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
- H04L1/0036—Systems modifying transmission characteristics according to link quality, e.g. power backoff arrangements specific to the receiver
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
- H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
- H04L1/1607—Details of the supervisory signal
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L5/00—Arrangements affording multiple use of the transmission path
- H04L5/003—Arrangements for allocating sub-channels of the transmission path
- H04L5/0053—Allocation of signaling, i.e. of overhead other than pilot signals
- H04L5/0055—Physical resource allocation for ACK/NACK
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
- H04L69/22—Parsing or analysis of headers
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L2101/00—Indexing scheme associated with group H04L61/00
- H04L2101/60—Types of network addresses
- H04L2101/618—Details of network addresses
- H04L2101/659—Internet protocol version 6 [IPv6] addresses
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Quality & Reliability (AREA)
- Computer Security & Cryptography (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a kind of difference explicit feedback transfer control method towards IPv6, quickly and accurately feeds back the rate that interchanger distributes in real time using the method for difference.For the problem that the slow phenomenon of convergence rate of 1 feedback marks position in ecn (explicit congestion notification) ECN mechanism and the multidigit feedback method of other explicit feedback mechanisms cause control overhead excessive, the present invention feeds back the difference of present rate and targeted rate using 4 bit traffic classification fields of data packet head in IPv6 agreement.The present invention can be controlled the expense of very little using limited digit and distribute rate come quick accurate feedback, improved the convergence of transmission, reduced the stream deadline.
Description
Technical field
It is aobvious that the present invention relates to a kind of data center network (DCN, Data Center Network) difference towards IPv6
Formula feedback transmission control method.
Background technique
With mobile Internet, Internet of Things and the continuous development of cloud computing, mobile terminal device, network equipment quantity
It is continuously increased, IP address is increasingly deficient, is transitioned into IPv6 from IPv4 and has become trend of the times.Compared to IPv4, the flow control of IPv6
Ability processed is obviously improved, and longer data packet header length means stronger information carrying capacity.Wherein 8 bit
Traffic classes field also carry congestion information other than compatible ecn (explicit congestion notification) (ECN) using idle component, mention
Rise network transmission efficiency.
Existing data center's transmission control protocol (DCTCP) passes through to congestion information feedback and congestion window adjusting side
The improvement improving data transmission efficiency of method.DCTCP feeds back 1 to transmitting terminal by ecn (explicit congestion notification) (ECN) when congestion occurs
The congestion information of position, and according to congestion window size is adjusted, slow down data transmission rate, congestion is avoided to aggravate.However, existing rank
The research of the Duan Youguan program also rests under the IPv4 environment being widely used at present.How utilize IPv6 characteristic, adjust and
Improving control of the DCTCP to network transmission becomes important research to further promote the performance of transmission control protocol
Problem.
DCTCP utilize ecn (explicit congestion notification) algorithm ECN mechanism, congestion state is judged by router, data packet head into
The explicit congestion marking of row.After transmission end main frame receives the ACK with congestion feedback label of receiving end return, Hownet is obtained
Congestion is occurring in network, is reducing and sends sliding window size, avoid the generation of congestion collapse phenomenon.But the problem of ECN, is
Only indicate whether congestion occurs with 1 feedback marks position, it is difficult to fast and accurately feed back the rate letter that router distributes in real time
Breath.Other typical high speed transport protocols (such as XCP) based on Explicit Congestion notifying mechanism are by modification head-coating structure, using big
The marker bit of amount feeds back distribution rate, although greatly improving convergence rate, also results in excessive control overhead, difficult
With the application deployment in large scale network.
Therefore, the flow control capability for how utilizing IPv6 agreement designs Control with Explicit Rate Feedback method with accurate and quick
Ground feedback transmission rate reduces convergence time, and the performance for promoting transmission control protocol is a urgent problem to be solved.
Summary of the invention
How technical problem solved by the invention is using mentioning in the traffic classes field in the packet header under IPv6 agreement
Data transmission efficiency is risen, rate convergence time is reduced, effectively distributes network bandwidth, more reasonably utilizes Internet resources.
Technical solution of the present invention the following steps are included:
A kind of difference explicit feedback transfer control method towards IPv6, comprising the following steps:
Step 1: the current concurrent TCP flow number N of initialization real-time exchange machinec, fluxion statistics timer expiry threshold value, hand over
Change planes egress rate C, real time rate rc, tidal flow volume Δ, targeted rate rt, maximum rate increment rmWith increment remainder rr;
Transmitting terminal is operated according to the following steps:
Step 2-1: ACK grouping is received;
Step 2-2: reading the received packet header ACK encoded information, and decoding obtains tidal flow volume Δ;
Step 2-3: it is used as transmission rate after present rate is increased tidal flow volume Δ, and with the rate sending data point
Group;
Step 2-4: judging whether grouping is sent, if so, terminating, otherwise goes to step 2-1.
Interchanger is operated according to the following steps:
Step 3-1: data grouping is received;
Step 3-2: judging whether current time reaches the timeout threshold of fluxion statistics timer, and step is gone to if reaching
Rapid 3-3;Otherwise step 3-6 is gone to;
Step 3-3: real-time fluxion N is updatedc, update the real time rate r of all streamsc, and initialize adjustment rate raValue
For rc;
Step 3-4: the targeted rate r of every stream is calculatedt=C/Nc;
Step 3-5: targeted rate rtThe maximum rate increment r carried divided by each ACK groupingm, obtained quotient is rounded, surplus
Under remainder be increment remainder rr;
Step 3-6: judge the adjustment rate r that flow table where received data grouping is shownaWith targeted rate rtIt is whether equal,
If so, going to step 3-7;Otherwise, 3-8 is gone to step;
Step 3-7: the value of tidal flow volume Δ is set as 0, goes to step 3-10;
Step 3-8: judge raWhether r is greater thant, if so, the value of setting tidal flow volume Δ is-rm, adjust rate ra's
Value is ra-rm, go to step 3-10;Otherwise, step 3-9 is gone to;
Step 3-9: judge rt-raWhether r is greater thanm, if so, the value of setting tidal flow volume Δ is rm, adjust rate ra's
Value is ra+rm, go to step 3-10;Otherwise, then the value that tidal flow volume Δ is arranged is rr, adjust rate raValue be ra+rr, turn
To step 3-10;
Step 3-10: encoding Δ, and is written in data packet header, forwards data grouping, goes to step 3-1;
Receiving end is operated according to the following steps:
Step 4-1: data grouping is received;
Step 4-2: the encoded information of data packet header is read;
Step 4-3: the encoded information read is written in ACK packet header;
Step 4-4: it sends ACK and is grouped to transmitting terminal, go to step 4-1.
The difference explicit feedback transfer control method towards IPv6, in the step 1: by real-time fluxion NcJust
Beginning turns to 1, by real time rate rcIt is initialized as 0;1ms is set by fluxion statistics timer expiry threshold value;Interchanger is exported
Rate C is set as current outlet port band width;The value of tidal flow volume Δ is initialized as 0;By targeted rate rtIt is initialized as 0;It will
Maximum rate increment rmIt is set as 7;By increment remainder rrValue be initialized as 0.
The difference explicit feedback transfer control method towards IPv6, in the step 3-10: taking in packet header 8
First 4 in traffic classes field encode rate increment Δ, and the 1st is sign bit, and 0 indicates just, and 1 indicates negative, and rear 3
Totally 8 binary numbers, sequence correspond to 0-7 totally 8 decimal numbers, respectively indicate 0~7Mbps for position expression 000 to 111, finally,
The rate increment Δ that range is -7~7Mbps is encoded to 4 bits that range is 1111~0111.
The difference explicit feedback transfer control method towards IPv6, in the step 2-2: taking 8 bit stream in packet header
First 4 in amount classification field are decoded rate increment Δ, and the 1st is sign bit, and 0 indicates just, and 1 indicates negative, and latter 3
Totally 8 binary numbers, sequence correspond to 0-7 totally 8 decimal numbers, respectively indicate 0~7Mbps for expression 000 to 111, finally, will
4 bits that range is 1111~0111 are decoded as the rate increment Δ that range is -7~7Mbps.
The difference explicit feedback transfer control method towards IPv6, the step 3-7 is into 3-10 step: aobvious
The rate increment Δ of formula feedback is the changing value of rate adaptation, and interchanger carries rate increment using 4 bit fields of packets headers
Δ, and feed back to sender.Sender carrys out regulations speed according to present rate and rate increment.
The technical effects of the invention are that being worked as using 4 bit traffic classification fields of data packet head in IPv6 agreement to feed back
The difference of preceding rate and targeted rate.The present invention can control the expense of very little using limited digit come quick accurate feedback
Rate is distributed, the convergence of transmission is improved, reduces the stream deadline.
Detailed description of the invention
Fig. 1 is flow chart of the invention.
Fig. 2 is test scene topological diagram.
Fig. 3 is the transmission performance figure for being convergence rate.Fig. 3 (a) is the test result of ECN mechanism, and Fig. 3 (b) is DECN machine
The test result of system.The present invention is named as DECN.
Fig. 4 is the different stream deadline comparison diagrams concurrently flowed under quantity.When Fig. 4 (a) is that the stream under two kinds of mechanism is completed
Between comparison diagram, Fig. 4 (b) is 99 quantile stream deadline comparison diagrams under two kinds of mechanism.
Fig. 5 is the stream deadline comparison diagram under different RTT.Fig. 5 (a) is the mean flow deadline under two kinds of mechanism
Figure, Fig. 5 (b) are 99 quantile stream deadline comparison diagrams under two kinds of mechanism.
Fig. 6 is the stream deadline comparison diagram under different concurrent intensity.Fig. 6 (a) is the stream deadline under two kinds of mechanism
Comparison diagram, Fig. 6 (b) are 99 quantile stream deadline comparison diagrams under two kinds of mechanism.
Specific embodiment
The present invention will be further described below with reference to the drawings.
The present embodiment the following steps are included:
Step 1: the current concurrent TCP flow number N of initialization real-time exchange machinec, fluxion statistics timer expiry threshold value, hand over
Change planes egress rate C, real time rate rc, tidal flow volume Δ, targeted rate rt, maximum rate increment rmWith increment remainder rr;
Transmitting terminal is operated according to the following steps:
Step 2-1: ACK grouping is received;
Step 2-2: reading the received packet header ACK encoded information, and decoding obtains tidal flow volume Δ;
Step 2-3: it is used as transmission rate after present rate is increased tidal flow volume Δ, and with the rate sending data point
Group;
Step 2-4: judging whether grouping is sent, if so, terminating, otherwise goes to step 2-1.
Interchanger is operated according to the following steps:
Step 3-1: data grouping is received;
Step 3-2: judging whether current time reaches the timeout threshold of fluxion statistics timer, and step is gone to if reaching
Rapid 3-3;Otherwise step 3-6 is gone to;
Step 3-3: real-time fluxion N is updatedc, update the real time rate r of all streamsc, and initialize adjustment rate raValue
For rc;
Step 3-4: the targeted rate r of every stream is calculatedt=C/Nc;
Step 3-5: targeted rate rtThe maximum rate increment r carried divided by each ACK groupingm, obtained quotient is rounded, surplus
Under remainder be increment remainder rr;
Step 3-6: judge the adjustment rate r that flow table where received data grouping is shownaWith targeted rate rtIt is whether equal,
If so, going to step 3-7;Otherwise, 3-8 is gone to step;
Step 3-7: the value of tidal flow volume Δ is set as 0, goes to step 3-10;
Step 3-8: judge raWhether r is greater thant, if so, the value of setting tidal flow volume Δ is-rm, adjust rate ra's
Value is ra-rm, go to step 3-10;Otherwise, step 3-9 is gone to;
Step 3-9: judge rt-raWhether r is greater thanm, if so, the value of setting tidal flow volume Δ is rm, adjust rate ra's
Value is ra+rm, go to step 3-10;Otherwise, then the value that tidal flow volume Δ is arranged is rr, adjust rate raValue be ra+rr, turn
To step 3-10;
Step 3-10: encoding Δ, and is written in data packet header, forwards data grouping, goes to step 3-1;
Receiving end is operated according to the following steps:
Step 4-1: data grouping is received;
Step 4-2: the encoded information of data packet header is read;
Step 4-3: the encoded information read is written in ACK packet header;
Step 4-4: it sends ACK and is grouped to transmitting terminal, go to step 4-1.
Wherein in step 1, specific initialization includes: by real-time fluxion NcIt is initialized as 1, by real time rate rcInitially
Turn to 0;1ms is set by fluxion statistics timer expiry threshold value;Current exit port band is set by interchanger egress rate C
It is wide;The value of tidal flow volume Δ is initialized as 0;By targeted rate rtIt is initialized as 0;By maximum rate increment rmIt is set as 7;It will
Increment remainder rrValue be initialized as 0.
Carrying out coding to Δ in step 3-10 is: taking first 4 in packet header in 8 bit traffic classification fields to rate increment
Δ is encoded, and the 1st is sign bit, and 0 indicates just, and 1 indicates negative, and latter 3 indicate 000 to 111 totally 8 binary numbers, sequence
0-7 totally 8 decimal numbers are corresponded to, 0~7Mbps is respectively indicated, finally, the rate increment Δ that range is -7~7Mbps is encoded
For 4 bits that range is 1111~0111.
In step 2-2, decoding is the following steps are included: take first 4 in packet header in 8 bit traffic classification fields to increase rate
Amount Δ is decoded, and the 1st is sign bit, and 0 indicates just, and 1 indicates negative, and latter 3 indicate 000 to 111 totally 8 binary numbers, suitable
Ordered pair answers 0-7 totally 8 decimal numbers, respectively indicates 0~7Mbps, finally, 4 bits for being 1111~0111 by range
It is decoded as the rate increment Δ that range is -7~7Mbps.
Step 3-7 is into 3-10 step: the rate increment Δ of explicit feedback is the changing value of rate adaptation, interchanger benefit
Rate increment Δ is carried with 4 bit fields of packets headers, and feeds back to sender.Sender is according to present rate and rate increment
Carry out regulations speed.
Referring to Fig. 1, the flow chart describe transmitting terminal according to the present invention, interchanger and the parts of receiving end three
Treatment process.Its process is as follows:
It initializes first, by the current concurrent TCP flow number N of the real-time exchange machine of first round statisticscIt is set as 1, by first
Take turns the real time rate r of statisticscIt is set as 0;0 is set by the value of tidal flow volume Δ;The targeted rate r that initial calculation is obtainedt
It is set as 0;By maximum rate increment rm7 are set as, by initial gain amount remainder rrValue be set as 0.And it is fixed to set fluxion statistics
When device timeout threshold.Then judge operating main body type, there are three types of operating main body types respectively: transmitting terminal, interchanger and reception
End.
For transmitting terminal: receiving ACK grouping, read the received packet header ACK encoded information, decoding obtains tidal flow volume, will
Present rate is used as transmission rate after increasing tidal flow volume Δ, and with rate sending data grouping.In decoding, according to number
Tidal flow volume is decoded as according to binary number in the 4 bit traffic classification fields in packet header.
For interchanger: receiving data grouping, judge whether current time reaches the overtime threshold of fluxion statistics timer
Value, i.e. whether statistics fluxion timer is overtime, if overtime, updates real-time fluxion NcWith the real time rate r of all streamsc, and just
Beginningization adjusts rate raValue be rc, calculate the targeted rate r of every streamt=C/ Nc, targeted rate rtDivided by can carry every time most
Big rate increment rm, quotient's round numbers, aliquant remainder is as increment remainder rr, judge that flow table where received data grouping is shown
Adjustment rate raWith targeted rate rtIt is whether equal, point or less the processing of 2 kinds of situations:
(1) if the two is equal, the value of tidal flow volume Δ is 0, carries out coding to Δ and is written in packets headers, directly turns
Send out data grouping.
(2) if the two etc., judge raWhether r is greater thantIf raGreater than rt, then the value that tidal flow volume Δ is arranged is-rm,
Adjust rate raValue be ra-rm, Δ is encoded, and is written in packets headers, forwards data grouping.If raLess than rt,
Then judge rt-raWhether r is greater thanmIf: rt-raGreater than rm, then the value that tidal flow volume Δ is arranged is rm, adjust rate raValue be
ra+rm, Δ is encoded, and is written in packets headers, forwards data grouping;Otherwise, then the value that tidal flow volume Δ is arranged is
rr, adjust rate raValue be ra+rr, Δ is encoded, and is written in packets headers, forwards data grouping.
When wherein carrying out binary coding to rate increment Δ, the 1st is sign bit, and 0 indicates just, and 1 indicates negative.3 afterwards
It can indicate 000 to 111 totally 8 binary numbers, the corresponding 0-7 of sequence totally 8 decimal numbers respectively indicate 0~7Mbps.Finally,
The rate increment Δ that range is -7~7Mbps is encoded to 4 bits that range is 1111~0111.
For receiving end: receiving data grouping, read the encoded information of data packet header, and be written in the packet header ACK
Encoded information finally sends ACK packet to transmitting terminal.
The present invention is realized using NS2.35 network simulation platform, and has carried out performance test.
Fig. 2 is test scene topological diagram.More transmitting terminal servers are by 1 interchanger to 1 receiving end server
Send data flow.Every 1 server sends a DCTCP stream respectively.The cache size of interchanger is set as 128 groupings,
ECN marking-threshold is set as 65 groupings.Link bandwidth is set as 10Gbps.
Fig. 3 is the transmission performance figure for being convergence rate.Experiment topology is as shown in Fig. 2, propagation delay RTT is set as 100 μ
S, tests 2 long streams, and the data volume of 2 stream is infinity.It is sent wherein 1 stream is since experiment i.e., in addition 1 stream
Started to send at 1 second.Now compare and ECN (display congestion feedback) and DECN proposed by the present invention is respectively adopted (difference, which is shown, gathers around
Plug feedback) throughputs of two stream under 2 kinds of mechanism.
It can be seen that ECN can help 2 stream to converge to the rate of fair bandwidth, but its convergence time is long in Fig. 3 (a)
Of about 50ms.And convergence is ineffective, and the throughput of 2 streams constantly vibrates, it is difficult to stablize.In Fig. 3 (b), DECN is helped
2 stream throughputs converge to rapidly fair bandwidth, and convergence time is by about 0.3ms, and throughput vibrates very little.
Fig. 4 is the different stream deadline comparison diagrams concurrently flowed under quantity.Experiment topology is as shown in Fig. 2, propagation delay
RTT is set as 100 μ s.Every stream starts to send with 200 μ s of interval, and fluxion increases to 100 from 10,20, the data volume of every stream
It is set as 165 data packets.Count two methods of mean flow deadline (AFCT) under ECN and DECN and 99 quantiles
Flow the deadline.
Fig. 4 (a) is the equal deadline comparison diagram of levelling under two kinds of mechanism.It can be seen that as fluxion increases, stream
Average completion time is increasing, and the AFCT of ECN is higher by about 30% than DECN always.Fig. 4 (b) is 99 under two kinds of mechanism
Quantile stream deadline comparison diagram, it can be seen that the 99 quantile stream deadlines of ECN compare DECN always under two kinds of mechanism
Want high, this illustrates that DECN significantly reduces the deadline of hangover stream.
Fig. 5 is the stream deadline comparison diagram under different RTT.Experiment topology is as shown in Fig. 2, fixed concurrent fluxion is 50
Item, changing RTT is 0.1ms, 1ms, 10ms, 100ms, 1s, and the data volume of every stream is set as 165 data packets, compares ECN
With two kinds of mechanism of DECN.
Fig. 5 (a) is the mean flow deadline figure under two kinds of mechanism, and wherein ordinate uses index coordinates.RTT is smaller
When ECN AFCT ratio DECN high about 30%, and as RTT becomes larger, the AFCT increasing degree of ECN constantly becomes larger, and DECN's changes
It is being continuously increased into amplitude.Fig. 5 (b) is 99 quantile stream deadline comparison diagrams under two kinds of mechanism.It can be seen that ECN
99 quantile stream deadlines are higher by 30% or more than DECN always, when this illustrates that DECN method can effectively reduce stream completion
Between.
Fig. 6 is the stream deadline comparison diagram under different concurrent intensity, and experiment topology is as shown in Fig. 2, fixed fluxion 50
It is 100 μ s with RTT.Change the random sending time section of stream, random interval increases to always 1s from 0.1s, 0.2s.Every stream
Data volume be set as 5000 data packets.Two kinds of mechanism of ECN and DECN of comparison.
Fig. 6 (a) is mean flow deadline comparison diagram, it can be seen that is become from small to large when stream random transmission interval time
When change, Congestion Level SPCC constantly declines, and the stream deadline is to reduce.But generally, the stream deadline of DECN ratio ECN is wanted
It is small.Fig. 6 (b) is 99 quantile stream deadline comparison diagrams, it can be seen that the stream deadline of DECN ratio ECN mechanism is small.
Claims (5)
1. a kind of difference explicit feedback transfer control method towards IPv6, which comprises the following steps:
Step 1: the current concurrent TCP flow number N of initialization real-time exchange machinec, fluxion statistics timer expiry threshold value, interchanger go out
Mouth rate C, real time rate rc, tidal flow volume Δ, targeted rate rt, maximum rate increment rmWith increment remainder rr;
Transmitting terminal is operated according to the following steps:
Step 2-1: ACK grouping is received;
Step 2-2: reading the received packet header ACK encoded information, and decoding obtains tidal flow volume Δ;
Step 2-3: it is used as transmission rate after present rate is increased tidal flow volume Δ, and with rate sending data grouping;
Step 2-4: judging whether grouping is sent, if so, terminating, otherwise goes to step 2-1;
Interchanger is operated according to the following steps:
Step 3-1: data grouping is received;
Step 3-2: judging whether current time reaches the timeout threshold of fluxion statistics timer, and step 3-3 is gone to if reaching;
Otherwise step 3-6 is gone to;
Step 3-3: real-time fluxion N is updatedc, update the real time rate r of all streamsc, and initialize adjustment rate raValue be rc;
Step 3-4: the targeted rate r of every stream is calculatedt=C/Nc;
Step 3-5: targeted rate rtThe maximum rate increment r carried divided by each ACK groupingm, obtained quotient is rounded, remaining
Remainder is increment remainder rr;
Step 3-6: judge the adjustment rate r that flow table where received data grouping is shownaWith targeted rate rtIt is whether equal, if so,
Then go to step 3-7;Otherwise, 3-8 is gone to step;
Step 3-7: the value of tidal flow volume Δ is set as 0, goes to step 3-10;
Step 3-8: judge raWhether r is greater thant, if so, the value of setting tidal flow volume Δ is-rm, adjust rate raValue be ra-
rm, go to step 3-10;Otherwise, step 3-9 is gone to;
Step 3-9: judge rt-raWhether r is greater thanm, if so, the value of setting tidal flow volume Δ is rm, adjust rate raValue be
ra+rm, go to step 3-10;Otherwise, then the value that tidal flow volume Δ is arranged is rr, adjust rate raValue be ra+rr, go to step
3-10;
Step 3-10: encoding Δ, and is written in data packet header, forwards data grouping, goes to step 3-1;
Receiving end is operated according to the following steps:
Step 4-1: data grouping is received;
Step 4-2: the encoded information of data packet header is read;
Step 4-3: the encoded information read is written in ACK packet header;
Step 4-4: it sends ACK and is grouped to transmitting terminal, go to step 4-1.
2. the difference explicit feedback transfer control method according to claim 1 towards IPv6, which is characterized in that the step
In rapid 1: by real-time fluxion NcIt is initialized as 1, by real time rate rcIt is initialized as 0;By fluxion statistics timer expiry threshold value setting
For 1ms;Current outlet port band width is set by interchanger egress rate C;The value of tidal flow volume Δ is initialized as 0;By target
Rate rtIt is initialized as 0;By maximum rate increment rmIt is set as 7;By increment remainder rrValue be initialized as 0.
3. the difference explicit feedback transfer control method according to claim 1 towards IPv6, which is characterized in that the step
In rapid 3-10: take first 4 in packet header in 8 bit traffic classification fields to encode rate increment Δ, the 1st is sign bit, 0
Just, 1 indicates negative for expression, and totally 8 binary numbers, sequence correspond to 0-7 totally 8 decimal numbers for rear 3 expressions 000 to 111, respectively table
Show 0~7Mbps, finally, by the rate increment Δ that range is -7~7Mbps be encoded to that range is 1111~0111 4 two into
Number processed.
4. the difference explicit feedback transfer control method according to claim 1 towards IPv6, which is characterized in that the step
In rapid 2-2: take first 4 in packet header in 8 bit traffic classification fields to be decoded rate increment Δ, the 1st is sign bit, 0
Just, 1 indicates negative for expression, and totally 8 binary numbers, sequence correspond to 0-7 totally 8 decimal numbers for rear 3 expressions 000 to 111, respectively table
Show 0~7Mbps, finally, 4 bits that range is 1111~0111 is decoded as the rate that range is -7~7Mbps and are increased
Measure Δ.
5. the difference explicit feedback transfer control method according to claim 1 towards IPv6, which is characterized in that the step
Rapid 3-7 is into 3-10 step: the rate increment Δ of explicit feedback is the changing value of rate adaptation, and interchanger utilizes the 4 of packets headers
Bit field carries rate increment Δ, and feeds back to sender;Sender carrys out regulations speed according to present rate and rate increment.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811568103.2A CN110098893B (en) | 2018-12-21 | 2018-12-21 | Differential explicit feedback transmission control method facing IPv6 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811568103.2A CN110098893B (en) | 2018-12-21 | 2018-12-21 | Differential explicit feedback transmission control method facing IPv6 |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110098893A true CN110098893A (en) | 2019-08-06 |
CN110098893B CN110098893B (en) | 2021-09-10 |
Family
ID=67443666
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811568103.2A Active CN110098893B (en) | 2018-12-21 | 2018-12-21 | Differential explicit feedback transmission control method facing IPv6 |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110098893B (en) |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7512066B2 (en) * | 2004-03-30 | 2009-03-31 | Hewlett-Packard Development Company, L.P. | Congestion control system |
EP2448195A1 (en) * | 2010-10-28 | 2012-05-02 | Samsung SDS Co. Ltd. | Apparatus and method for ensuring fairness of udp data transmission in ethernet environment |
US20130124753A1 (en) * | 2011-11-15 | 2013-05-16 | New Jersey Institute Of Technology | Fair quantized congestion notification (fqcn) to mitigate transport control protocol (tcp) throughput collapse in data center networks |
WO2014155043A1 (en) * | 2013-03-28 | 2014-10-02 | British Telecommunications Public Limited Company | Re-marking of packets for queue control |
CN106059951A (en) * | 2016-06-08 | 2016-10-26 | 中南大学 | Transmission control method for DCN (Data Center Network) based on multilevel congestion feedback |
WO2017009715A1 (en) * | 2015-07-14 | 2017-01-19 | Alcatel Lucent | Method and apparatus for managing network congestion |
CN106911583A (en) * | 2017-03-14 | 2017-06-30 | 国网四川省电力公司经济技术研究院 | A kind of transmission layer congestion control method of explicit feedback Congestion Level SPCC sliding average |
CN107026716A (en) * | 2017-05-12 | 2017-08-08 | 中南大学 | A kind of transfer control method perceived in data center network based on concurrency |
CN107135164A (en) * | 2017-06-27 | 2017-09-05 | 中国联合网络通信集团有限公司 | Jamming control method and device |
CN108270691A (en) * | 2018-01-11 | 2018-07-10 | 电子科技大学 | A kind of alternative ecn (explicit congestion notification) labeling method for data center |
-
2018
- 2018-12-21 CN CN201811568103.2A patent/CN110098893B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7512066B2 (en) * | 2004-03-30 | 2009-03-31 | Hewlett-Packard Development Company, L.P. | Congestion control system |
EP2448195A1 (en) * | 2010-10-28 | 2012-05-02 | Samsung SDS Co. Ltd. | Apparatus and method for ensuring fairness of udp data transmission in ethernet environment |
US20130124753A1 (en) * | 2011-11-15 | 2013-05-16 | New Jersey Institute Of Technology | Fair quantized congestion notification (fqcn) to mitigate transport control protocol (tcp) throughput collapse in data center networks |
WO2014155043A1 (en) * | 2013-03-28 | 2014-10-02 | British Telecommunications Public Limited Company | Re-marking of packets for queue control |
WO2017009715A1 (en) * | 2015-07-14 | 2017-01-19 | Alcatel Lucent | Method and apparatus for managing network congestion |
CN106059951A (en) * | 2016-06-08 | 2016-10-26 | 中南大学 | Transmission control method for DCN (Data Center Network) based on multilevel congestion feedback |
CN106911583A (en) * | 2017-03-14 | 2017-06-30 | 国网四川省电力公司经济技术研究院 | A kind of transmission layer congestion control method of explicit feedback Congestion Level SPCC sliding average |
CN107026716A (en) * | 2017-05-12 | 2017-08-08 | 中南大学 | A kind of transfer control method perceived in data center network based on concurrency |
CN107135164A (en) * | 2017-06-27 | 2017-09-05 | 中国联合网络通信集团有限公司 | Jamming control method and device |
CN108270691A (en) * | 2018-01-11 | 2018-07-10 | 电子科技大学 | A kind of alternative ecn (explicit congestion notification) labeling method for data center |
Non-Patent Citations (2)
Title |
---|
I. A. QAZI等: "Congestion control using efficient explicit feedback", 《IEEE INFOCOM 2009》 * |
JIANXIN WANG: "An Explicit Congestion Control Protocol Based on Bandwidth Estimation", 《2011 IEEE GLOBAL TELECOMMUNICATIONS CONFERENCE - GLOBECOM 2011》 * |
Also Published As
Publication number | Publication date |
---|---|
CN110098893B (en) | 2021-09-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103167359B (en) | The transmission method of RTP Media Stream and device | |
CN107171842B (en) | Multipath transmission protocol congestion control method based on reinforcement learning | |
KR102478440B1 (en) | Congestion control method and network device | |
CN101924603B (en) | Self-adaption adjusting method, device and system of data transmission rate | |
CN106059951B (en) | A kind of transfer control method for based on multilevel congestion feedback in DCN | |
CN102256314B (en) | Method for adjusting service coding rate and communication node | |
CN106330761A (en) | Congestion control method and device based on queue delay | |
CN103533451A (en) | Method and system for regulating jitter buffer | |
CN101969408B (en) | Active queue management method based on packet DSCP (Differentiated Services Code Point) marks | |
CN104113884A (en) | Real-time multimedia transmission rate control mechanism in wireless network | |
CN112491658A (en) | Bandwidth estimation method and device, electronic equipment and computer-readable storage medium | |
Nguyen et al. | Performance evaluation of TCP congestion control algorithms in data center networks | |
Xia et al. | Deep reinforcement learning for smart city communication networks | |
CN107154897A (en) | Isomery stream partition method based on bag scattering in DCN | |
CN103401778A (en) | Receiving-end buffer overflow probability guarantee based multi-path transmission packet scheduling method | |
CN103079231A (en) | Device and method for multicast-to-unicast based on wireless local area network in AP (Access Point) | |
CN101557273A (en) | Method simultaneously suitable for wired network real-time streaming media transport protocol and wireless network real-time streaming media transport protocol | |
CN107483990A (en) | Dynamic code rate adjusting method, device and the Transmission system of a kind of streaming media | |
CN103607255A (en) | Wireless channel video communication packet loss rate control method | |
CN110198273A (en) | Multi-path transmission method in data center network based on network code | |
CN110098893A (en) | Difference explicit feedback transfer control method towards IPv6 | |
CN111417029B (en) | Method, system, terminal and storage medium for adjusting video code rate | |
CN107872291A (en) | A kind of method of adjustment and terminal of voice code check | |
US20110225476A1 (en) | System And Method For Enabling A Buffer Less Or Substantially Buffer Less Core Network | |
CN109348259A (en) | A kind of the transmission code rate adjustment method and system of wireless network video |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |