CN108322401A - The control method and device of network transmission congestion - Google Patents
The control method and device of network transmission congestion Download PDFInfo
- Publication number
- CN108322401A CN108322401A CN201710037306.8A CN201710037306A CN108322401A CN 108322401 A CN108322401 A CN 108322401A CN 201710037306 A CN201710037306 A CN 201710037306A CN 108322401 A CN108322401 A CN 108322401A
- Authority
- CN
- China
- Prior art keywords
- network
- delay
- window
- state parameters
- parameter
- 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
- 230000005540 biological transmission Effects 0.000 title claims abstract description 104
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000001514 detection method Methods 0.000 claims abstract description 105
- 230000002159 abnormal effect Effects 0.000 claims description 19
- 230000003111 delayed effect Effects 0.000 claims description 16
- 238000012790 confirmation Methods 0.000 claims description 13
- 241000208340 Araliaceae Species 0.000 claims description 4
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims description 4
- 235000003140 Panax quinquefolius Nutrition 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 4
- 235000008434 ginseng Nutrition 0.000 claims description 4
- 238000004891 communication Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 4
- 241001269238 Data Species 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000002035 prolonged effect Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011084 recovery Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 101100048435 Caenorhabditis elegans unc-18 gene Proteins 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/12—Avoiding congestion; Recovering from congestion
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a kind of control method and device of network transmission congestion.Wherein, this method includes:Obtain network state parameters, wherein the network state parameters indicate the delay of network;Network bandwidth detection mode is determined based on the network state parameters;Congestion avoidance algorithm is selected based on the network bandwidth detection mode, calculates the parameter of congestion window;According to the parameter of the congestion window, to opposite end transmission data packet.The present invention solves the slow technical problem of transmission speed of wireless network.
Description
Technical field
The present invention relates to wireless communication fields, in particular to a kind of control method and device of network transmission congestion.
Background technology
With the universal and application of mobile Internet, user more and more accesses internet site's resource by mobile device
To obtain the network information and amusement and recreation, optionally, the mode that user accesses internet is not limited to office and family, may be used also
To select wireless mode to surf the Internet;However since mobile device is complicated to the environment between network server, wireless communications environment is anti-
Multiple variation accesses stream Bandwidth-Constrained and the factors such as network environment is uncertain, user's speed sometimes during website visiting
Slowly, it is easy to which the user for causing network application server complains and customer loss.In addition, customer access network server resource exists
For transport layer due to the unstability of wireless network, packet loss retransmission rate is relatively high, and the factors such as link bandwidth asymmetry also result in
The partially slow problem of speed, or and existing congestion avoidance algorithm be based on packet loss or whether gathering around based on delay judgement network
Plug and information is single can all influence the problem to some extent.
Specifically, the network server at user's access enterprise networks station drawback existing for transmission layered scheme is as follows:It is existing to answer
With trust server in IDC computer rooms, mostly use operating system default configuration, the application server of default configuration that can not be directed to nothing
The congestion control of thread environment optimizes;The application server of default configuration is practical in the algorithm performance of wireless environment transmission control
Test effect is simultaneously bad.
For the slow-footed problem of above-mentioned wireless network transmissions, currently no effective solution has been proposed.
Invention content
An embodiment of the present invention provides a kind of control method and device of network transmission congestion, at least to solve wireless network
The slow technical problem of transmission speed.
One side according to the ... of the embodiment of the present invention provides a kind of control method of network transmission congestion, including:It obtains
Network state parameters, wherein the network state parameters indicate the delay of network;Network is determined based on the network state parameters
Bandwidth detection mode;Congestion avoidance algorithm is selected based on the network bandwidth detection mode, calculates the parameter of congestion window;According to
The parameter of the congestion window, to opposite end transmission data packet.
Further, obtaining network state parameters includes:Average delay is determined by the first formula, wherein described first
Formula is:AvgRTT=((scale-1) * avgRTT+rtt)/scale, wherein avgRTT indicates that average delay, scale indicate
First delay parameter, rtt indicate current delay;Transmission delay is determined based on the current delay;Based on the average delay with
The transmission delay determines queuing delay.
Further, determine that network bandwidth detection mode includes based on the network state parameters:It detects described network-like
Whether the state of state parameter is effective status;If it is detected that the state of the network state parameters is effective status, base is used
In the mode detection network bandwidth of network delay;If it is detected that the state of the network state parameters is not effective status and examines
It measures the network and is in normal operating conditions, then the detection network bandwidth by the way of packet loss detection, obtains congestion window
Window parameter, wherein the detection mode includes the mode of the packet loss detection;If it is detected that the shape of the network state parameters
State is not effective status and detects that the network is in abnormal operation, then obtains the window ginseng of preset congestion window
Number.
Further, detection network bandwidth includes by the way of packet loss detection:The information of the N items connection of configuration is obtained,
Wherein, N is natural number;The window parameter of the congestion window of the network bandwidth is determined by the second formula, wherein described second
Formula is:ωi(t1)=ω+N* ωi(t0), wherein the ωi(t1) indicate the window parameter of the current congestion window,
ω indicates the congestion window information that the information connected based on the N items is determined, ωi(t0) indicate the last time using packet loss detection
Mode carries out the window parameter for the congestion window that network bandwidth obtains.
Further, detection network bandwidth includes by the way of based on network delay:Described in the determination of third formula
The window parameter of congestion window, wherein the third formula is: Wherein, the ωi(t1) indicate current congestion window window parameter, ωi(t0) indicate last based on net
The mode of network delay carries out the window parameter for the congestion window that network bandwidth obtains, and γ is convergence coefficient, αiIt indicates to arrange in network
The discreet of team's data packet;D indicates that the queuing delay in the network state parameters, d indicate in the network state parameters
Transmission delay.
Further, before the window parameter for determining the congestion window by third formula, the method further includes:
Detect whether the current delay is less than or equal to history delay, wherein the history delay indicates that last time receives data packet
Delay;If the current delay is delayed less than or equal to the history, the convergence coefficient is chosen from the first interval
Value;If the current delay is delayed more than the history, the value of the convergence coefficient is chosen from the second interval,
In, the minimum value in first interval is more than the maximum value in second interval.
Further, before whether the state for detecting the network state parameters is effective status, the method is also wrapped
It includes:Detect whether network is in normal operating conditions, wherein detect whether to receive the confirmation instruction of repetition transmission, if detection
The confirmation sent instruction is repeated to described, then detects that the network is in abnormal operation;It detects whether to receive time-out
Retransmission instructions detect that the network is in abnormal operation if detecting the Retransmission timeout instruction.
Another aspect according to the ... of the embodiment of the present invention additionally provides a kind of network transmission congestion control device, including:It obtains
Unit, for obtaining network state parameters, wherein the network state parameters indicate the delay of network;Determination unit is used for base
Network bandwidth detection mode is determined in the network state parameters;Selecting unit, for being based on the network bandwidth detection mode
Select congestion avoidance algorithm;Computing unit, the parameter for calculating congestion window;Transmission unit, for according to the congestion window
The parameter of mouth, to opposite end transmission data packet.
Further, the acquiring unit includes:Determining module determines average delay for passing through the first formula, wherein
First formula is:AvgRTT=((scale-1) * avgRTT+rtt)/scale, wherein avgRTT indicates average delay,
Scale indicates that the first delay parameter, rtt indicate current delay;First determining module is passed for being determined based on the current delay
Defeated delay;Second determining module, for determining queuing delay based on the average delay and the transmission delay.
Further, the determination unit includes:First detection module, the state for detecting the network state parameters
Whether it is effective status;First detecting module, if it is detected that the state of the network state parameters is effective status, for using
Mode detection network bandwidth based on network delay;Second detecting module, if it is detected that the state of the network state parameters is not
For effective status and detect that the network is in normal operating conditions, for the detection network band by the way of packet loss detection
It is wide, wherein the detection mode includes the mode of the packet loss detection;Acquisition module, if it is detected that the network state parameters
State be not effective status and to detect that the network is in abnormal operation, for obtaining preset congestion window
Window parameter.
Further, second detecting module includes:Acquisition submodule, the information of the N items connection for obtaining configuration,
Wherein, N is natural number;First determination sub-module, the window of the congestion window for determining the network bandwidth by the second formula
Mouthful parameter, wherein second formula is:ωi(t1)=ω+N* ωi(t0), wherein the ωi(t1) indicate currently described
The window parameter of congestion window, ω indicate the congestion window information that the information connected based on the N items is determined, ωi(t0) in expression
The primary window parameter that the congestion window that network bandwidth obtains is carried out by the way of packet loss detection.
Further, first detecting module includes:Second determination sub-module, described in being determined by third formula
The window parameter of congestion window, wherein the third formula is: Wherein, the ωi(t1) indicate current congestion window window parameter, ωi(t0) indicate last based on net
The mode of network delay carries out the window parameter for the congestion window that network bandwidth obtains, and γ is convergence coefficient, αiIt indicates to arrange in network
The discreet of team's data packet;D indicates that the queuing delay in the network state parameters, d indicate in the network state parameters
Transmission delay.
Further, described device further includes:First detection unit, for determining the congestion window by third formula
Before the window parameter of mouth, detect whether the current delay is less than or equal to history delay, wherein the history delay indicates
The secondary delay for receiving data packet;First selection unit, if the current delay is delayed less than or equal to the history, for from the
The value of the convergence coefficient is chosen in one interval;Second selection unit, if the current delay is delayed more than the history,
Value for choosing the convergence coefficient from the second interval, wherein the minimum value in first interval is more than
Maximum value in second interval.
Further, described device further includes:Second detection unit, in the state for detecting the network state parameters
Before whether being effective status, detection network whether be in normal operating conditions, wherein the second detection module, for detect be
The no confirmation instruction for receiving repetition transmission detects if detecting the confirmation instruction for repeating transmission at the network
In abnormal operation;Third detection module, for detecting whether Retransmission timeout instruction is received, if detecting the time-out weight
Teletype command then detects that the network is in abnormal operation.
By above-described embodiment, after obtaining network state parameters, determine that network bandwidth is visited based on network state parameters
Survey mode is then based on network bandwidth detection mode selection congestion avoidance algorithm, the parameter of congestion window is calculated, according still further to congestion
The parameter of window, to opposite end transmission data packet.In the above-described embodiments, it when needing transmission packet, can be based on indicating network
The network state parameters of delay determine detection mode, and the parameter of congestion window is determined using determining detection mode, with into
Row data transmission can accurately determine the window parameter of congestion window based on network state parameters, be passed according to the window parameter
The quick transmission of data may be implemented in transmission of data packet, solves the problems, such as that wireless network transmissions are slow-footed in the prior art.
Description of the drawings
Attached drawing described herein is used to provide further understanding of the present invention, and is constituted part of this application, this hair
Bright illustrative embodiments and their description are not constituted improper limitations of the present invention for explaining the present invention.In the accompanying drawings:
Fig. 1 is a kind of system structure diagram of the control method of network transmission congestion according to the ... of the embodiment of the present invention;
Fig. 2 is a kind of flow chart of the control method of optional network transmission congestion according to the ... of the embodiment of the present invention;
Fig. 3 is congestion avoidance algorithm normal workflow figure according to the ... of the embodiment of the present invention;
Fig. 4 is congestion avoidance algorithm packet loss work flow diagram according to the ... of the embodiment of the present invention;
Fig. 5 is a kind of schematic diagram of optional network transmission congestion control device according to the ... of the embodiment of the present invention.
Specific implementation mode
In order to enable those skilled in the art to better understand the solution of the present invention, below in conjunction in the embodiment of the present invention
Attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only
The embodiment of a part of the invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill people
The every other embodiment that member is obtained without making creative work should all belong to the model that the present invention protects
It encloses.
It should be noted that term " first " in description and claims of this specification and above-mentioned attached drawing, "
Two " etc. be for distinguishing similar object, without being used to describe specific sequence or precedence.It should be appreciated that using in this way
Data can be interchanged in the appropriate case, so as to the embodiment of the present invention described herein can in addition to illustrating herein or
Sequence other than those of description is implemented.In addition, term " comprising " and " having " and their any deformation, it is intended that cover
It includes to be not necessarily limited to for example, containing the process of series of steps or unit, method, system, product or equipment to cover non-exclusive
Those of clearly list step or unit, but may include not listing clearly or for these processes, method, product
Or the other steps or unit that equipment is intrinsic.
According to embodiments of the present invention, a kind of embodiment of the method for the control method of network transmission congestion is provided, is needed
Bright, step shown in the flowchart of the accompanying drawings can be held in the computer system of such as a group of computer-executable instructions
Row, although also, logical order is shown in flow charts, and it in some cases, can be with different from sequence herein
Execute shown or described step.
Include as shown in Figure 1:Terminal 13, communication link 15, Provider Equipment 17 and application server 19.User passes through
Terminal accesses application server by Provider Equipment (such as base station and gateway device).Wherein, terminal and Provider Equipment via
Communication link connects, and Provider Equipment is also to be connect via communication link with application server.Terminal can be personal computer
And mobile terminal, such as mobile phone and tablet computer, the application does not limit the specific manifestation form of terminal.
Fig. 2 is the flow chart of the control method of network transmission congestion according to the ... of the embodiment of the present invention, as shown in Fig. 2, the party
Method includes the following steps:
Step S202 obtains network state parameters, wherein network state parameters indicate the delay of network;
Step S204 determines network bandwidth detection mode based on network state parameters;
Step S206 selects congestion avoidance algorithm based on network bandwidth detection mode, calculates the parameter of congestion window;
Step S208, according to the parameter of congestion window, to opposite end transmission data packet.
By above-described embodiment, after obtaining network state parameters, determine that network bandwidth is visited based on network state parameters
Survey mode is then based on network bandwidth detection mode selection congestion avoidance algorithm, the parameter of congestion window is calculated, further according to congestion
The parameter of window is to opposite end transmission data packet.It in the above-described embodiments, can be based on expression network when needing transmission packet
The network state parameters of delay determine detection mode, and the parameter of congestion window is determined using determining detection mode, to carry out
Data transmission can accurately determine the window parameter of congestion window based on network state parameters, be transmitted according to the window parameter
The quick transmission of data may be implemented in data packet, solves the problems, such as that wireless network transmissions are slow-footed in the prior art.
In the technical solution that step S202 is provided, delay may include:Since Provider Equipment receive data to
Time interval between destination interface transmission data packet can also include time and reality that server should receive data packet
Border receives the time interval between data packet.It is alternatively possible to indicate the delay of network using network state parameters, this prolongs
When may include delay duration and the state whether network is delayed.
In the technical solution that step S204 is provided, network bandwidth refers to the data volume that can be transmitted within the unit interval, can
Selection of land, the unit interval can be 1 second.
In the technical solution that step S206 is provided, congestion window refers to be source (such as server) in congestion control situation
The quantity for the data packet that most multipotency is sent next time, transmitting terminal (such as server) default one according to the Congestion Level SPCC of network
Sizes values, and this value is exactly the window parameter of congestion window, what the window parameter of the congestion window can indicate once to send
The quantity of data packet.
After determining the window parameter of congestion window, according to the quantity for the data packet that determining primary most multipotency is sent,
To client transmission data packet.
In the above-described embodiments, in the case where needing transmission data, the network state parameters for indicating network delay, profit are determined
Network bandwidth detection is carried out with the network state parameters, to determine the quantity for the data packet that once can at most send, according to this
The quantity for the data packet that determining primary most multipotency is sent in the above scheme, can be based on net to client transmission data packet
The size of congestion window is rationally estimated in network delay, so as to transmission packet as quickly as possible.
Wherein, above-mentioned network state parameters include:Average delay avgRTT, transmission delay baseRTT and queuing are prolonged
When queueRTT, optionally, average delay avgRTT, transmission delay baseRTT and queuing delay queueRTT initial value
It is 0.
Optionally, obtaining network state parameters includes:Average delay is determined by the first formula, wherein the first formula is:
AvgRTT=((scale-1) * avgRTT+rtt)/scae, wherein avgRTT indicates that average delay, scale indicate the first delay
Parameter, rtt indicate current delay;Transmission delay is determined based on current delay;It is determined and is lined up based on average delay and transmission delay
Delay.
Optionally, queuing delay is the difference of the transmission delay of network and the average delay of network.
In an alternative embodiment, the statistical calculation method of average delay avgRTT is:AvgRTT=((scale-
1) * avgRTT+rtt)/scale, wherein the first delay parameter scale is identified not yet according to having been sent in network
Obtained from data packet (i.e. first kind data packet) is estimated, optionally, scale values are min (256, scale),
That is, the minimum value of first delay parameter, which is 256, rtt, indicates current delay.
In another optional embodiment, the evaluation method of transmission delay baseRTT is the minimum rtt that observation obtains
(i.e. current delay), that is, transmission delay baseRTT=min (avgRTT, rtt).
In there are one optional embodiment, the difference by calculating average delay and transmission delay obtains queuing delay
QueueRTT, that is, queuing delay queueRTT=avgRTT-baseRTT.
Before executing step S202, method further includes:The working condition of network is detected, and records the work shape of the determination
State, the working condition may include normal operating conditions and abnormal operation, wherein abnormal operation includes packet loss and hair
The state of raw congestion.Fig. 3 is the flow chart of congestion avoidance algorithm normal work according to the ... of the embodiment of the present invention, and Fig. 4 is according to this
The congestion avoidance algorithm packet loss work flow diagram of inventive embodiments.
Optionally, detection network whether in normal operating conditions include at least one following:It detects whether to receive weight
The confirmation instruction ack sent is recurred, if detecting the confirmation instruction ack that repetition is sent, detects that network is in abnormal work shape
State;It detects whether to receive Retransmission timeout instruction, if detecting, Retransmission timeout instructs, and detects that network is in abnormal work shape
State.
Optionally, according to detection mode corresponding with the state of network state parameters, detection network bandwidth includes:Detect net
Whether the state of network state parameter is effective status;If it is detected that the state of network state parameters is effective status, it is based on net
Network state parameter determines the window parameter of congestion window, wherein detection mode includes the mode of network delay;If it is detected that network
The state of state parameter is not effective status and detects that network is in normal operating conditions, then by the way of packet loss detection
Detection network bandwidth obtains the window parameter of congestion window, wherein detection mode includes the mode of packet loss detection;If it is detected that
The state of network state parameters is not effective status and detects that network is in abnormal operation, then obtains preset congestion
The window parameter of window.
The application optional embodiment is described in detail with reference to Fig. 3 and Fig. 4.
As shown in figure 3, may include steps of in an alternative embodiment:
Step S301 receives request or confirms instruction ack.
If detecting transmission request or confirming instruction ack, it is determined that the working condition of network is normal operating conditions.
Effectively whether step S302, detection network state parameters are estimated.
If it is detected that network state parameters are estimated effectively, S303 is thened follow the steps, if it is detected that network state parameters are estimated
In vain, S304 is thened follow the steps.
Optionally, in the case where network state parameters include transmission delay and average delay, network state parameters are detected
State whether be that effective status includes:If it is detected that transmission delay and average delay are not zero, it is determined that detect network-like
The state of state parameter is effective status;If it is detected that transmission delay and average delay are zero, it is determined that detect network state
The state of parameter is not effective status.
Wherein, when receiving confirmation instruction ack, to average delay avgRTT, transmission delay baseRTT and queuing are prolonged
When queueRTT estimated.If the value of network state parameters is all 0, network state parameters estimation does not come into force;If average
The avgRTT and transmission delay baseRT that is delayed is not 0, it is determined that collected network state parameters are effective.
When network state parameters are effective, the detection network bandwidth by the way of based on network delay;Network state parameters
When estimation does not come into force, the detection network bandwidth by the way of packet loss detection obtains the window parameter of congestion window.
Step S303 sends window and updates timers trigger.
Wherein, window update timer is used to record the update duration of congestion window, and in the update of last time congestion window
Time interval between moment and current time triggers the update of congestion window in the case of reaching predetermined time interval.
Step S305, configuration N items link.
Step S306 calculates target and sends window size, and transmission window is arranged.
Wherein, target sends the window parameter that window size is above-mentioned congestion window (sending window).
It optionally, can be real with S305 through the above steps and S306 in the case of window update clocked flip-flop triggering
The window parameter of congestion window is now determined based on network state parameters.
Further, which includes:It is gathered around by the determination of third formula
Fill in the window parameter of window, wherein third formula is: Its
In, ωi(t1) indicate current congestion window window parameter, ωi(t0) indicate that the last mode based on network delay carries out
The window parameter for the congestion window that network bandwidth obtains, γ are convergence coefficient, αiData queued packet estimates number in expression network
Amount;D indicates that the queuing delay in network state parameters, d indicate the transmission delay in network state parameters.
Optionally, it when network state parameters estimate effective, needs to judge whether window update timer triggers, in window
In the case of updating timers trigger, network bandwidth is calculated by the way of based on network delay, the mode according to FastTCP is estimated
Window size is calculated, that is to say and pass through formula: Window is carried out
It estimates.Here γ is convergence coefficient, optionally, fixed constant 0.5, ω can be taken in FastTcpi(t0) it is that last RTT is adopted
Congestion control window when sample, ωi(t1) it is estimation to congestion control window, αiIt is the estimation to data queued packet in network,
D is queuing delay queueRTT, that is, difference the queueRTT=avgRTT-baseRTT, d of average delay and transmission delay
It is transmission delay baseRTT, the evaluation method of transmission delay baseRTT is the minimum rtt, baseRTT=min of observation
(avgRTT,rtt)。
Further, before determining the window parameter of congestion window by third formula, method further includes:Detection is current
Whether delay, which is less than or equal to history, is delayed, wherein history delay indicates that last time receives the delay of data packet;If current delay is small
It is delayed in equal to history, then chooses the value of convergence coefficient from the first interval;If current delay is delayed more than history, from
The value of convergence coefficient is chosen in second interval, wherein the minimum value in the first interval is more than in the second interval
Maximum value.
Optionally, if window updates timers trigger, if current RTT (current delay)<=SRTT (history delay),
SRTT is that RTT is calculated in last time packet receiving, then it is assumed that network load reduces, and should increase window, and γ values are [0.6,0.8], are write from memory
Think 0.75;It is on the contrary, then it is assumed that network load increases, and should reduce congestion window, γ is set as between [0.3,0.5], is defaulted as
0.4;So the value of γ is just [0.4,0.75], calculates and updates target window size.
Step S304 is based on packet loss detective bandwidth.
Step S307 is calculated and transmission window size is arranged.
Detection network bandwidth optionally uses in such a way that step S305 and step S307 are realized using packet loss detection
Packet loss detection mode detection network bandwidth include:Obtain the information of the N items connection of configuration, wherein N is natural number, i.e. N is non-
Negative integer;The window parameter of the congestion window of network bandwidth is determined by the second formula, wherein the second formula is:ωi(t0)=
ω+N*ωi(t0), wherein ωi(t1) indicate that the window parameter of current congestion window, ω indicate the information that is connected based on N items
Determining congestion window information, ωi(t0) indicate the last congestion window for carrying out network bandwidth by the way of packet loss detection and obtaining
The window parameter of mouth.
Optionally, if packet loss mode detection network bandwidth, N FastTCP and Hybla is simulated, calculates congestion window
Size;ωi(t0)=ω+N* ωi(t0), ω is the size for the congestion window that hybla algorithms are calculated, ω herei(t0)
It is the window size of last FastTcp estimations, N is natural number, i.e. nonnegative integer, N=2.Its main purpose improved in this way
The case where being desirable to RTT connections can obtain fast transmission rate as cable network.Its rudimentary algorithm thought:Long RTT meetings
The relatively slow of rate is caused to increase, so having introduced a normalized RTT value.The advantages of Hybla, is a cancellation long RTT to TCP
The influence of congestion control mechanism so that TCP provides higher transmission rates under long RTT condition.
Wherein, FastTCP is considered as the current state-of-the-art unilateral deployment acceleration technique of carrier class ICP/IP protocol in the whole world,
It has significant effect to the TCP flow amount on wide area network and wireless network, is especially delayed in height, the TCP network rings of high packet loss
Effect is more obvious in border, can in the TCP network environments of high packet loss:The response time of application is reduced, TCP is improved and handles up
The utilization rate of the speed and raising wireless network and wide area network bandwidth of amount and effective discharge.Hybla is primarily directed to long delay
TCP improvement, while also efficiently solving fairness problems of the TCP on different delayed time chain road.It is by a series of process groups
At, including the slow turn-on and congestion avoidance phase that enhance, the SACK strategies used and channel width algorithm for estimating are forced, and
Timestamp and new packaging technique are used.
Optionally, application server receives duplicate acknowledgment instruction ack RTO time-out and detects packet loss, indicates at network
In abnormal operation.
Wherein, RTO is TCP Retransmission timeouts mechanism (Retransmission Timeout), is the one of TCP function timers
Kind.Datagram is lost in order to prevent, as soon as when TCP transmission messages, starts re-transmission timer, there are two types of situations:If 1.
The confirmation that special packet is had received before timer expired, then cancel this timer;2. specific data message is super in timer
When before do not receive confirmation, then retransmit the datagram, and the numerical value of timer reset RTO based on smooth two-way time and its
Deviation.
Step S308 updates network state parameters.
Optionally, network state parameters are updated, recalculate average delay avgRTT, transmission delay baseRTT, queuing is prolonged
When queueRTT, in order to next time determine congestion window window size when can directly acquire the network state parameters.
Step S309, according to transmission window transmission data packet.
Optionally, give out a contract for a project according to transmission window setting size, continue waiting for confirming instruction ack or new request.
Alternative embodiment as shown in Figure 4 may include steps of:
Step S401 detects packet loss or congestion occurs.
Optionally, it detects packet loss or congestion occurs, it may be determined that network is in abnormal operation.
Effectively whether step S402, detection network state parameters are estimated.
If it is detected that network state parameters are estimated effectively, S403 is thened follow the steps, if it is detected that network state parameters are estimated
In vain, S406 is thened follow the steps.
The specific implementation of the step S302 of the realization method of the step as shown in figure 3 is consistent, and details are not described herein.
Step S403 records current network state parameter.
Whether step S404, detection packet loss restore to complete.
If so, thening follow the steps S405;If it is not, then continuing to execute step S404.
Step S405 sends window according to the setting of current network state parameter prediction.
Optionally, in the realization method and above-described embodiment of the step step S305 and step S306 realization method one
It causes, details are not described herein.
Optionally, it is based on network bandwidth detection mode selection congestion avoidance algorithm, the parameter for calculating congestion window includes:
In the case of network recovery normal operating conditions, the parameter of congestion window is calculated.
Specifically, it if loss recovery is completed, sends window size and estimates (i.e. according to network-like according to current network bandwidth
State parameter) setting.
Step S406, it is the initial value for sending window that setting, which sends window,.
Step S407, according to transmission window transmission data packet.
Optionally, the realization method of the step is consistent with the realization method of step S309 in above-described embodiment, herein no longer
It repeats.
Pass through obtain network state parameters, wherein network state parameters indicate network delay, then according to it is network-like
Then the corresponding detection mode of state of state parameter, detection network bandwidth select congestion avoidance algorithm corresponding with detection mode
To calculate the window parameter of congestion window, wherein the state of network state parameters for indicating whether network state parameters effective,
And then the window parameter according to congestion window, to opposite end transmission data packet.In the above-described embodiments, hybla congestion controls are based on
The a plurality of TCP connection of simulated implementation in system makes full use of bandwidth as far as possible, and congestion avoidance algorithm is after packet loss generation, according to right
The bandwidth estimation of network restores rapidly the state before packet loss.It has been used simultaneously based on two kinds of transmission congestion control sides of packet loss and delay
Method, it is preferential to select hybla algorithms and TCP_FAST algorithms on the basis of based on experimental analysis, but it is not limited to both algorithms
Combination, and then solve the slow-footed technical problem of wireless network transmissions.
Specific embodiment of the invention optimum results are given below, the application an optional embodiment is carried out specifically
It is bright:
For example, for 5.13,5.14,5.15 three days on line signaling datas, be extracted lan1 106 network element datas and
203 network element datas of lan2 are compared, and test data shows that our algorithm has larger promotion, table 1 to give mobile mutual
Correspondence after networking content information, optimization, before optimization and after optimization.
Table 1
In embodiments of the present invention, by obtaining network state parameters, wherein network state parameters indicate prolonging for network
When, then according to detection mode corresponding with the state of network state parameters, detection network bandwidth, then selection and detection mode
Corresponding congestion avoidance algorithm is to calculate the window parameter of congestion window, wherein the state of network state parameters is for indicating net
Whether network state parameter is effective, and then according to the window parameter of congestion window, to opposite end transmission data packet.Realize congestion control
Algorithm processed is easy to transplant all kinds of nix operating systems, these systems can be that can be integrated in cdn service nodes, web server,
Or function is realized by way of module loading, the carry out network bandwidth and parameter Estimation of congestion avoidance algorithm are also achieved,
Based on FastTCP congestion avoidance algorithms be based on delay judgement network whether congestion, improve the parameter of state equation, realize congestion
The more acurrate estimation of window.The a plurality of TCP connection of simulated implementation, makes full use of band as far as possible on based on hybla congestion controls
It is wide.Realize that congestion avoidance algorithm after packet loss generation, restores rapidly the state before packet loss according to the bandwidth estimation to network.Simultaneously
It has used and has preferentially selected hybla on the basis of based on experimental analysis based on two kinds of transmission jamming control methods of packet loss and delay
Algorithm and TCP_FAST algorithms, but be not limited to the combination of both algorithms, and then the transmission speed that solves wireless network is slow
Technical problem.
Wherein, cdn (Content Delivery Network, i.e. content distributing network) is by existing
Increase by one layer of new network architecture in Internet, the content of website is published to the network " edge " closest to user, uses
Family can obtain required content nearby, solve Internet network congestion condition, improve the response speed that user accesses website.
Technically comprehensively solve is since network bandwidth is small, user's visit capacity is big, the unequal reason of network point distribution, solves user and accesses net
The slow basic reason of the response speed stood.
According to another embodiment of the invention, a kind of device embodiment of network transmission congestion control is additionally provided, such as
Include shown in Fig. 5:Acquiring unit 51, determination unit 53, selecting unit 55, computing unit 57 and transmission unit 59.
Wherein, acquiring unit 51, for obtaining network state parameters, wherein network state parameters indicate the delay of network;
Determination unit 53, for determining network bandwidth detection mode based on network state parameters;Selecting unit 55, for being based on Netowrk tape
Wide detection mode selects congestion avoidance algorithm;Computing unit 57, the parameter for calculating congestion window;Transmission unit 59, is used for
According to the parameter of congestion window, to opposite end transmission data packet.
Optionally, acquiring unit includes:Determining module determines average delay for passing through the first formula, wherein first is public
Formula is:AvgRTT=((scale-1) * avgRTT+rtt)/scale, wherein avgRTT indicates that average delay, scale indicate the
One delay parameter, rtt indicate current delay;First determining module, for determining transmission delay based on current delay;Second determines
Module, for determining queuing delay based on average delay and transmission delay.
Optionally it is determined that unit includes:First detection module, whether the state for detecting network state parameters is effective
State;First detecting module, if it is detected that the state of network state parameters is effective status, for using based on network delay
Mode detection network bandwidth;Second detecting module, if it is detected that the state of network state parameters is not effective status and detects
Network is in normal operating conditions, for the detection network bandwidth by the way of packet loss detection, wherein detection mode includes packet loss
The mode of detection;Acquisition module, if it is detected that the state of network state parameters is not effective status and detects that network is in different
Normal working condition, the window parameter for obtaining preset congestion window.
Optionally, the second detecting module includes:Acquisition submodule, the information of the N items connection for obtaining configuration, wherein N
For natural number;First determination sub-module, the window parameter of the congestion window for determining network bandwidth by the second formula,
In, the second formula is:ωi(t1)=ω+N* ωi(t0), wherein ωi(t1) indicate current congestion window window parameter, ω
Indicate the congestion window information that the information connected based on N items is determined, ωi(t0) indicate it is last detected using packet loss by the way of into
The window parameter for the congestion window that row network bandwidth obtains.
Optionally, the first detecting module includes:Second determination sub-module, for determining congestion window by third formula
Window parameter, wherein third formula is:Wherein, ωi
(t1) indicate current congestion window window parameter, ωi(t0) indicate that the last mode based on network delay carries out Netowrk tape
The window parameter for the congestion window that width obtains, γ are convergence coefficient, αiIndicate the discreet of data queued packet in network;D tables
Show that the queuing delay in network state parameters, d indicate the transmission delay in network state parameters.
Optionally, device further includes:First detection unit, in the window ginseng for determining congestion window by third formula
Before number, whether detection current delay, which is less than or equal to history, is delayed, wherein history delay indicates that last time receives prolonging for data packet
When;First selection unit, if current delay is delayed less than or equal to history, for choosing convergence coefficient from the first interval
Value;Second selection unit, if current delay is delayed more than history, the value for choosing convergence coefficient from the second interval,
Wherein, the minimum value in the first interval is more than the maximum value in the second interval.
Optionally, device further includes:Whether second detection unit is effective for the state in detection network state parameters
Before state, whether detection network is in normal operating conditions, wherein the second detection module, for detecting whether receiving weight
The confirmation instruction ack sent is recurred, if detecting the confirmation instruction that repetition is sent, detects that network is in abnormal operation;
Third detection module, for detecting whether receiving Retransmission timeout instruction, if detecting, Retransmission timeout instructs, and detects network
In abnormal operation.
By above-described embodiment, after determining network state parameters, according to corresponding with the state of network state parameters
Detection mode, detection network bandwidth are then based on detection mode selection congestion avoidance algorithm, the ginseng for calculating congestion window
Number, according to the window parameter of the congestion window of the determination, to terminal transmission data packet.In the above-described embodiments, number need to transmitted
When according to packet, detection mode can be determined based on the network state parameters of the delay state of network are indicated, and utilize determining detection
Mode determines the parameter of congestion window, and to carry out data transmission, congestion window can be accurately determined based on network state parameters
Window parameter, the quick transmission of data may be implemented according to the window parameter transmission packet, solve in the prior art without
The slow problem of line network transfer speeds.
The embodiments of the present invention are for illustration only, can not represent the quality of embodiment.
In the above embodiment of the present invention, all emphasizes particularly on different fields to the description of each embodiment, do not have in some embodiment
The part of detailed description may refer to the associated description of other embodiment.
In several embodiments provided herein, it should be understood that disclosed technology contents can pass through others
Mode is realized.Wherein, the apparatus embodiments described above are merely exemplary, for example, unit division, can be one kind
Division of logic function, formula that in actual implementation, there may be another division manner, such as multiple units or component can combine or can
To be integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual
Coupling, direct-coupling or communication connection can be by some interfaces, the INDIRECT COUPLING or communication connection of unit or module,
Can be electrical or other forms.
The unit illustrated as separating component may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, you can be located at a place, or may be distributed over multiple
On unit.Some or all of unit therein can be selected according to the actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it can also
It is that each unit physically exists alone, it can also be during two or more units be integrated in one unit.Above-mentioned integrated list
The form that hardware had both may be used in member is realized, can also be realized in the form of SFU software functional unit.
It, can if integrated unit is realized in the form of SFU software functional unit and when sold or used as an independent product
To be stored in a computer read/write memory medium.Based on this understanding, technical scheme of the present invention substantially or
Say that all or part of the part that contributes to existing technology or the technical solution can embody in the form of software products
Out, which is stored in a storage medium, including some instructions are used so that a computer equipment
(can be personal computer, server or network equipment etc.) executes all or part of step of each embodiment method of the present invention
Suddenly.And storage medium above-mentioned includes:USB flash disk, read-only memory (ROM, Read-Only Memory), random access memory
The various media that can store program code such as (RAM, Random Access Memory), mobile hard disk, magnetic disc or CD.
It the above is only the preferred embodiment of the present invention, it is noted that those skilled in the art are come
It says, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications also should be regarded as
Protection scope of the present invention.
Claims (10)
1. a kind of control method of network transmission, which is characterized in that including:
Obtain network state parameters, wherein the network state parameters indicate the delay of network;
Network bandwidth detection mode is determined based on the network state parameters;
Congestion avoidance algorithm is selected based on the network bandwidth detection mode, calculates the parameter of congestion window;
According to the parameter of the congestion window, to opposite end transmission data packet.
2. according to the method described in claim 1, it is characterized in that, acquisition network state parameters include:
Average delay is determined by the first formula, wherein first formula is:
AvgRTT=((scale-1) * avgRTT+rtt)/scale, wherein avgRTT indicates that average delay, scale indicate the
One delay parameter, rtt indicate current delay;
Transmission delay is determined based on the current delay;
Queuing delay is determined based on the average delay and the transmission delay.
3. according to the method described in claim 1, it is characterized in that, determining that network bandwidth detects based on the network state parameters
Mode includes:
Whether the state for detecting the network state parameters is effective status;
If it is detected that the state of the network state parameters is effective status, the detection network by the way of based on network delay
Bandwidth;
If it is detected that the state of the network state parameters is not effective status and detects that the network is in normal work shape
State, then the detection network bandwidth by the way of packet loss detection, wherein the detection mode includes the mode of packet loss detection;
If it is detected that the state of the network state parameters is not effective status and detects that the network is in abnormal work shape
State then obtains the window parameter of preset congestion window.
4. according to the method described in claim 3, it is characterized in that, detection network bandwidth includes by the way of packet loss detection:
Obtain the information of the N items connection of configuration, wherein N is natural number;
The window parameter of the congestion window of the network bandwidth is determined by the second formula, wherein second formula is:
ωi(t1)=ω+N* ωi(t0), wherein the ωi(t1) indicate the window parameter of the current congestion window, ω tables
Show the congestion window information that the information connected based on the N items is determined, ωi(t0) indicate last by the way of packet loss detection
Carry out the window parameter for the congestion window that network bandwidth obtains.
5. according to the method described in claim 3, it is characterized in that, by the way of based on network delay detection network bandwidth packet
It includes:
The window parameter of the congestion window is determined by third formula, wherein the third formula is:
Wherein, the ωi(t1) indicate current and gather around
Fill in the window parameter of window, ωi(t0) indicate that the last mode based on network delay carries out the congestion window that network bandwidth obtains
Window parameter, γ is convergence coefficient, αiIndicate the discreet of data queued packet in network;D indicates the network state ginseng
Queuing delay in number, d indicate the transmission delay in the network state parameters.
6. according to the method described in claim 5, it is characterized in that, in the window for determining the congestion window by third formula
Before parameter, the method further includes:
Detect whether the current delay is less than or equal to history delay, wherein the history delay indicates that last time receives data
The delay of packet;
If the current delay is delayed less than or equal to the history, the convergence coefficient is chosen from the first interval
Value;
If the current delay is delayed more than the history, the value of the convergence coefficient is chosen from the second interval,
In, the minimum value in first interval is more than the maximum value in second interval.
7. according to the method described in claim 3, it is characterized in that, whether being to have detecting the state of the network state parameters
Before effect state, the method further includes:Detect whether network is in normal operating conditions, wherein
Detect whether that the confirmation instruction for receiving repetition transmission detects if detecting the confirmation instruction for repeating transmission
The network is in abnormal operation;
It detects whether to receive Retransmission timeout instruction, if detecting the Retransmission timeout instruction, detects that the network is in
Abnormal operation.
8. a kind of control device of network transmission, which is characterized in that including:
Acquiring unit, for obtaining network state parameters, wherein the network state parameters indicate the delay of network;
Determination unit, for determining network bandwidth detection mode based on the network state parameters;
Selecting unit, for selecting congestion avoidance algorithm based on the network bandwidth detection mode;
Computing unit, the parameter for calculating congestion window;
Transmission unit, for the parameter according to the congestion window, to opposite end transmission data packet.
9. device according to claim 8, which is characterized in that the acquiring unit includes:
Determining module determines average delay for passing through the first formula, wherein first formula is:
AvgRTT=((scale-1) * avgRTT+rtt)/scale, wherein avgRTT indicates that average delay, scale indicate the
One delay parameter, rtt indicate current delay;
First determining module, for determining transmission delay based on the current delay;
Second determining module, for determining queuing delay based on the average delay and the transmission delay.
10. device according to claim 8, which is characterized in that the determination unit includes:
First detection module, whether the state for detecting the network state parameters is effective status;
First detecting module, if it is detected that the state of the network state parameters is effective status, for prolonging using based on network
When mode detection network bandwidth;
Second detecting module, if it is detected that the state of the network state parameters is not effective status and detects the network
In normal operating conditions, for the detection network bandwidth by the way of packet loss detection, wherein the detection mode includes described
The mode of packet loss detection;
Acquisition module, if it is detected that the state of the network state parameters is not effective status and detects that the network is in
Abnormal operation, the window parameter for obtaining preset congestion window.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710037306.8A CN108322401B (en) | 2017-01-17 | 2017-01-17 | Method and device for controlling network transmission congestion |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710037306.8A CN108322401B (en) | 2017-01-17 | 2017-01-17 | Method and device for controlling network transmission congestion |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108322401A true CN108322401A (en) | 2018-07-24 |
CN108322401B CN108322401B (en) | 2022-01-25 |
Family
ID=62892024
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710037306.8A Expired - Fee Related CN108322401B (en) | 2017-01-17 | 2017-01-17 | Method and device for controlling network transmission congestion |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108322401B (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110061925A (en) * | 2019-04-22 | 2019-07-26 | 深圳市瑞云科技有限公司 | A kind of image based on Cloud Server avoids congestion and accelerates transmission method |
CN110856214A (en) * | 2019-10-29 | 2020-02-28 | 广东省电信规划设计院有限公司 | TCP congestion control method and device |
CN111212126A (en) * | 2019-12-27 | 2020-05-29 | 百度在线网络技术(北京)有限公司 | Data transmission method, device, equipment and medium of block chain network |
CN111371692A (en) * | 2020-03-13 | 2020-07-03 | 中科驭数(北京)科技有限公司 | Window control method and device based on TCP (Transmission control protocol) and electronic equipment |
CN111404823A (en) * | 2020-03-05 | 2020-07-10 | 广东睿江云计算股份有限公司 | Method and system for tcp dynamic selection of congestion algorithm |
WO2021013260A1 (en) * | 2019-07-25 | 2021-01-28 | 中兴通讯股份有限公司 | Network transmission control method and apparatus |
CN113556213A (en) * | 2020-04-23 | 2021-10-26 | 华为技术有限公司 | Method for determining time-out Retransmission Time (RTO) and related device |
CN114389975A (en) * | 2022-02-08 | 2022-04-22 | 北京字节跳动网络技术有限公司 | Network bandwidth estimation method, device, system, electronic equipment and storage medium |
CN115086235A (en) * | 2022-05-31 | 2022-09-20 | 北京达佳互联信息技术有限公司 | Network congestion detection method and device, electronic equipment and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090073975A1 (en) * | 2007-09-19 | 2009-03-19 | Nec Corporation | Communication method and communication device |
CN101917342A (en) * | 2010-08-25 | 2010-12-15 | 天津理工大学 | Pervasive service-oriented adaptive network congestion control method |
CN105227484A (en) * | 2015-10-16 | 2016-01-06 | 中国人民解放军国防科学技术大学 | A kind of data transfer control method towards satellite network |
CN106059950A (en) * | 2016-05-25 | 2016-10-26 | 四川大学 | Adaptive network congestion control method based on SCPS-TP |
CN106131208A (en) * | 2016-08-09 | 2016-11-16 | 中南大学 | A kind of HTTP height concurrent data transfer method in data center network |
-
2017
- 2017-01-17 CN CN201710037306.8A patent/CN108322401B/en not_active Expired - Fee Related
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090073975A1 (en) * | 2007-09-19 | 2009-03-19 | Nec Corporation | Communication method and communication device |
CN101917342A (en) * | 2010-08-25 | 2010-12-15 | 天津理工大学 | Pervasive service-oriented adaptive network congestion control method |
CN105227484A (en) * | 2015-10-16 | 2016-01-06 | 中国人民解放军国防科学技术大学 | A kind of data transfer control method towards satellite network |
CN106059950A (en) * | 2016-05-25 | 2016-10-26 | 四川大学 | Adaptive network congestion control method based on SCPS-TP |
CN106131208A (en) * | 2016-08-09 | 2016-11-16 | 中南大学 | A kind of HTTP height concurrent data transfer method in data center network |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110061925A (en) * | 2019-04-22 | 2019-07-26 | 深圳市瑞云科技有限公司 | A kind of image based on Cloud Server avoids congestion and accelerates transmission method |
WO2021013260A1 (en) * | 2019-07-25 | 2021-01-28 | 中兴通讯股份有限公司 | Network transmission control method and apparatus |
CN110856214B (en) * | 2019-10-29 | 2023-01-10 | 广东省电信规划设计院有限公司 | TCP congestion control method and device |
CN110856214A (en) * | 2019-10-29 | 2020-02-28 | 广东省电信规划设计院有限公司 | TCP congestion control method and device |
CN111212126A (en) * | 2019-12-27 | 2020-05-29 | 百度在线网络技术(北京)有限公司 | Data transmission method, device, equipment and medium of block chain network |
CN111404823A (en) * | 2020-03-05 | 2020-07-10 | 广东睿江云计算股份有限公司 | Method and system for tcp dynamic selection of congestion algorithm |
CN111371692A (en) * | 2020-03-13 | 2020-07-03 | 中科驭数(北京)科技有限公司 | Window control method and device based on TCP (Transmission control protocol) and electronic equipment |
CN113556213A (en) * | 2020-04-23 | 2021-10-26 | 华为技术有限公司 | Method for determining time-out Retransmission Time (RTO) and related device |
CN113556213B (en) * | 2020-04-23 | 2022-12-06 | 华为技术有限公司 | Method for determining time-out Retransmission Time (RTO) and related device |
CN114389975A (en) * | 2022-02-08 | 2022-04-22 | 北京字节跳动网络技术有限公司 | Network bandwidth estimation method, device, system, electronic equipment and storage medium |
CN114389975B (en) * | 2022-02-08 | 2024-03-08 | 北京字节跳动网络技术有限公司 | Network bandwidth estimation method, device and system, electronic equipment and storage medium |
CN115086235A (en) * | 2022-05-31 | 2022-09-20 | 北京达佳互联信息技术有限公司 | Network congestion detection method and device, electronic equipment and storage medium |
CN115086235B (en) * | 2022-05-31 | 2024-03-26 | 北京达佳互联信息技术有限公司 | Network congestion detection method, device, electronic equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN108322401B (en) | 2022-01-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108322401A (en) | The control method and device of network transmission congestion | |
JP7173587B2 (en) | Packet transmission system and method | |
US10594596B2 (en) | Data transmission | |
CN105991462B (en) | Sending method, sending device and the system of transmission control protocol TCP data packet | |
US8661295B1 (en) | Monitoring and detecting causes of failures of network paths | |
CN103986548B (en) | A kind of method and terminal for determining packet loss reason | |
CN104836732B (en) | The automatic selecting method and system of network connection | |
EP3100420A2 (en) | Buffer sizing for multi-hop networks | |
US10355961B2 (en) | Network traffic capture analysis | |
CN101895466A (en) | Method for reducing influence of data packet disorder on SCTP multipath transmission | |
CN108809847A (en) | Realize the method, apparatus and network system of load balancing | |
CN102638407A (en) | Message forwarding method, device and network equipment | |
CN108696449A (en) | A kind of data dispatching method and device | |
CN106612284A (en) | Streaming data transport method and streaming data transport device | |
EP2993867A1 (en) | Communication system, control apparatus, and communication method | |
CN106302230B (en) | A kind of data transmission method and device | |
CN108429682A (en) | A kind of optimization method and system of network transmission link | |
JP5775214B2 (en) | Data packet loss reduction system and method using adaptive transmission queue length | |
JP2015154325A (en) | Apparatus selection network system, apparatus selection server, and program | |
CN110808917B (en) | Multilink aggregation data retransmission method and transmitting equipment | |
CN105939404B (en) | The acquisition methods and device of NAT resource | |
JP5128556B2 (en) | Traffic information collecting apparatus, traffic information collecting method and program thereof | |
CN110995593A (en) | Intelligent information forwarding method, device, equipment and storage medium | |
Pournaghshband et al. | End-to-End Detection of Middlebox Interference | |
CN105580330B (en) | A kind of method, apparatus and system of data transmission |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20220125 |
|
CF01 | Termination of patent right due to non-payment of annual fee |