CN105791144A - Method and apparatus for sharing link traffic - Google Patents

Method and apparatus for sharing link traffic Download PDF

Info

Publication number
CN105791144A
CN105791144A CN201410802351.4A CN201410802351A CN105791144A CN 105791144 A CN105791144 A CN 105791144A CN 201410802351 A CN201410802351 A CN 201410802351A CN 105791144 A CN105791144 A CN 105791144A
Authority
CN
China
Prior art keywords
interval
flow
timing statistics
line card
time
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.)
Pending
Application number
CN201410802351.4A
Other languages
Chinese (zh)
Inventor
万伟
朱明全
付东阳
连会彬
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
ZTE Corp
Original Assignee
ZTE Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by ZTE Corp filed Critical ZTE Corp
Priority to CN201410802351.4A priority Critical patent/CN105791144A/en
Priority to PCT/CN2015/078617 priority patent/WO2016095410A1/en
Publication of CN105791144A publication Critical patent/CN105791144A/en
Pending legal-status Critical Current

Links

Landscapes

  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The invention discloses a method and apparatus for sharing the link traffic. The method comprises the steps of receiving traffic report information sent by at least one line card, wherein the traffic report information comprises a traffic value counted in a statistical time interval and the statistical time interval; acquiring an end moment of the statistical time interval in traffic report information sent for the last time by the at least one line card in a sampling time interval, wherein the sampling time interval comprises at least one statistical time interval; according to the end moment, a time period of the statistical time interval in the sampling time interval, and preset parameters, obtaining a calculation time interval; and according to the traffic value counted in the statistical time interval and the statistical time interval, acquiring a total traffic value of data transmitted in the calculation time interval, and according to the total traffic value, carrying out link traffic sharing. By the method and apparatus for sharing the link traffic, which are disclosed by the invention, at least the problem that a real-time bandwidth with high accuracy cannot be accurately obtained in the prior art is solved.

Description

The method and apparatus that link flow is shared
Technical field
The present invention relates to the communications field, in particular to the method and apparatus that a kind of link flow is shared.
Background technology
Fig. 1 is the load sharing scene schematic diagram of traditional distributed system in correlation technique, as it is shown in figure 1, the outbound of user network adopts dual link up.Wherein a link is the transmission link rented, and the priority renting link is set to the highest by user.The converting flow of access arrangement, what prioritizing selection priority was the highest rent link forwards, and when the LI(link interface) flow that priority is the highest reaches max threshold, then is shared on the interface of low priority by partial discharge.By dynamic load share of command, the real-time bandwidth of total interface can be monitored, according to user configured load-sharing strategy, equilibrium assignment flow.
In traditional distributed system, each CPU disposal ability differs.Master control focuses on business datum, and the process of performance data should not be taken too much system resource, should not frequently gather line card flow, it is therefore desirable for line card periodically reports statistical value according to mastering capability, updates real-time bandwidth.But the statistics of line card is often integrated in chip, but each veneer carrying business is different, enabling time is strict conformance not, each veneer statistics gatherer cycle is inconsistent, gathering message arrival master control poor if having time, Fig. 2 is conventional statistics data acquisition flow figure in correlation technique, as shown in Figure 2, the t1 moment only has line card 1 to report flow, and in line card 2, in TB time interval, data just reported in the later t2 moment.
Therefore, in the related, it is impossible to be accurately obtained the real-time bandwidth that precision is higher.
Summary of the invention
The invention provides the method and apparatus that a kind of link flow is shared, with the problem at least solving to be accurately obtained the higher real-time bandwidth of precision in correlation technique.
According to an aspect of the present invention, provide a kind of method that link flow is shared, including: receiving the flow reporting information that at least one line card sends, wherein, described flow reporting information includes flow value and the described timing statistics interval of statistics in timing statistics interval;Obtaining the end time that the described timing statistics in the flow reporting information that at least one line card described sends for the last time in sampling time interval is interval, wherein, it is interval that described sampling time interval includes at least one described timing statistics;According to the time period that described timing statistics in described end time, described sampling time interval is interval, and parameter preset obtains calculating time interval;Flow value according to statistics in described timing statistics interval and described timing statistics interval obtain the total flow value of transmission in described calculating time interval, and carry out link flow according to described total flow value and share.
Preferably, before receiving the described flow reporting information that at least one line card described sends, described method also includes: sends at least one line card described and reports request message;Receive the flow reporting information that at least one line card described sends to include: receive at least one line card described according to described in report the described flow reporting information of request message transmission.
Preferably, obtain described calculating time interval include according to described end time, described timing statistics interval time period and described parameter preset: pass through formula: [tlast-λ*TL-S, tlast-λ*TL] obtain calculating time interval, wherein, tlastFor described end time, TLFor the time period that described timing statistics is interval, described parameter preset includes: S is described pre-designed evaluation time, and λ is the coefficient of variation between the sampling period between line card.
Preferably, described λ is be more than or equal to 0 and less than or equal to 2.
Preferably, obtain the total flow value of transmission in described calculating time interval according to the flow value of statistics in described timing statistics interval and described timing statistics interval to include: pass through formula: L=Q/ (tend-tstart) obtain described at least one line card flow value in each moment at least one timing statistics interval described, wherein, L is described at least one line card flow value in each moment at least one timing statistics interval described, Q is the flow value that at least one line card described is added up at least one timing statistics interval described, tendFor the end time that at least one timing statistics described is interval, tstartFor the initial time that at least one timing statistics described is interval;The flow value of whole line cards corresponding each moment at least one timing statistics interval described is added and obtains the flow total value of transmission of each moment at least one timing statistics interval described;According to the flow total value of transmission of each moment at least one timing statistics interval described, the flow total value addition corresponding whole moment in described calculating time interval is obtained described total flow value.
According to a further aspect in the invention, provide the device that link flow is shared, including: receiver module, for receiving the flow reporting information that at least one line card sends, wherein, described flow reporting information includes flow value and the described timing statistics interval of statistics in timing statistics interval;Acquisition module, for obtaining the end time in the described timing statistics interval in the flow reporting information that at least one line card described sends for the last time in sampling time interval, wherein, it is interval that described sampling time interval includes at least one described timing statistics;Processing module, for the time period interval according to described timing statistics in described end time, described sampling time interval, and parameter preset obtains calculating time interval;Link shares module, obtains the total flow value of transmission in described calculating time interval for the flow value according to statistics in described timing statistics interval and described timing statistics interval, and carries out link flow according to described total flow value and share.
Preferably, sending module, report request message for sending at least one line card described;Described receiver module, is additionally operable to receive the described flow reporting information reporting request message to send that at least one line card described sends according to described sending module.
Preferably, described processing module, it is additionally operable to pass through formula: [tlast-λ*TL-S, tlast-λ*TL] obtain calculating time interval, wherein, tlastFor described end time, TLFor the time period that described timing statistics is interval, described parameter preset includes: S is described pre-designed evaluation time, and λ is the coefficient of variation between the sampling period between line card.
Preferably, described λ is be more than or equal to 0 and less than or equal to 2.
Preferably, described link is shared module and is included: obtains unit, is used for passing through formula: L=Q/ (tend-tstart) obtain described at least one line card flow value in each moment at least one timing statistics interval described, wherein, L is described at least one line card flow value in each moment at least one timing statistics interval described, Q is the flow value that at least one line card described is added up at least one timing statistics interval described, tendFor the end time that at least one timing statistics described is interval, tstartFor the initial time that at least one timing statistics described is interval;First processing unit, obtains the flow total value of transmission of each moment at least one timing statistics interval described for being added by the flow value of whole line cards corresponding each moment at least one timing statistics interval described;Second processing unit, for obtaining described total flow value according to the flow total value of transmission of each moment at least one timing statistics interval described by the flow total value addition corresponding whole moment in described calculating time interval.
Pass through the present invention, adopt and receive the flow reporting information that at least one line card sends, wherein, this flow reporting information includes flow value and this timing statistics interval of statistics in timing statistics interval, obtain the end time that this timing statistics in the flow reporting information that this at least one line card sends for the last time in sampling time interval is interval, wherein, it is interval that this sampling time interval includes at least one this timing statistics, and according to this end time, the time period that in this sampling time interval, this timing statistics is interval, and parameter preset obtains calculating time interval, and the flow value and this timing statistics interval according to statistics in this timing statistics interval obtains the total flow value of transmission in this calculating time interval, and carry out link flow according to this total flow value and share.Owing to line card carries the timing statistics interval of the flow value of statistics in the flow reporting information reported to link flow sharing device, so, link flow sharing device reports, according to this, the flow value that timing statistics interval is accurately obtained in this timing statistics interval, thus being accurately obtained the real-time bandwidth that precision is higher.
Accompanying drawing explanation
Accompanying drawing described herein is used for providing a further understanding of the present invention, constitutes the part of the application, and the schematic description and description of the present invention is used for explaining the present invention, is not intended that inappropriate limitation of the present invention.In the accompanying drawings:
Fig. 1 is the load sharing scene schematic diagram of traditional distributed system in correlation technique;
Fig. 2 is conventional statistics data acquisition flow figure in correlation technique;
Fig. 3 is the flow chart of the method that link flow according to embodiments of the present invention is shared;
Fig. 4 is the structured flowchart of link flow sharing device according to embodiments of the present invention;
Fig. 5 is the preferred structure block diagram of link flow sharing device according to embodiments of the present invention;
Fig. 6 is the preferred structure block diagram that link flow sharing device link according to embodiments of the present invention shares module;
Fig. 7 is real-time bandwidth computational methods schematic diagram in distributed system according to embodiments of the present invention;
Fig. 8 is S second Mean Speed computational methods schematic diagram according to embodiments of the present invention;
Fig. 9 is the preferred schematic diagram of real-time bandwidth computational methods in distributed system according to embodiments of the present invention;
Figure 10 is 10 seconds real time rate computational methods schematic diagrams according to the preferred embodiment of the present invention.
Detailed description of the invention
Below with reference to accompanying drawing and describe the present invention in detail in conjunction with the embodiments.It should be noted that when not conflicting, the embodiment in the application and the feature in embodiment can be mutually combined.
Providing a kind of method that link flow is shared in the present embodiment, Fig. 3 is the flow chart of the method that link flow according to embodiments of the present invention is shared, as it is shown on figure 3, this flow process comprises the steps:
Step S302, receives the flow reporting information that at least one line card sends, and wherein, this flow reporting information includes flow value and this timing statistics interval of statistics in timing statistics interval.
Step S304, obtains the end time that this timing statistics in the flow reporting information that this at least one line card sends for the last time in sampling time interval is interval, and wherein, it is interval that this sampling time interval includes at least one this timing statistics.
Step S306, according to the time period that this timing statistics in this end time, this sampling time interval is interval, and parameter preset obtains calculating time interval;
Step S308, obtains in this calculating time interval the total flow value of transmission according to the flow value of statistics in this timing statistics interval and this timing statistics interval, and carries out link flow according to this total flow value and share.
So, owing to line card carries the timing statistics interval of the flow value of statistics in the flow reporting information reported to link flow sharing device, so, link flow sharing device reports, according to this, the flow value that timing statistics interval is accurately obtained in this timing statistics interval, thus being accurately obtained the real-time bandwidth that precision is higher.
Preferably, before receiving this flow reporting information that this at least one line card sends, send to this at least one line card and report request message, and receive this at least one line card according to this this flow reporting information reporting request message to send.So, link flow sharing device active request line card reports flow reporting information, with the control that each line card flow is reported by realization.
It should be noted that this flow reporting information can be { [tbegin, tend], Q}, wherein, [tbegin, tend] interval for this timing statistics, Q is the flow value of statistics in this timing statistics interval, owing to the collection of flow is lasting by line card, and end time (the i.e. t that timing statistics that therefore this line card current reports is intervalend) it is initial time (the i.e. t that this line card reports next time timing statistics is intervalbegin)。
Preferably, obtain described calculating time interval can adopt various ways according to described end time, described timing statistics interval time period and described parameter preset, for instance, it is possible to pass through formula: [tlast-λ*TL-S, tlast-λ*TL] obtain calculating time interval, wherein, tlastFor this end time, TLFor the time period that this timing statistics is interval, this parameter preset includes: S is this pre-designed evaluation time, and λ is the coefficient of variation between the sampling period between line card, so, the calculating time interval determined by above-mentioned formula can more accurately add up total flow value, shares realizing more accurate link.
Preferably, this λ is be more than or equal to 0 and less than or equal to 2.In a kind of possible implementation of the present invention, if the flow obtained can be directly used in calculating every time, it is incomplete to be absent from certain timing statistics interval flow, or during only one of which line card, this λ can take 0;The time delay when the timing statistics interval that each line card reports exists, this λ can take 2, is merely illustrative a kind of implementation of offer, and this is not construed as limiting by the present invention.
It should be noted that in actual applications, the setting of this S may be excessive, thus causing tlast-λ*TLThe result that-S draws is negative value, now, by this tlast-λ*TL-S takes 0.
When the flow value according to statistics in described timing statistics interval and described timing statistics interval obtain the total flow value of transmission in described calculating time interval, multiple processing mode can also be adopted, introduce a kind of preferred processing mode at this, first pass through formula: L=Q/ (tend-tstart) flow value in each moment that obtains this at least one line card in this at least one timing statistics interval, wherein, L is this at least one line card in this at least one the timing statistics interval flow value in each moment, Q is the flow value that this at least one line card is added up in this at least one timing statistics interval, tendFor the end time that this at least one timing statistics is interval, tstartFor the initial time that this at least one timing statistics is interval;Afterwards, the flow value of whole line cards corresponding each moment in this at least one timing statistics interval is added and obtains the flow total value of transmission of each moment in this at least one timing statistics interval, finally, according to the flow total value of transmission of each moment in this at least one timing statistics interval, the flow total value addition corresponding whole moment in this calculating time interval is obtained this total flow value, so, by the above-mentioned formula flow value in each moment that can calculate at least one line card more accurately in this at least one timing statistics interval.
Additionally providing a kind of link flow sharing device in the present embodiment, this device is used for realizing above-described embodiment and preferred implementation, has be carried out repeating no more of explanation.As used below, term " module " can realize the software of predetermined function and/or the combination of hardware.Although the device described by following example preferably realizes with software, but hardware, or the realization of the combination of software and hardware is also likely to and is contemplated.
Fig. 4 is the structured flowchart of link flow sharing device according to embodiments of the present invention, and as shown in Figure 4, this link flow sharing device 40 includes: receiver module 42, acquisition module 44, processing module 46 and link share module 48, below this device are illustrated.
Receiver module 42, for receiving the flow reporting information that at least one line card sends, wherein, this flow reporting information includes flow value and this timing statistics interval of statistics in timing statistics interval;Acquisition module 44, it is connected to above-mentioned acquisition module 44, for obtaining the end time in this timing statistics interval in the flow reporting information that at least one line card that this first receiver module 42 receives in sampling time interval sends for the last time, wherein, this sampling time interval includes at least one this timing statistics interval;Processing module 46, is connected to above-mentioned acquisition module 44, and for the time period that this timing statistics in this end time obtained according to this acquisition module 44, this sampling time interval is interval, and parameter preset obtains calculating time interval;Link shares module 48, it is connected to above-mentioned processing module 46, obtain the total flow value of transmission in this calculating time interval for the flow value according to statistics in this timing statistics interval and this timing statistics interval, and the total flow value obtained according to this processing module 46 carries out link flow and shares.
Fig. 5 is the preferred structure block diagram of link flow sharing device according to embodiments of the present invention, as it is shown in figure 5, this device is except including all structures shown in Fig. 4, also includes: sending module 52, below this sending module 52 is illustrated.
Sending module 52, is connected to above-mentioned receiver module 42, for, before this receiver module 42 receives this flow reporting information that this at least one line card sends, sending to this at least one line card and report request message;This receiver module 42, is additionally operable to receive this flow reporting information reporting request message to send that this at least one line card sends according to this sending module 52.So, link flow sharing device active request line card reports flow reporting information, with the control that each line card flow is reported by realization.
Preferably, above-mentioned processing module 46, it is additionally operable to pass through formula: [tlast-λ*TL-S, tlast-λ*TL] obtain calculating time interval, wherein, tlastFor this end time, TLFor the time period that this timing statistics is interval, this parameter preset includes: S is this pre-designed evaluation time, and λ is the coefficient of variation between the sampling period between line card.
Preferably, this λ is be more than or equal to 0 and less than or equal to 2.
Fig. 6 is the preferred structure block diagram that link flow sharing device link according to embodiments of the present invention shares module 48, as shown in Figure 6, this link is shared module 48 and is included: obtains unit the 62, first processing unit 64 and the second processing unit 66, below this link is shared module 48 and illustrate.
Obtain unit 62, be used for passing through formula: L=Q/ (tend-tstart) obtain described at least one line card flow value in each moment at least one timing statistics interval described, wherein, L is described at least one line card flow value in each moment at least one timing statistics interval described, Q is the flow value that at least one line card described is added up at least one timing statistics interval described, tendFor the end time that at least one timing statistics described is interval, tstartFor the initial time that at least one timing statistics described is interval;First processing unit 64, is connected to above-mentioned acquisition unit 62, obtains the flow total value of transmission of each moment at least one timing statistics interval described for being added by the flow value of whole line cards corresponding each moment at least one timing statistics interval described;Second processing unit 66, it is connected to above-mentioned first processing unit 64, for the flow total value addition corresponding whole moment in described calculating time interval being obtained described total flow value according to the flow total value of transmission of each moment at least one timing statistics interval described.It should be noted that above-mentioned link flow sharing device can be master control borad.
For the problems referred to above in correlation technique, the real-time bandwidth higher in order to obtain precision, it is necessary to the problem solving three aspects: first, how the period of performance of balanced master control and line card is inconsistent, and master control can take performance data in the self-defined cycle, it is impossible to couples with line card;Secondly, if a certain interface data takes from polylith veneer, the moment that each veneer statistical value periodically reports is inconsistent, and the flow renewal once received in master control is a certain piece of veneer, cannot estimate other veneer flow values and on call time, the real-time bandwidth therefore calculated is less than normal.Finally, in distributed system, statistical value has time delay in message process, when portfolio is big, the statistical value that sampling obtains is delivered to master cpu and inter-process messages, and mutual often intermittence is delayed, the flow that master control receives was before the TA time in fact, and the error of the flow that TA is interval needs to eliminate.
Based in the above-mentioned distributed system to existing in correlation technique, measurement period is inconsistent, bandwidth calculation Statistic Source not accurate, many cannot the analysis of the problem of computation bandwidth in real time, in the present embodiment, it is proposed that the scheme of real-time bandwidth data acquisition and calculating in a kind of distributed system.Fig. 7 is real-time bandwidth computational methods schematic diagram in distributed system according to embodiments of the present invention, as it is shown in fig. 7, the program generally comprises following process: resident statistics agent on line card, and Statistic Source is (T periodicallyL) past statistics agent send statistical data.Statistical fractals end in master control, periodically (TR) the process requested statistical result of past resident statistics agent, statistics agent returns statistics request to statistical fractals end, and statistical fractals end passes to data processing module after carrying out converging storage according to the flow that different veneers report.Data processing module, carries out real-time bandwidth calculating according to the flow reported and time according to corresponding strategy by flow value.Wherein, T in figureRFor master control sampling period, TLFor the line card sampling period.
The method below wherein real-time bandwidth and S second Mean Speed calculated illustrates.
S1, resident statistics agent on line card.It is used for collecting and storing the statistical data of Statistic Source, responds the request message that master control is sent simultaneously.Owing to statistics agent stores statistical data, (cycle is T once to ask message to be probably repeatedly Statistic SourceL) data accumulation that reports.In Fig. 7, two Statistic Sources have all returned response ACK to service end.
S2, reports data on flows QtCarrying flow time of origin interval, namely reporting unit is { [tbegin,tend],Qt, such as Fig. 7.Owing to the collection of flow is lasting, the t that therefore the N+1 time returnsbeginT for n-thend.The time interval reported, end to end, tlastFor the t that the last flow renewal time namely reports for the last timeend
S3, statistical disposition module, by axle sliced time second, by flow QtTemporally proportional linearity is divided in corresponding time interval.Fig. 8 is S second Mean Speed computational methods schematic diagram according to embodiments of the present invention, as shown in Figure 8, in this interval, and every second flow qt=Qt/(tend-tbegin) such as Fig. 8 (a).If the t second has repeatedly flow to report, t second flow: qt=qt 1+qt 2+qt 3……+qt n
S4, sets the time interval that can be used for computation rate.It is confidence interval referred to herein as this interval.Consider that each plate collection period is inconsistent, take coefficient lambda and reflect above-mentioned difference.TL-MAXTake each veneer report cycle maximum.Then on time shaft [0, tlast-λ*TL-MAX] for confidence interval, such as Fig. 8 (b).
Below the value of λ is illustrated.
λ is 0.Representing that the flow every time obtained can be directly used in calculating, it is incomplete to be absent from certain time interval flow.Namely physical interface is so, and its flow is from a certain line card.
λ is 2.Namely each veneer reports its time beginning and end of flow arriving data processing module maximum to differ two TL-MAX.Typically when two line card cycles were homogeneously all TL-MAX, in Fig. 8 (c).tlastThe flow of moment veneer 1 updates, but the flow Q of veneer 2t 2Message is likely to could update after a while master control, therefore Q in figuret 1Qt 2The time interval flow covered is all on the low side, should not be used to calculated flow rate.Limiting case down-off time interval on the low side is 2*TL-MAX
By the impact of each veneer cycle differentiation in real system, the suggestion of λ span is [0,2].
S5, takes [tlast-λ*TL-MAX-S, tlast-λ*TL-MAX] t flow rate calculation S second Mean Speed in time interval.The flow divided by step 3 in S second interval adds up.That is:
qt-1+qt-2+qt-3……+qt-s
S6, the flow in untrusted interval is probably on the low side, it is impossible to as the reference that current bandwidth calculates.Follow-up along with the continuous updating of flow, it is in confidence interval just meaningful when the time.
In the above-described embodiments, by statistics agent resident in step 1, the time difference that each Statistic Source reports, T are shieldedLAnd TRAllowing inconsistent, master control can configure T according to current traffic situationsR, effectively reduce interacting message between master control and line card, solve the first problem that legacy equipment exists;The time that statistics agent record flow updates, together report with flow, owing in step S2, timestamp is carried at data source place, eliminate the time loss of Asynchronous message transfer;The calculation of sampling step S3, S4, S5 eliminates the impact that in distributed system, real-time bandwidth is calculated by Statistic Source straddle very well.nullIllustratively,Illustrate in conjunction with Fig. 9,As shown in Figure 9,The embodiment of the present invention is with two line cards (being designated as line card 1 and line card 2 respectively),And to include two timing statistics intervals in a sampling time interval be that example illustrates,Wherein,Sampling time interval takes [t0,t4],Two timing statistics intervals of line card 1 are designated as timing statistics interval 10 and timing statistics interval 11 respectively,Two timing statistics intervals of line card 2 are designated as timing statistics interval 20 and timing statistics interval 21 respectively,Q1 is the flow value that line card 1 reports,Timing statistics interval 10 takes [t1,t2],Timing statistics interval 10 takes [t2,t3],Q2 is the flow value that line card 2 reports,Timing statistics interval 20 takes [t0,t1],Timing statistics interval 21 takes [t1,t2],Then link flow sharing device is according to the above-mentioned formula flow value in each moment that calculates line card 1 respectively in timing statistics interval 10、The flow value in line card 1 each moment in timing statistics interval 11 and the flow value in line card 2 each moment in timing statistics interval 20 and line card 2 flow value in each moment in timing statistics interval 21,If the calculating time interval obtained is [t1,t2],Then the flow value of the line card 1 in each moment in this calculating time interval and the flow value of line card 2 are added the transmission total value obtaining each moment line card 1 and line card 2 by this link flow sharing device,And the transmission total value addition in each moment in this calculating time interval is obtained total flow value,So,Link flow sharing device is determining that this total flow value is when reaching the preset flow upper limit,Carry out link flow to share,Reduce the load sharing weight of this interface,Flow too much on this interface is made to be distributed on other interfaces,It should be noted that,Include two timing statistics intervals in above-mentioned two line card and sampling time interval to be merely illustrative,This is not construed as limiting by the present invention.
Preferably, above-mentioned S can take 10 seconds, for two line cards, the preferred embodiment for the present invention is illustrated below.
In this embodiment, the real time rate that in distributed system, load sharing calculates is generally adopted 10 seconds Mean Speeds.Set master control and equal 10 second sampling period of line card.This link flow sharing arrangement includes following process:
S1, link flow sharing device can be master control borad, rents link port and enables dynamic load share of command function, and link priority is configured non-zero, and setting bandwidth transfinites alarm threshold and the preset flow upper limit, and configuration-direct is following table such as:
S2, master control borad and line card sampling time acquiescence are 10 seconds, and link aggregation (Smartgroup1) mouth acquiescence enables performance statistics simultaneously.Configuration straddle link aggregation port service, Statistic Source perception link aggregation mouth flow collection is from line card 1 and line card 2, and configuration-direct is as follows:
S3, Statistic Source sends the statistical data of link aggregation mouth for every 10 seconds to statistics agent on two line plates, and is done data accumulation by statistics agent, the present flow rate that statistics agent receives, recording time interval, the starting point of time interval is exactly the time point that last data on flows updates.Such as { [t0, t1], Q1} etc..Statistics agent needs to process following two situation:
A () gathers data and does not report, gathering data in (b) measurement period is zero;
The first situation, Statistic Source is not fully complete this collection, currently cannot obtain definite flow, and time record does not also update.Second case, gathering data is zero, and this situation flow generates, and therefore time value needs to update, it is necessary to update time record.
S4, on master control borad, every 10 seconds of performance statistics is to the statistics agent request data of line card, currently cumulative data are returned to master control borad (such as by statistics agent, can be the master control performance statistic module (sharing the partial function of module 48 with above-mentioned link) of master control borad), afterwards master control borad these data are directly synchronized to interface capability statistics process carry out bandwidth calculation.Fig. 9 is 10 seconds real time rate computational methods schematic diagrams according to the preferred embodiment of the present invention, and as shown in Figure 10, n-th is asked, and line card 1 returns { [t0,t1],Q1, line card 2 returns { [t0',t1'],Q1', the N+1 time request, line card 1 returns { [t1,t2],Q2, line card 2 returns { [t1',t2'],Q2'}。
S5, statistical disposition module is with 1 second for granularity division flow.Q1, Q1', Q2, Q2' share data on flows according to the time interval reported equally toward relevant position respectively.Taking λ for data source straddle interface is 2, namely thinks that difference that the timing of Statistic Source between veneer reports is not over a measurement period, is finally updated 10 seconds Mean Speeds by algorithm above.The time delay aspect of flow rate, the selection calculating time interval causes that the data in nearest 20 seconds cannot participate in this and calculate, and therefore the calculated speed of above-mentioned interface exists the time delay of 20 seconds.
S6, real time rate is synchronized to dynamic load share of command module, and obtains peak load and share bandwidth, if it exceeds 80%, then adjust associated loadings and share the weight of list item, reduce the load sharing weight of this interface, make flow too much on this interface be distributed on other interfaces.
By above-mentioned preferred implementation, report, according to this, the flow value that timing statistics interval is accurately obtained in this timing statistics interval, thus being accurately obtained the real-time bandwidth that precision is higher.
Obviously, those skilled in the art should be understood that, each module of the above-mentioned present invention or each step can realize with general calculation element, they can concentrate on single calculation element, or it is distributed on the network that multiple calculation element forms, alternatively, they can realize with the executable program code of calculation element, thus, can be stored in storage device is performed by calculation element, and in some cases, shown or described step can be performed with the order being different from herein, or they are fabricated to respectively each integrated circuit modules, or the multiple modules in them or step are fabricated to single integrated circuit module realize.So, the present invention is not restricted to the combination of any specific hardware and software.
The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for a person skilled in the art, the present invention can have various modifications and variations.All within the spirit and principles in the present invention, any amendment of making, equivalent replacement, improvement etc., should be included within protection scope of the present invention.

Claims (10)

1. the method that a link flow is shared, it is characterised in that including:
Receiving the flow reporting information that at least one line card sends, wherein, described flow reporting information includes flow value and the described timing statistics interval of statistics in timing statistics interval;
Obtaining the end time that the described timing statistics in the flow reporting information that at least one line card described sends for the last time in sampling time interval is interval, wherein, it is interval that described sampling time interval includes at least one described timing statistics;
According to the time period that described timing statistics in described end time, described sampling time interval is interval, and parameter preset obtains calculating time interval;
Flow value according to statistics in described timing statistics interval and described timing statistics interval obtain the total flow value of transmission in described calculating time interval, and carry out link flow according to described total flow value and share.
2. method according to claim 1, it is characterised in that before receiving the described flow reporting information that at least one line card described sends, described method also includes:
Send at least one line card described and report request message;
The flow reporting information receiving described at least one line card transmission includes:
Receive at least one line card described according to described in report the described flow reporting information of request message transmission.
3. method according to claim 1, it is characterised in that the time period interval according to described end time, described timing statistics and described parameter preset obtain described calculating time interval and include:
Pass through formula: [tlast-λ*TL-S, tlast-λ*TL] obtain calculating time interval, wherein, tlastFor described end time, TLFor the time period that described timing statistics is interval, described parameter preset includes: S is described pre-designed evaluation time, and λ is the coefficient of variation between the sampling period between line card.
4. method according to claim 3, it is characterised in that described λ is be more than or equal to 0 and less than or equal to 2.
5. the method according to any one of Claims 1-4, it is characterised in that obtain the total flow value of transmission in described calculating time interval according to the flow value of statistics in described timing statistics interval and described timing statistics interval and include:
Pass through formula: L=Q/ (tend-tstart) obtain described at least one line card flow value in each moment at least one timing statistics interval described, wherein, L is described at least one line card flow value in each moment at least one timing statistics interval described, Q is the flow value that at least one line card described is added up at least one timing statistics interval described, tendFor the end time that at least one timing statistics described is interval, tstartFor the initial time that at least one timing statistics described is interval;
The flow value of whole line cards corresponding each moment at least one timing statistics interval described is added and obtains the flow total value of transmission of each moment at least one timing statistics interval described;
According to the flow total value of transmission of each moment at least one timing statistics interval described, the flow total value addition corresponding whole moment in described calculating time interval is obtained described total flow value.
6. the device that a link flow is shared, it is characterised in that including:
Receiver module, for receiving the flow reporting information that at least one line card sends, wherein, described flow reporting information includes flow value and the described timing statistics interval of statistics in timing statistics interval;
Acquisition module, for obtaining the end time in the described timing statistics interval in the flow reporting information that at least one line card described sends for the last time in sampling time interval, wherein, it is interval that described sampling time interval includes at least one described timing statistics;
Processing module, for the time period interval according to described timing statistics in described end time, described sampling time interval, and parameter preset obtains calculating time interval;
Link shares module, obtains the total flow value of transmission in described calculating time interval for the flow value according to statistics in described timing statistics interval and described timing statistics interval, and carries out link flow according to described total flow value and share.
7. device according to claim 6, it is characterised in that described device also includes:
Sending module, reports request message for sending at least one line card described;
Described receiver module, is additionally operable to receive the described flow reporting information reporting request message to send that at least one line card described sends according to described sending module.
8. device according to claim 6, it is characterised in that
Described processing module, is additionally operable to pass through formula: [tlast-λ*TL-S, tlast-λ*TL] obtain calculating time interval, wherein, tlastFor described end time, TLFor the time period that described timing statistics is interval, described parameter preset includes: S is described pre-designed evaluation time, and λ is the coefficient of variation between the sampling period between line card.
9. device according to claim 8, it is characterised in that described λ is be more than or equal to 0 and less than or equal to 2.
10. the device according to any one of claim 6 to 9, it is characterised in that described link is shared module and included:
Obtain unit, be used for passing through formula: L=Q/ (tend-tstart) obtain described at least one line card flow value in each moment at least one timing statistics interval described, wherein, L is described at least one line card flow value in each moment at least one timing statistics interval described, Q is the flow value that at least one line card described is added up at least one timing statistics interval described, tendFor the end time that at least one timing statistics described is interval, tstartFor the initial time that at least one timing statistics described is interval;
First processing unit, obtains the flow total value of transmission of each moment at least one timing statistics interval described for being added by the flow value of whole line cards corresponding each moment at least one timing statistics interval described;
Second processing unit, for obtaining described total flow value according to the flow total value of transmission of each moment at least one timing statistics interval described by the flow total value addition corresponding whole moment in described calculating time interval.
CN201410802351.4A 2014-12-19 2014-12-19 Method and apparatus for sharing link traffic Pending CN105791144A (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201410802351.4A CN105791144A (en) 2014-12-19 2014-12-19 Method and apparatus for sharing link traffic
PCT/CN2015/078617 WO2016095410A1 (en) 2014-12-19 2015-05-08 Link traffic distributing method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410802351.4A CN105791144A (en) 2014-12-19 2014-12-19 Method and apparatus for sharing link traffic

Publications (1)

Publication Number Publication Date
CN105791144A true CN105791144A (en) 2016-07-20

Family

ID=56125763

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410802351.4A Pending CN105791144A (en) 2014-12-19 2014-12-19 Method and apparatus for sharing link traffic

Country Status (2)

Country Link
CN (1) CN105791144A (en)
WO (1) WO2016095410A1 (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107733805A (en) * 2016-08-12 2018-02-23 腾讯科技(深圳)有限公司 Business load dispatching method and device
CN111211987A (en) * 2019-11-29 2020-05-29 清华大学 Method and system for dynamically adjusting flow in network, electronic equipment and storage medium
CN112714028A (en) * 2021-01-29 2021-04-27 紫光云技术有限公司 Monitoring method for load balancing SLB flow and data packet parameters
CN113328906A (en) * 2021-04-22 2021-08-31 成都欧珀通信科技有限公司 Flow real-time monitoring method and device, storage medium and electronic equipment
CN113347107A (en) * 2020-03-02 2021-09-03 ***通信集团浙江有限公司 Flow scheduling method and device based on uplink message and computing equipment
WO2022100146A1 (en) * 2020-11-10 2022-05-19 北京市天元网络技术股份有限公司 Internet performance monitoring method and system

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110011926B (en) * 2019-03-07 2022-08-23 新华三技术有限公司 Method, device, equipment and storage medium for adjusting message sending time

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1984119A (en) * 2005-12-13 2007-06-20 中兴通讯股份有限公司 Method for controlling flow by time-division technology
CN101257415A (en) * 2008-01-18 2008-09-03 东南大学 Network flow real time self-adapting measuring method based on fixed storage space
CN102143042A (en) * 2010-07-09 2011-08-03 华为技术有限公司 Virtual cluster router system and flow sharing method thereof, controller and sub routers
CN102237957A (en) * 2010-05-07 2011-11-09 电信科学技术研究院 Channel state information reporting method, system and device
US20140185454A1 (en) * 2006-03-09 2014-07-03 Firetide, Inc. Effective Bandwidth Path Metric and Path Computation Method for Wireless Mesh Networks with Wired Links

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020085547A1 (en) * 1998-09-17 2002-07-04 Mcnamara Tod W. High density network topology
US7027448B2 (en) * 2000-04-21 2006-04-11 At&T Corp. System and method for deriving traffic demands for a packet-switched network
CN100566288C (en) * 2004-12-28 2009-12-02 杭州华三通信技术有限公司 On router or switch, carry out the method and apparatus of multiple exit traffic management
CN102932268A (en) * 2012-11-05 2013-02-13 杭州迪普科技有限公司 Data packet transmitting method and device

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1984119A (en) * 2005-12-13 2007-06-20 中兴通讯股份有限公司 Method for controlling flow by time-division technology
US20140185454A1 (en) * 2006-03-09 2014-07-03 Firetide, Inc. Effective Bandwidth Path Metric and Path Computation Method for Wireless Mesh Networks with Wired Links
CN101257415A (en) * 2008-01-18 2008-09-03 东南大学 Network flow real time self-adapting measuring method based on fixed storage space
CN102237957A (en) * 2010-05-07 2011-11-09 电信科学技术研究院 Channel state information reporting method, system and device
CN102143042A (en) * 2010-07-09 2011-08-03 华为技术有限公司 Virtual cluster router system and flow sharing method thereof, controller and sub routers

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107733805A (en) * 2016-08-12 2018-02-23 腾讯科技(深圳)有限公司 Business load dispatching method and device
CN107733805B (en) * 2016-08-12 2021-04-16 腾讯科技(深圳)有限公司 Service load scheduling method and device
US11025709B2 (en) 2016-08-12 2021-06-01 Tencent Technology (Shenzhen) Company Limited Load processing method and apparatus
CN111211987A (en) * 2019-11-29 2020-05-29 清华大学 Method and system for dynamically adjusting flow in network, electronic equipment and storage medium
CN111211987B (en) * 2019-11-29 2021-07-27 清华大学 Method and system for dynamically adjusting flow in network, electronic equipment and storage medium
CN113347107A (en) * 2020-03-02 2021-09-03 ***通信集团浙江有限公司 Flow scheduling method and device based on uplink message and computing equipment
WO2022100146A1 (en) * 2020-11-10 2022-05-19 北京市天元网络技术股份有限公司 Internet performance monitoring method and system
CN112714028A (en) * 2021-01-29 2021-04-27 紫光云技术有限公司 Monitoring method for load balancing SLB flow and data packet parameters
CN113328906A (en) * 2021-04-22 2021-08-31 成都欧珀通信科技有限公司 Flow real-time monitoring method and device, storage medium and electronic equipment
CN113328906B (en) * 2021-04-22 2023-01-06 成都欧珀通信科技有限公司 Flow real-time monitoring method and device, storage medium and electronic equipment

Also Published As

Publication number Publication date
WO2016095410A1 (en) 2016-06-23

Similar Documents

Publication Publication Date Title
CN105791144A (en) Method and apparatus for sharing link traffic
CN107508719B (en) Method and device for measuring network time delay and network node
CN101272290B (en) Measuring method, measuring apparatus for route congestion state in IP network
EP1931084A1 (en) A method for getting the link estimating parameters
US10623313B2 (en) Path computation method, message responding method, and related device
CN105591843A (en) Network performance detection method and system based on receiving end in TCP transmission stream
US8437264B1 (en) Link microbenchmarking with idle link correction
US9331803B2 (en) System and method of synchronization among a control apparatus and a plurality of terminals
CN110855741A (en) Service self-adaptive access method and device, storage medium and electronic device
CN117424638B (en) Network slice management method, device and storage medium based on satellite-ground fusion
WO2021147319A1 (en) Data processing method, apparatus, device, and medium
CN100512162C (en) Method and system and device for collecting flow data
US7761528B2 (en) Method for processing multiple continuous Top-K queries
JP2003534715A (en) Method and system for measuring one-way delay variation
CN105282025A (en) Method of determining end-to-end routing and apparatus thereof
CN109787861A (en) Network data delay control method
CN104270466A (en) Data reporting method and related equipment
CN109213778B (en) Stream data sliding window gathering query method
CN108234658A (en) A kind of method, apparatus and server of aware services device cluster health status
CN104579812A (en) Method for measuring available bandwidth of computer network
CN108347390A (en) The method and device of Limit Rate in a kind of network equipment
CN110086592B (en) Data processing method and device based on virtual time axis and electronic equipment
CN105611406A (en) Method for access network service provider to monitor delay feature from user to video server
CN112312209A (en) Comprehensive alarm generation method, device, server and storage medium
CN106301899A (en) A kind of quality to network direct broadcasting content carries out the method and system of quantitative analysis

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WD01 Invention patent application deemed withdrawn after publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20160720