CN112771816B

CN112771816B - Method and device for predicting network rate

Info

Publication number: CN112771816B
Application number: CN201880097672.5A
Authority: CN
Inventors: 罗勇
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2018-09-19
Filing date: 2018-09-19
Publication date: 2022-09-16
Anticipated expiration: 2038-09-19
Also published as: WO2020056633A1; CN112771816A

Abstract

A method and a device for predicting network speed are provided. The method comprises the following steps: the prediction device collects flow data of an object to be measured (S201); carrying out flow grouping on the collected flow data according to the user label and the service type (S202); calculating a network peak rate of each traffic packet (S203); and adding the network peak rates of the flow packets to obtain the network peak rate of the object to be detected (S204). The peak rate of the network can be accurately predicted.

Description

Method and device for predicting network rate

Technical Field

The present application relates to the field of network technologies, and in particular, to a method and a device for predicting a network rate.

Background

With the continuous development of network technology, the scale of the existing network is continuously enlarged, and the heterogeneity and complexity of the network are increasingly improved. The method can accurately grasp the network state in real time, especially the bandwidth information required by the network, and has important effects on improving the network performance and optimizing the service quality of users. The bandwidth required by the network is determined by the network peak rate. Therefore, the prediction of the network peak rate is a very important basic work for network planning and design; the network peak rate here refers to the maximum rate that can be achieved by network data transmission on the premise of a certain packet loss rate; if the "east-west traffic" (i.e. the access traffic between two clients in the network) in the network is not considered, the peak rate of the whole network can be gradually converged by the peak rate of each network element, so that it is a precondition to predict the peak rate of a single network element or a certain link on the single network element.

The future daily peak rate for a network element or a link in a network can be described by the following formula:

the peak rate of a day in the future is the peak rate of the day of today + the variation of the first time period + the variation of the second time period + the random variation; the first time period variable quantity is used for representing the change condition of the network peak value rate in a longer time period, such as gradual increase of the network peak value rate according to months, seasons or years; the second time period variation is used to represent the variation of the network peak rate in a shorter time period, for example, the variation of the network peak rate according to a cycle of a week (for example, the network peak rate may rise when every week reaches the weekend), and the random variation represents the random variation of the network peak rate every day.

In the prior art, the conventional network peak rate prediction usually calculates an expected value and a standard deviation of a network peak rate according to average data obtained by sampling, and then predicts a network peak rate possible in a future day in a short period according to the expected value and the standard deviation, so that only the variation of a second time period, that is, the variation of the network peak rate in a shorter time period can be predicted in such a manner. The prediction results are not accurate.

Disclosure of Invention

The technical problem to be solved by the embodiments of the present application is to provide a method and a device for predicting a network rate, so as to accurately predict a peak rate of a network.

In a first aspect, an embodiment of the present application provides a method for predicting a network rate, which may include:

the method comprises the steps that a prediction device collects flow data of an object to be measured, wherein the object to be measured is a network element to be measured or a link to be measured;

carrying out flow grouping on the collected flow data according to a user label and a service type, wherein the user label is used for indicating a network package signed by a user and/or an identity of the user, and the service type comprises an internet service, an internet private line service and a video service;

calculating the network peak rate of each flow packet;

and adding the network peak rates of the flow packets to obtain the network peak rate of the object to be detected.

By means of grouping calculation, accurate random variation of the network peak rate can be obtained, and prediction accuracy is improved.

In a possible implementation manner, the traffic packet includes an internet service group, and when a network peak rate of the internet service group is calculated, the method includes:

according to the first traffic data corresponding to the internet access service group, the daily peak average rate of a single user in the data acquisition period is counted;

calculating a first expected value and a first standard deviation of the daily peak average rate of the single user according to the data obtained by statistics;

calculating the single-user network peak rate of the internet service group according to the first expected value and the first standard deviation;

and calculating the network peak rate of the internet service group according to the number of the users of the internet service group and the single-user network peak rate.

In one possible implementation, the method further includes:

calculating the peak average rate of the preset period of a single user according to the first flow data and the preset period to obtain expected values of at least two preset periods, wherein the preset period is more than 1 day;

performing curve fitting according to the expected values of the at least two preset periods to obtain a peak average rate increase function of the preset periods;

and calculating the peak rate of the single-user network of the internet service group according to the first expected value, the first standard deviation and the peak average rate increase function of the preset period.

By introducing calculation of the long period variable quantity, the long period growth rate of the network peak rate can be obtained, and therefore more accurate network peak rate can be predicted.

In one possible implementation, the traffic packet includes a private interconnection group, and when calculating a network peak rate of the private interconnection group, the method includes:

acquiring a signed bandwidth of the private interconnection service from an operation support system, and taking the signed bandwidth as a single-user network peak rate of the private interconnection traffic group;

and calculating the network peak rate of the interconnected private line group according to the number of users of the interconnected private line group and the single-user network peak rate.

In one possible implementation, the traffic packet includes a unicast video group, and when calculating a network peak rate of the unicast video group, the method includes:

if the average code rate of the unicast video is the same, taking the product of the average code rate and a first preset coefficient as the peak rate of the single user network of the unicast video group;

if the average code rates of the unicast videos are different, counting the daily peak average rate of a single user during data acquisition according to the second flow data corresponding to the unicast video group; calculating a second expected value and a second standard deviation of the daily peak average rate of the single user according to the data obtained by statistics; calculating the single user network peak rate of the unicast video group according to the second expected value and the second standard deviation;

and calculating the network peak rate of the unicast video group according to the number of the users of the unicast video group and the network peak rate of the single user.

In one possible implementation, the traffic packet includes a multicast video group, and when calculating a network peak rate of the multicast video group, the method includes:

if the multicast replication point for replicating the multicast video playing is not located on the object to be tested, taking the product of the average code rate of the multicast video and a second preset coefficient as the peak rate of the single-user network of the multicast video group;

and calculating the network peak rate of the multicast video group according to the number of users of the multicast video group and the network peak rate of a single user.

if the multicast replication point for replicating the multicast video playing is located on the object to be tested, determining a third expected value and a third standard deviation of the online c channels according to the number c of channels required to be provided by the service and the number u of users borne by the object to be tested;

determining the online number of peak frequency channels according to the third expected value and the third standard deviation;

and calculating the network peak rate of the multicast video group according to the peak channel online number and the average code rate of the multicast video.

In one possible implementation, the method further includes:

according to the flow data corresponding to the flow grouping, the daily peak value online rate of the number of users in the flow grouping during data acquisition is counted;

calculating expected values and standard deviations of daily peak online rates according to the data obtained through statistics;

calculating a user online rate threshold according to the expected value and the standard deviation of the daily peak online rate;

and taking the product of the online rate threshold of the user and the number of users in the flow packet as the number of users when the peak rate of the flow packet network is calculated.

The maximum possible online user number is reflected by introducing the online rate threshold of the user, so that the method can be closer to the actual situation and obtain a better prediction result.

In one possible implementation, the method further includes:

and allocating network bandwidth resources to the object to be detected according to the network peak rate of the object to be detected.

In a second aspect, an embodiment of the present application provides a prediction apparatus, which may include:

the system comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring flow data of an object to be detected, and the object to be detected is a network element to be detected or a link to be detected;

the system comprises a grouping unit, a service type determining unit and a service processing unit, wherein the grouping unit is used for carrying out flow grouping on collected flow data according to a user label and the service type, the user label is used for indicating a network package signed by a user and/or an identity of the user, and the service type comprises an internet service, an internet dedicated line service and a video service;

the calculating unit is used for calculating the network peak rate of each flow packet; and adding the network peak rates of the flow packets to obtain the network peak rate of the object to be detected.

In a possible implementation manner, the traffic packet includes an internet service group, and the computing unit is configured to:

In one possible implementation, the computing unit is further configured to:

performing curve fitting according to the expected values of the at least two preset periods to obtain a preset period peak value average rate increasing function;

In one possible implementation, the traffic packet includes a private interconnection group, and the computing unit is configured to:

In one possible implementation, the traffic packet includes a unicast video group, and the computing unit is configured to:

In one possible implementation, the traffic packet includes a multicast video group, and the computing unit is configured to:

In one possible implementation manner, the traffic packet includes a multicast video group, and the computing unit is configured to:

In one possible implementation, the computing unit is further configured to:

In one possible implementation, the prediction apparatus further includes:

and the distribution unit is used for distributing network bandwidth resources to the object to be detected according to the network peak rate of the object to be detected.

In a third aspect, an embodiment of the present application provides a prediction apparatus, which may include:

the processor and the memory are connected through the bus, wherein the memory is used for storing a group of program codes, and the processor is used for calling the program codes stored in the memory and executing the steps in the first aspect of the embodiment of the present application or any implementation manner of the first aspect.

In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which stores instructions that, when executed on a computer, implement the method according to the first aspect or any implementation manner of the first aspect.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments or the background art of the present application, the drawings required to be used in the embodiments or the background art of the present application will be described below.

FIG. 1 is a schematic diagram of a system architecture applied to a method for predicting a network rate according to an embodiment of the present disclosure;

fig. 2 is a flowchart illustrating a method for predicting a network rate according to an embodiment of the present disclosure;

fig. 3 is a schematic flowchart of a method for predicting a network rate of an internet service group according to an embodiment of the present disclosure;

fig. 4 is a schematic flowchart of another method for predicting a network rate of an internet service group according to an embodiment of the present application;

fig. 5 is a flowchart illustrating a method for calculating a user online rate threshold when predicting a network rate according to an embodiment of the present application;

fig. 6 is a flowchart illustrating a method for predicting a network rate of a multicast video group according to an embodiment of the present disclosure;

FIG. 7 is a schematic diagram illustrating a prediction apparatus according to an embodiment of the present disclosure;

fig. 8 is a schematic composition diagram of another prediction apparatus according to an embodiment of the present application.

Detailed Description

Embodiments of the present application are described below with reference to the drawings in the embodiments of the present application.

The terms "including" and "having," and any variations thereof, in the description and claims of this application and the drawings described above, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus.

Fig. 1 is a schematic diagram of a system architecture applied to a method for predicting a network rate according to an embodiment of the present application. In the system architecture. May include, but is not limited to: a prediction apparatus 10, at least one object 20 to be measured, and at least one terminal 30.

The prediction device 10 is configured to collect traffic data of the object 20 to be measured, group the traffic according to the traffic data, and calculate a network peak rate of each group to obtain the network peak rate of the object 20 to be measured. The cloud computing system can be configured on a cloud computer or a server, and can also be configured on a local computer or a server, and the embodiment of the application is not limited at all.

Alternatively, the predictive device 10 may include, but is not limited to: a Central Processing Unit (CPU), memory, a network interface, and a user interface.

The central processing can be used to read computer instructions and process data in computer software. Is the core component in a computer system that reads, decodes, and executes instructions. The data in the memory can also be read, and the data is output after being correspondingly processed, for example, in the embodiment of the application, the central processing unit can acquire the flow data of the object to be detected through the network interface; and reading the calculation instruction from the memory to perform grouping and calculation on the traffic data.

The Memory, which may also be referred to as a storage medium, a storage device or a storage device, may include a high-speed Random Access Memory (RAM) and may also include a non-volatile Memory (non-volatile Memory), such as at least one disk Memory. The disk storage may be used to store an operating system and various application programs as well as non-program data. The cpu can read the program and run in the RAM (memory), which can temporarily store the program read and executed by the cpu and store the data obtained after the program is executed and required to be transmitted.

And the network interface is used for carrying out data transmission with the object to be detected and acquiring flow data of the object to be detected.

Optionally, the detection device may further include a user interface for connecting with an external device, such as a display screen and a keyboard, so as to facilitate a user to view information and operate.

And the at least one object to be tested 20 is connected with the prediction device 10 and can be used for service transmission, flow forwarding and the like. The Network device may be a Network device such as an Optical Line Terminal (OLT), an Optical Network Unit (ONU), a Broadband Remote Access Server (BRAS), a switch, a router, and the like. Or may be a link to be tested, and the embodiment of the present application is not limited at all.

And the at least one terminal 30 can be connected with the at least one object to be tested 20 to realize various internet access services.

The method for predicting the network rate of the present application is described in detail below with reference to fig. 2 to 6.

Referring to fig. 2, fig. 2 is a schematic flowchart illustrating a method for predicting a network rate according to an embodiment of the present disclosure; the method comprises the following steps:

s201, the prediction device collects flow data of the object to be measured.

The object to be tested is a network element to be tested or a link to be tested.

In order to accurately predict the peak rate of the network, basic data for prediction needs to be collected, accumulated and processed from the network, and the more sufficient the sampled data is, the more accurate the basic data is, the higher the prediction accuracy of the model is. Therefore, the collection points can be placed where the number of users is large.

For more convenient implementation, ports of some devices may be selected as anchor points for sampling statistics. Such as selecting several BRAS or OLT upstream ports as sampling anchors. When the object to be detected needs to be predicted, the related flow data can be selected for analysis and calculation.

And S202, carrying out flow grouping on the acquired flow data according to the user label and the service type.

The user label is used for indicating a network package signed by a user and/or an identity of the user, and the service types comprise an internet service, an internet dedicated line service and a video service.

Optionally, the term "and/or" in this embodiment is only one kind of association relationship describing an association object, and indicates that three relationships may exist, for example, a and/or B may indicate: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" herein generally indicates that the former and latter related objects are in an "or" relationship.

The difference between the service types and the users may cause inconsistent influences (rate, delay and packet loss rate) on the network, for example, the actual peak internet access rate of a student user in a 20M package is much higher than that of a 20M ordinary home user; the actual peak internet access rate of 200M normal users is also greater than the peak internet access rate of 50M normal users.

Alternatively, the traffic packets may include, but are not limited to: the network service group comprises an internet service group, an interconnection private line group, a unicast video group and a multicast video group. In addition, some commercial internet special lines of internet access services are processed in a similar way to the internet access service group.

The unicast video and the multicast video are used for indicating the self-operated video of the operator, and the Over The Top (OTT) video watched by the user can be classified into the internet service.

It should be noted that the same broadband user may belong to several traffic packets at the same time, for example, a 100M triple play (triple play) user, the internet traffic of the user belongs to a 100M common user internet group, the unicast video traffic watched by the set-top box of the user belongs to a unicast video group, and the multicast video traffic watched by the set-top box of the user belongs to a multicast video group.

The service type can be obtained through service identification, access destination address, identification of forwarding pipeline, session identification and the like, and the identity identification of the user can be obtained through big data analysis or provided by uploading of the user. Therefore, when grouping the traffic, the division can be performed by combining the acquired service type and/or the user identity.

With the rapid development of network functions and user requirements, in order to accurately predict the peak rate of the network, more, richer and more detailed traffic packets may be divided, for example, some users like to download movie collections very much, and a download service packet may be established for the users, which is not limited in this embodiment of the present application.

And S203, calculating the network peak rate of each flow packet.

According to the collected traffic data and the traffic groups, the daily peak value internet access rate can be counted according to the traffic data corresponding to the specific traffic groups, and then the expected value and the standard deviation of the traffic data are calculated by using a Poisson model. Thereby estimating the network peak rate for that particular traffic packet.

And S204, adding the network peak rates of the flow packets to obtain the network peak rate of the object to be detected.

Optionally, when the network peak rate of the object to be detected is obtained, the network bandwidth resource may be allocated to the object to be detected according to the network peak rate of the object to be detected. Here, the allocation may be performed by the prediction apparatus, or may be performed by reporting, by the prediction apparatus, to a device responsible for allocating network bandwidth resources, which is not limited in any way in the embodiments of the present application.

In the embodiment of the application, the traffic data is subjected to traffic grouping, then the network rate peak value of each group is calculated according to the grouping, and then the results are accumulated to obtain the network peak rate of the object to be measured. The variation of the network peak rate in a short time period is fully considered, and the difference of various users is fully considered in a grouping mode, so that the random variation of the network peak rate is more accurately described. The prediction result is more accurate. And the allocation and planning of network bandwidth resources are facilitated.

The following describes the calculation of the peak internet traffic for each packet in detail.

For The dedicated interconnection line group, when calculating The network peak rate of The dedicated interconnection line group, because The bandwidth of The dedicated interconnection line must be fully satisfied, The signed bandwidth of The dedicated interconnection line service can be obtained from an Operation Support System (OSS), and The signed bandwidth is used as The single-user network peak rate of The dedicated interconnection line traffic group;

and calculating to obtain the network peak rate of the interconnection private line group according to the number of users of the interconnection private line group and the single-user network peak rate.

For a unicast video group, in calculating a network peak rate for the unicast video group, comprising:

if the average code rate of the unicast video is the same, taking the product of the average code rate and a first preset coefficient as the peak rate of the single user network of the unicast video group; at this point the standard deviation is considered to be 0.

If the average bit rates of the unicast videos are different, for example, the unicast videos are distributed by using an adaptive bit rate method, the bit rates of programs that each user may watch are different, and at this time, a statistical method needs to be used to obtain the daily peak average rate statistical expectation and standard deviation of the single user of the unicast video group. Then, according to the second streaming data corresponding to the unicast video group, the daily peak average rate of the single user during the data collection period can be counted; calculating a second expected value and a second standard deviation of the daily peak average rate of the single user according to the data obtained by statistics; calculating the peak rate of the single user network of the unicast video group according to the second expected value and the second standard deviation;

For a multicast video group, when calculating a network peak rate of the multicast video group, the method comprises:

For a detailed calculation process of the internet service group, please refer to the embodiment of fig. 3-4, and for a detailed calculation process when the multicast replication point in the multicast video group is located on the object to be measured, please refer to the embodiment of fig. 6.

Referring to fig. 3, fig. 3 is a schematic flowchart illustrating a method for predicting a network rate of an internet service group according to an embodiment of the present disclosure; the method can comprise the following steps:

s301, according to the first traffic data corresponding to the internet service group, the daily peak average rate of a single user during the data acquisition period is counted.

For example, data collection for a sample anchor (BRAS or OLT port used for sampling) is from day 9.1 00: days 00 to 9.6 23: 45, collected every 15 minutes, 96 points a day, six days of rate data can be obtained.

If portstraxic (t) is the port rate at time t and onlineusernumber (t) is the number of online users of the group at time t, then the average rate of individual users in the group at the port is:

AverageUserSpeed(t)＝porttraffic(t)/onlineusernumber(t)，t＝00.00 to 24:00；

the average daily peak rate for a single user in the group is then:

PeakAverageUserSpeed(t)＝Max(porttraffic(t)/onlineusernumber(t))，t＝00.00 to 24:00。

s302, calculating a first expected value and a first standard deviation of the daily peak average rate of the single user according to the data obtained through statistics.

Assuming that the daily peak average rate for a single user is counted daily, we will have a data sequence PeakAverageUserSpeed (day): X1, X2, X3 … XN.

Then during data acquisition, the mathematical expectation of the peak average number rate of the single user in the internet service group is:

expeafpeakarateperiod sum (xn)/N, N1 to N, sum representing the summation operation.

The variance is: varpeakavaverageusesperated sum ((Xn-expafpeakerateprady) ² ) And (N-1), N is 1 to N.

The standard deviation is the square root of the variance and can be expressed as:

stddevpeakaverageusespeed ═ sqrt (varpeavagaverageusesped), sqrt denotes the square root operation.

And S303, calculating the peak rate of the single-user network of the internet service group according to the first expected value and the first standard deviation.

Alternatively, the first expected value plus a preset multiple of the standard deviation may be selected as the single-user network peak rate. For example: the peak rate exceeding "expected value +3 × standard deviation" can be considered as "small probability event", and the occurrence probability only meets the standard of the network packet loss rate, and the value can be used as a prediction of the future single-user network peak rate. The preset multiple in the embodiment of the present application may be selected according to actual needs, except for 3, and the embodiment of the present application is not limited at all.

S304, calculating to obtain the network peak rate of the internet service group according to the number of the users of the internet service group and the single-user network peak rate.

Since the network and user demands are constantly developing, the peak internet access rate of the user may also be increased over time, and thus, the variation of the longer period is fully considered when predicting the peak network rate.

Referring to fig. 4, fig. 4 is a schematic flowchart illustrating a method for predicting a network rate of an internet service group according to an embodiment of the present application; the method can comprise the following steps:

s401, calculating the peak average rate of the preset period of a single user according to the first flow data and the preset period, and obtaining expected values of at least two preset periods.

The preset period is greater than 1 day.

S402, performing curve fitting according to the expected values of the at least two preset periods to obtain a preset period peak value average rate increase function.

And S403, calculating the peak rate of the single-user network of the internet service group according to the first expected value, the first standard deviation and the peak average rate increase function of the preset period.

For example, the monthly expectation of the average peak rate of a single user in the internet service group can be counted by the collected traffic data to research the monthly growth function of the value.

Every month we can calculate the average peak speed monthly expectation of single users in the internet service group of the month:

expafpeaktamonthly (1 month 2018) ═ sum (xn)/N, N ═ 1 to 30;

expafpeaktamontonthy (2.2018) sum (xn)/N, N1 to 30;

…

expafpeaktamontonthy (12 months 2018) sum (xn)/N, N1 to 30;

..

we can perform curve fitting based on the above statistics.

This increasing curve may be exponential, or logarithmic, or linear, and a high confidence curve may be selected for fitting. For example: when the monthly expectation values for a plurality of months are distributed around a straight line, a linear fit can be used to predict the future curve, for example, the average monthly peak rate growth function obtained by the fit is: if Y is 1.08X + B, the average daily peak rate of the internet service group can be increased by 1.8% per month. The predicted value may be multiplied by (1+ 1.8%) when calculating the peak rate of the single-user network for the internet service group.

Optionally, since the number of users in the packet may not be online at the same time, the prediction of the network peak rate by using the number of users in the packet is usually large, and at this time, a user online rate threshold may be introduced to estimate the maximum number of users that may be online actually, so as to obtain a more accurate prediction effect.

The online rate of the internet service group means that the state of the PPPOE link for the user to access the internet is 'online', the online rate of the unicast video group means whether the user set top box is watching the unicast video, and the online rate of the multicast video group means whether the user set top box is watching the multicast video.

Referring to fig. 5, fig. 5 is a flowchart illustrating a method for calculating a user online rate threshold when predicting a network rate according to an embodiment of the present application; in the embodiment shown in fig. 5, the method may specifically include the following steps:

s501, according to the flow data corresponding to the flow grouping, the daily peak value online rate of the number of users in the flow grouping during data collection is counted.

The daily online rate curve of a certain group is also a curve which periodically changes according to days, the peak online rate generally appears between 8 and 10 points in the evening, and the peak online rate of a certain day in the group x is as follows:

Peakonlinerate.groupx＝MAX(onlinerate.groupx(t)),t＝00:00 to 24:00；

and S502, calculating expected values and standard deviations of daily peak online rates according to the data obtained through statistics.

The peak online rate can be regarded as a random variable, the statistical distribution of the peak online rate also conforms to the poisson distribution, and after the peak online rate is counted according to the day, a probability distribution curve can be obtained.

And S503, calculating a user online rate threshold according to the expected value and the standard deviation of the daily peak online rate.

For example, then the "user online rate threshold" is:

PeakOnlineratethreshold is the expected value of daily peak online rate +3 standard deviation of daily peak online rate.

"user online rate threshold", the statistical meaning is that a peak online rate exceeding this threshold is a very small probability event. This value can be used as the future maximum online rate for the packet in the model.

S504, the product of the online rate threshold of the user and the number of the users in the flow grouping is used as the number of the users when the peak rate of the flow grouping network is calculated.

For example, if the number of users in a currently processed traffic packet is 100 and the user online rate threshold is 65%, the network peak rate of the packet is predicted by multiplying (100 × 65%) by the network peak rate of a single user of the packet as the number of users to be actually predicted.

Referring to fig. 6, fig. 6 is a flowchart illustrating a method for predicting a network rate of a multicast video group according to an embodiment of the present disclosure; in the embodiment of the present application, the multicast replication point for replicating the multicast video playing is located on the object to be tested, it is assumed that each user independently determines which channel should be watched, and when the channel is randomly changed, the chances of watching each channel are the same. And setting u as the number of users borne on the object to be tested and c as the number of channels c required to be provided by the video service. The calculation method may specifically include the steps of:

s601, determining a third expected value and a third standard deviation of the online channels of the c channels according to the number c of the channels provided by the service requirement and the number u of the users borne by the object to be detected.

The function of each channel online is:

its probability distribution function is:

and S602, determining the online number of the peak frequency channel according to the third expected value and the third standard deviation.

The mathematical expectation of the number of peak channels on-line is:

the variance is: sigma ² (1)＝(1-E(1)) ² E(1)+(0-E(1)) ² (1-E(1))＝E(1)-E ² (1)

The mathematical expectation of the third expectation, c channels, is therefore:

the variance is:

σ ² (c)＝cσ ² (1)＝c[E(1)-E ² (1)]

the third standard deviation is σ.

The online distribution function of the c channels follows a normal distribution. Therefore, it is known that, on average, e (c) channels are online, and at most, e (c) +3 σ channels are online (the probability of exceeding this value is extremely small, and it can be considered that this does not occur). It is known that if the number of users is not much larger than the number of channels, then a certain percentage of the channels are not watched by anyone. That is, the multicast group that the Digital Subscriber Line Access Multiplexer (DSLAM) needs to support does not need to be estimated for all channels. In particular, if the number of users and channels is comparable, the number of channels on-line will be about 65% of the total number of channels.

It should be noted that the above derivation is not dependent on a specific multicast processing model, and therefore, this conclusion can be considered to have a certain general meaning.

And S603, calculating to obtain the network peak rate of the multicast video group according to the peak channel online number and the average code rate of the multicast video.

And finally, calculating the network peak value rate of the multicast video group according to the average code rate and the peak value channel online number of the multicast video.

The following describes a process for predicting a network peak rate according to an embodiment of the present application.

First, taking a 100M general user internet service Group3 as an example, it is assumed that statistics have already acquired the following data:

average rate statistical expectation value expafpeaker for single user of Group3 Group3

Group3 statistical standard deviation sdevofpeaktraependegy of average rate of single user peak value Group3 ═ B

The growth multiple in the single user monthly peak average rate growth function of Group3 is:

monthlyincreasetimesPeakrateperday.group3＝C

group3 with a statistical expectation value expofopeakeonlineperdata for the single-user peak online rate

Group3 standard deviation of single user peak online rate STDdeofpeakOnlinerelatedday, Group3 ═ E

The "user presence rate threshold" for the single-user peak presence rate of Group3 is:

peakOnlineratethreshold.group3＝F

if the network peak rate needs to be predicted for the next g months, it is known,

predicted time point (month) BWfcstmonth ═ g

Maximum allowable packet loss rate maxPLLRlimit of uplink port i

The 100M common maximum user number MaxuserNumber.Group3 ═ j on the object to be measured

Then the need to predict:

how much bandwidth will be needed by the network element Group3 in the future g months?

Still further, consider all traffic packets on the object to be tested (assuming there are 6 internet service groups), how large bandwidth is needed for the total uplink port?

The specific calculation process is as follows:

the peak rate of the single-user network in Group3 conforms to Gaussian distribution N (A, B) ² ) Then the net peak rate of j users also follows the gaussian distribution N (j x F a, j) ² *F ² *B ² ) Expressed as a mathematical formula:

X～N(j*F*A，j ² *F ² *B ² ) And X is the daily peak rate of the j 100M ordinary users for surfing the internet.

It should be noted that, here, simplified processing is performed, the presence rate is not actually a constant, the presence rate and the J user network peak rate are two independent events, and the product of the presence rate and the J user network peak rate should satisfy cauchy distribution, but here, for simple processing, the presence rate is directly replaced by a constant of "user presence rate threshold".

Another known condition is that the probability that the network peak rate is less than the upstream port capacity is equal to 1-i, then the probability distribution of X must satisfy: p (X < ═ Capacity) > (1-i);

the distribution of X is converted to a standard gaussian distribution:

Y＝(X-J*F*A)/(j*F*B)～N(0,1)

the probability P (X < ═ Capacity) can be replaced by P ((X-J × F × a)/(J × F × B) < ═ Capacity-J × F × a)/(J × F × B)) >, then 1-i

The problem is converted into:

STDNormdist ((Capacity-J F A)/(J F B)) > < 1-i (STDNormdist is a cumulative probability function of a standard Gaussian distribution and can be obtained by table lookup)

(Capacity-J F A)/(J F B) ═ STDnormminv (1-i,0,1) (STDNormdist function: the known probabilities in the Gaussian distribution are summed up and the corresponding cut-off values are determined)

The required capacity of the upstream port is:

Capacity＝STDnormminv(1-i,0,1)*j*F*B+J*F*A

and correcting the A and the B by using the increase multiple in the monthly peak average rate increase function to obtain the network peak rate of Group 3:

Capacity＝STDnormminv(1-i,0,1)*j*F*B*C ^g +J*F*A*C ^g

note that: the standard deviation part after the future g months is also simplified, and the standard deviation is directly expanded by C ^g 。

Assuming that there are 6 internet service groups, the peak rate of each internet group satisfies the gaussian distribution

X1～N(j1*F1*A1*C ^g ，(j1*F1*B1*Cn ^g ) ² )

X2～N(j2*F2*A2*C ^g ，(j2*F2*B2*Cn ^g ) ² )

…

X6～N(j2*F2*A2*C ^g ，(j2*F2*B2*Cn C ^g ) ² )

Then the superposition of the network peak rates of all the internet service groups also conforms to the gaussian distribution:

Sum(Xn)～N(sum(jn*Fn*An*Cn ^g ),sum((jn*Fn*Bn*Cn ^g ) ² ) N ═ 1 to 6).

The expected value is the sum of the expected values for each packet and the variance is the sum of the variances for all packets.

It should be noted that the standard deviation is not the sum of all standard deviations, but the standard deviation is obtained by summing the variances and then obtaining the square root.

If Jn is the number of users in the nth group, Fn is the 'user online rate threshold' in the nth group, An is the expected value of the peak rate of the network of the nth group of single users, Bn is the standard deviation of the peak rate of the network of the nth group of single users, Cn is the multiple of the monthly peak average rate growth function, and g is the number of months.

The superposed maximum uplink capacity requirement is as follows:

Fcstcapacity＝STDnormminv(1-i,0,1)*sqrt(sum((jn*Fn*Bn*Cn ^g ) ² ))+sum(jn*Fn*An*Cn ^g )。

if there are also network peak rates of other packets that need to be superimposed, such as superimposed groups of private interconnect lines.

Since the bandwidth of the interconnecting private line group is fixedly ensured, the bandwidth can be regarded as a linear function only related to the number of users

For example, if j7 is the number of users in the "1G interconnection private line group (group 7)", the peak rate of the uplink port after overlapping the interconnection private line groups conforms to the following gaussian distribution

Sum(Xn)+X7～N(sum(jn*Fn*An*Cn ^g )+j7*10 ⁹ ,sum((jn*Fn*Bn*Cn ^g ) ² ))

The maximum uplink bandwidth requirement that is superimposed:

Fcstcapacity＝STDnormminv(1-i,0,1)*sqrt(sum((jn*Fn*Bn*Cn ^g ) ² ))+sum(jn*Fn*An*Cn ^g )+j7*10 ⁹ 。

when the network peak rate of the unicast video Group8 needs to be superimposed, for simplicity, assuming that the code rates of the unicast videos are all the same, the unicast video rate of the uplink port can be regarded as a linear function related to the number of users

Sum(Xn)+X7+X8～N(sum(jn*Fn*An*Cn ^g )+j7*10 ⁹ +j8*f8*M8*L8,sum((jn*Fn*Bn*Cn ^g )^2))

J8 is the number of IPTV users, F8 is the "user online rate threshold" of the unicast video group, L8 is the unicast video average bitrate, and M8 is the peak bitrate multiple.

The maximum uplink bandwidth requirement that is superimposed:

Fcstcapacity＝STDnormminv(1-i,0,1)*sqrt(sum((jn*Fn*Bn*Cn ^g ) ² ))+sum(jn*Fn*An*Cn ^g )+j7*10 ⁹ +j8*f8*M8*L8

when the network peak rate of the multicast video Group9 needs to be superimposed, if the multicast replication point is not on the object to be measured, the multicast bandwidth is not converged, and the calculation formula of the multicast bandwidth is as follows:

Group9.capacity＝L9*M9*J9*F9

assuming that there is only one set-top box per IPTV, L9 is the average multicast program bitrate, M9 is the multiple of the multicast peak bitrate (mostly CBR, multiple is 1), J9 is the number of IPTV users, and F9 is the "user online rate threshold" for the multicast video group.

The maximum uplink port bandwidth requirement is

Fcstcapacity＝STDnormminv(1-i,0,1)*sqrt(sum((jn*Fn*Bn*Cn ^g ) ² ))+sum(jn*Fn*An*Cn ^g )+j7*10 ⁹ +j8*f8*M8*L8+L9*M9*J9*F9

If the multicast replication point is on the object to be measured, the convergence rate of multicast replication needs to be calculated. The model is as follows

The maximum multicast rate of the uplink port of the object to be tested is as follows: group9.capacity ═ L9 × M9: (e) (ch) +3 × stddev (ch)

The number of channels of a user in the CH channels is expected: e (CH) ═ CH (1- (1/e) ^(J9*F9/CH) 。

Among the CH channels, there is the standard deviation of the number of channels viewed by the user:

STDdev(CH)＝SQRT(CH*((1-(1/e) ^(J9*F9/CH) )-(1-(1/e) ^(J9*F9/CH) ) ² ))

j9 is the number of IPTV users on the object to be tested, F9 is the "user online rate threshold" of the multicast video group, CH is the number of multicast channels, e is the natural logarithm, L9 is the average code rate of the multicast video, and M9 is the multiple of the peak code rate.

The maximum uplink bandwidth obtained by the superposition is:

Fcstcapacity＝STDnormminv(1-i,0,1)*sqrt(sum((jn*Fn*Bn*Cn ^g ) ² ))+sum(jn*Fn*An*Cn ^g )+j7*10 ⁹ +j8*f8*M8*L8+L9*M9*J9*F9+group9.capacity

it should be noted that, the simplified processing is performed here, the multicast rate and the other group rates are not in an additive relationship, and they may also be statistically multiplexed, and the simplified processing is that the two rates are completely independent.

Please refer to fig. 7, which is a schematic diagram illustrating a prediction apparatus according to an embodiment of the present disclosure; the method comprises the following steps:

the system comprises an acquisition unit 100, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring flow data of an object to be detected, and the object to be detected is a network element to be detected or a link to be detected;

a grouping unit 200, configured to perform traffic grouping on the acquired traffic data according to a user tag and a service type, where the user tag is used to indicate a network package signed by a user and/or an identity of the user, and the service type includes an internet service, an internet dedicated line service, and a video service;

a calculating unit 300, configured to calculate a network peak rate of each traffic packet; and adding the network peak rates of the flow packets to obtain the network peak rate of the object to be detected.

Optionally, the traffic packet includes an internet service group, and the computing unit 300 is configured to:

Optionally, the computing unit 300 is further configured to:

Optionally, the traffic packet includes an interconnection private line group, and the computing unit 300 is configured to:

Optionally, the traffic packet includes a unicast video group, and the computing unit 300 is configured to:

and calculating the network peak value rate of the unicast video group according to the number of the users of the unicast video group and the network peak value rate of the single user.

Optionally, the traffic packet includes a multicast video group, and the computing unit 300 is configured to:

Optionally, the computing unit 300 is further configured to:

and taking the product of the online rate threshold of the user and the number of the users in the flow grouping as the number of the users when the peak rate of the flow grouping network is calculated.

Optionally, the prediction apparatus further comprises:

an allocating unit (not shown) configured to allocate network bandwidth resources to the object to be tested according to the network peak rate of the object to be tested.

For the concepts, explanations, details and other steps related to the technical solutions provided in the embodiments of the present application related to the prediction apparatus, reference is made to the description of these contents in the foregoing method embodiments, which are not repeated herein.

Please refer to fig. 8, which is a schematic composition diagram of another prediction apparatus provided in an embodiment of the present application; as shown in fig. 8, the controller may include a processor 110, a memory 120, and a bus 130. The processor 110 and the memory 120 are connected by a bus 130, the memory 120 is used for storing instructions, and the processor 110 is used for executing the instructions stored by the memory 120 to realize the steps in the method corresponding to fig. 2-6.

Further, the controller may also include an input port 140 and an output port 150. Wherein the processor 110, the memory 120, the input port 140, and the output port 150 may be connected by a bus 130.

The processor 110 is configured to execute the instructions stored in the memory 120 to control the input port 140 to collect traffic data, and optionally, control the output port 150 to issue allocation information of network bandwidth resources, so as to complete the steps executed by the prediction apparatus in the above method. Wherein input port 140 and output port 150 may be the same or different physical entities. When they are the same physical entity, they may be collectively referred to as an input-output port. The memory 120 may be integrated in the processor 110 or may be provided separately from the processor 110.

As an implementation manner, the functions of the input port 140 and the output port 150 may be implemented by a transceiver circuit or a dedicated chip for transceiving. The processor 110 may be considered to be implemented by a dedicated processing chip, processing circuit, processor, or a general-purpose chip.

As another implementation manner, a manner of using a general-purpose computer to implement the prediction apparatus provided in the embodiment of the present application may be considered. Program code that implements the functionality of processor 110, input ports 140 and output ports 150 is stored in memory, and a general purpose processor implements the functionality of processor 110, input ports 140 and output ports 150 by executing the code in memory.

As another implementation manner, a single board may be considered to implement the prediction apparatus provided in the embodiment of the present application. That is, configuring a coupled main control board and an interface board, the processor 110 and the memory 120 may be configured on the main control board, the input port 140 and the output port 150 are configured on the interface board, the main control board executes a program to generate a test packet and complete bandwidth detection, and the sending and receiving of the test packet is completed through the interface board. Alternatively, the memory 120 may be configured on the interface board.

For the concepts, explanations, details and other steps related to the technical solutions provided in the embodiments of the present application related to the prediction apparatus, reference is made to the descriptions of the foregoing methods or other embodiments, which are not repeated herein.

Those skilled in the art will appreciate that fig. 8 shows only one memory and processor for ease of illustration. In an actual controller, there may be multiple processors and memories. The memory may also be referred to as a storage medium or a storage device, and the like, which is not limited in this application. In the embodiments of the present application, the processor may be a Central Processing Unit (CPU), and the processor may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The memory may include both read-only memory and random access memory, and provides instructions and data to the processor. The portion of memory may also include non-volatile random access memory. The bus may include a power bus, a control bus, a status signal bus, and the like, in addition to the data bus. But for clarity of illustration the various buses are labeled as buses in the figures.

According to the method and the prediction apparatus provided in the embodiments of the present application, the embodiments of the present application further provide a computer system, which includes a CPU, a controller and a storage medium, and the relationship and instruction flow of the three may refer to the description and illustration in the embodiments of fig. 1 to 6, which are not described herein again.

It should be understood that, in the various embodiments of the present application, the sequence numbers of the above-mentioned processes do not mean the execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

Those of ordinary skill in the art will appreciate that the various illustrative logical blocks and steps (step) described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the application to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.

The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method of predicting a network rate, comprising:

carrying out flow grouping on the acquired flow data according to a user label and a service type, wherein the user label is used for indicating a network package signed by a user and an identity of the user, the identity is obtained through big data analysis or is uploaded and provided by the user, and the service type comprises an internet service, an internet dedicated line service and a video service;

calculating the user online rate threshold of each group, and calculating the network peak rate of each flow group according to the user online rate threshold of each group;

2. The method of claim 1, wherein the traffic packets comprise a group of internet traffic, and wherein calculating the network peak rate of the group of internet traffic comprises:

according to the first traffic data corresponding to the internet service group, counting the daily peak average rate of a single user during data acquisition;

calculating the peak rate of the single-user network of the internet service group according to the first expected value and the first standard deviation;

3. The method of claim 2, further comprising:

4. The method of claim 1, wherein the traffic packets comprise a private interconnect wire group, and wherein calculating the peak internet rate for the private interconnect wire group comprises:

acquiring a signed bandwidth of the private interconnection line service from an operation support system, and taking the signed bandwidth as a single-user network peak rate of the private interconnection line group;

5. The method of claim 1, wherein the traffic packet comprises a unicast video group, and wherein the calculating of the network peak rate of the unicast video group comprises:

6. The method of claim 1, wherein the traffic packet comprises a multicast video group, and wherein calculating the network peak rate of the multicast video group comprises:

if the multicast replication point for replicating the multicast video playing is not located on the object to be measured, taking the product of the average code rate of the multicast video and a second preset coefficient as the peak rate of the single-user network of the multicast video group;

7. The method of claim 1, wherein the traffic packet comprises a multicast video group, and wherein calculating the network peak rate of the multicast video group comprises:

if the multicast replication point for replicating the multicast video playing is located on the object to be tested, determining a third expected value and a third standard deviation of the online channels of the c channels according to the number c of the channels required to be provided by the service and the number u of the users borne by the object to be tested;

8. The method of claim 2, 3, 5 or 6, wherein the calculating the user presence rate threshold for each group comprises:

calculating the expected value and the standard deviation of the daily peak online rate according to the data obtained by statistics;

9. The method of claim 1, further comprising:

10. A prediction apparatus, comprising:

the system comprises a grouping unit, a service type obtaining unit and a service processing unit, wherein the grouping unit is used for carrying out flow grouping on collected flow data according to a user label and the service type, the user label is used for indicating a network package signed by a user and an identity of the user, the identity is obtained through big data analysis or is uploaded and provided by the user, and the service type comprises an internet access service, an internet dedicated line service and a video service;

the calculating unit is used for calculating the user online rate threshold of each group and calculating the network peak rate of each flow group according to the user online rate threshold of each group; and adding the network peak rates of the flow packets to obtain the network peak rate of the object to be detected.

11. The prediction apparatus according to claim 10, wherein the traffic packet comprises an internet service group, and the calculation unit is configured to:

12. The prediction apparatus according to claim 11, wherein the computing unit is further configured to:

13. The prediction apparatus of claim 10, wherein the traffic packet comprises a private interconnect wire group, and wherein the computing unit is configured to:

14. The prediction apparatus according to claim 10, wherein the traffic packet comprises a unicast video group, and the computing unit is configured to:

15. The prediction apparatus according to claim 10, wherein the traffic packet comprises a multicast video group, and the calculation unit is configured to:

16. The prediction apparatus according to claim 10, wherein the traffic packet comprises a multicast video group, and the calculation unit is configured to:

17. The prediction apparatus according to claim 11, 12, 14 or 15, wherein the computing unit is further configured to:

according to flow data corresponding to the flow grouping, counting the daily peak value online rate of the number of users in the flow grouping every day during data acquisition;

18. The prediction apparatus according to claim 10, wherein the prediction apparatus further comprises:

19. A prediction apparatus, comprising:

a processor, a memory and a bus, the processor and the memory being connected by the bus, wherein the memory is configured to store a set of program codes, and the processor is configured to call the program codes stored in the memory to perform the method according to any one of claims 1-9.

20. A computer-readable storage medium, comprising:

the computer-readable storage medium has stored therein instructions which, when run on a computer, implement the method of any one of claims 1-9.