CN111756646B

CN111756646B - Network transmission control method, device, computer equipment and storage medium

Info

Publication number: CN111756646B
Application number: CN202010652960.1A
Authority: CN
Inventors: 刘岩; 杨文正
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2020-07-08
Filing date: 2020-07-08
Publication date: 2023-09-29
Anticipated expiration: 2040-07-08
Also published as: CN111756646A

Abstract

The application relates to a network transmission control method, a device, a computer device and a storage medium, which relate to the technical field of network transmission, and the method comprises the following steps: acquiring a historical data set, wherein the historical data set comprises network transmission data at each historical time point; each historical time point corresponds to a periodic characteristic, and the periodic characteristic is used for indicating the time domain position of the corresponding time point in the belonged time period; acquiring periodic characteristics of a target time point; according to the periodic characteristics of the target time points, the periodic characteristics of each historical time point and the historical data set, obtaining the prediction strategy characteristics corresponding to the target time points; and in a time interval corresponding to the target time point, performing network transmission control according to the prediction strategy characteristics. The scheme fully considers the periodic change of the network transmission condition, thereby improving the accuracy of the prediction of the network transmission control strategy and the accuracy of the network transmission control.

Description

Network transmission control method, device, computer equipment and storage medium

Technical Field

The embodiment of the application relates to the technical field of network transmission, in particular to a network transmission control method, a network transmission control device, computer equipment and a storage medium.

Background

In the technical field of network communication, a server and a client communicate based on a network transmission protocol, for example, the transmission control protocol is a transmission layer communication protocol, and in order to prevent the congestion phenomenon of a network, a series of congestion control strategies are proposed by a developer, and each strategy has a set of parameters for realizing congestion control.

In practical applications, the data transmission quality of different networks varies greatly, for example, optical fiber, wireless local area network, wireless cellular network, etc., and even the same network has different network quality at different times. In the related art, a time window is manually determined through manual experience, a plurality of (several or tens of) candidate parameter sets are empirically determined, and policy parameters with optimal transmission performance are found through a random search mode.

However, the policy optimization process has less consideration on the change of the network transmission condition along with time, so that the accuracy of acquiring the policy features is lower, and the accuracy of controlling the network transmission is lower.

Disclosure of Invention

The embodiment of the application provides a network transmission control method, a device, computer equipment and a storage medium, which can improve the accuracy of acquiring strategy characteristics and further improve the accuracy of network transmission control, and the technical scheme is as follows:

In one aspect, a network transmission control method is provided, and the method includes:

acquiring a historical data set, wherein the historical data set comprises network transmission data at each historical time point; each historical time point corresponds to a periodic characteristic, and the periodic characteristic is used for indicating the time domain position of the corresponding time point in the belonged time period; the network transmission data is used for indicating a strategy characteristic and network transmission performance at a corresponding time point, and the strategy characteristic is used for indicating at least one of a network transmission control strategy and a strategy parameter of the network transmission control strategy;

acquiring periodic characteristics of a target time point;

according to the periodic characteristics of the target time points, the periodic characteristics of each historical time point and the historical data set, obtaining the prediction strategy characteristics corresponding to the target time points;

and in a time interval corresponding to the target time point, performing network transmission control according to the prediction strategy characteristic.

In another aspect, there is provided a network transmission control apparatus, the apparatus including:

the data set acquisition module is used for acquiring a historical data set, wherein the historical data set comprises network transmission data at each historical time point; each historical time point corresponds to a periodic characteristic, and the periodic characteristic is used for indicating the time domain position of the corresponding time point in the belonged time period; the network transmission data is used for indicating a strategy characteristic and network transmission performance at a corresponding time point, and the strategy characteristic is used for indicating at least one of a network transmission control strategy and a strategy parameter of the network transmission control strategy;

The first feature acquisition module is used for acquiring periodic features of the target time point;

the strategy characteristic acquisition module is used for acquiring the prediction strategy characteristic corresponding to the target time point according to the periodic characteristic of the target time point, the periodic characteristic of each historical time point and the historical data set;

and the transmission control module is used for carrying out network transmission control according to the prediction strategy characteristics in a time interval corresponding to the target time point.

In one possible implementation manner, the policy feature obtaining module includes:

a fitting function construction sub-module, configured to construct a fitting function based on the historical data set, where the fitting function is used to indicate a relationship between transmission state information of each historical time point and network transmission performance corresponding to each historical time point; the transmission state information is used for indicating the periodic characteristics of the corresponding time points and the strategy characteristics of the corresponding time points;

and the strategy feature extraction sub-module is used for extracting the prediction strategy features from the fitting function based on the periodic features of the target time points.

In one possible implementation, the policy feature extraction submodule is configured to, in use,

Extracting a reference evaluation point from the fitting function based on the periodic characteristic of the target time point, wherein the reference evaluation point comprises a strategy characteristic which enables network transmission performance to be maximized at the target time point;

and extracting the strategy features contained in the reference evaluation points as the prediction strategy features.

In one possible implementation, the fitting function constructs a sub-module for,

extracting each reference time point from each historical time point based on the periodic features of the target time point and the periodic features of each historical time point;

and constructing the fitting function based on the periodic characteristics of the reference time points and network transmission data of the reference time points.

In one possible implementation, the apparatus further includes:

a second feature acquisition module, configured to, before the data set acquisition module acquires a historical data set, obtain a periodic feature of a first time point by taking a model of the first time point according to a period length of the time period in response to the respective historical time points being expressed in terms of absolute time differences from an initial time point; the first time point is any one of the respective history time points.

In one possible implementation, the apparatus further includes:

a third feature acquisition module, configured to, before the data set acquisition module acquires the historical data sets, respond to the respective historical time points to be represented in the form of clock times, and the time period is a period with a duration of 24 hours, and acquire a clock time at a first time point as a periodic feature of the first time point; the first time point is any one of the respective history time points.

In one possible implementation, the apparatus further includes:

a merging module, configured to merge network transmission performance at the first time point and network transmission performance at the second time point in response to the transmission state information at the first time point being the same as the transmission state information at the second time point in the respective historical time points;

the transmission state information is used for indicating the periodic characteristics of the corresponding time points and the strategy characteristics of the corresponding time points.

In one possible implementation, the combining module is configured to, in response to a request from the host device,

modifying the network transmission performance of the time point with the front time point to the network transmission performance of the time point with the rear time point in the first time point and the second time point;

Or alternatively, the process may be performed,

and adjusting the network transmission performance of the time point with the later time point based on the network transmission performance of the time point with the earlier time point in the first time point and the second time point.

In one possible implementation, the apparatus further includes:

and the strategy characteristic generation module is used for responding to the fact that the historical data set is empty and randomly generating the prediction strategy characteristic.

In one possible implementation, the apparatus further includes:

the correction module is configured to correct, before the policy feature obtaining module obtains the predicted policy feature corresponding to the target time point according to the periodic feature of the target time point, the periodic feature of each historical time point, and the historical dataset, the network transmission performance of each historical time point according to the periodic feature of the target time point and the periodic feature of each historical time point.

In one possible implementation manner, the transmission control module is configured to perform at least one of the following network transmission control according to the policy feature in a time interval corresponding to the target time point: transmission control protocol TCP control, audio and video transmission performance control, content delivery network CDN scheduling control, and network bandwidth scheduling control.

In another aspect, a network transmission control system is provided, the system comprising: the system comprises a log monitoring server, a strategy parameter optimizing server and a transmission control server;

the log monitoring server is used for recording a historical data set, wherein the historical data set comprises network transmission data at each historical time point; each historical time point corresponds to a periodic characteristic, and the periodic characteristic is used for indicating the time domain position of the corresponding time point in the belonged time period; the network transmission data is used for indicating a strategy characteristic and network transmission performance at a corresponding time point, and the strategy characteristic is used for indicating at least one of a network transmission control strategy and a strategy parameter of the network transmission control strategy;

the strategy parameter optimization server is used for acquiring periodic characteristics of a target time point; according to the periodic characteristics of the target time points, the periodic characteristics of each historical time point and the historical data set, obtaining the prediction strategy characteristics corresponding to the target time points;

and the transmission control server is used for carrying out network transmission control according to the prediction strategy characteristics in a time interval corresponding to the target time point.

In another aspect, a computer device is provided, the computer device including a processor and a memory having stored therein at least one instruction, at least one program, a set of codes, or a set of instructions, the at least one instruction, the at least one program, the set of codes, or the set of instructions being loaded and executed by the processor to implement a network transmission control method as described above.

In another aspect, a computer readable storage medium having stored therein at least one instruction, at least one program, a set of codes, or a set of instructions loaded and executed by a processor to implement a network transmission control method as described above is provided.

In another aspect, a computer program product or computer program is provided, the computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the network transmission control method provided in the above-described various alternative implementations.

The technical scheme provided by the application can comprise the following beneficial effects:

by combining the relation of time domain positions of the historical time points and the target time points in the time periods of the historical time points, and the strategy parameters and the network transmission performance of the historical time points, the strategy characteristics of the target time points are predicted, so that the periodical change of the network transmission condition is fully considered in the process of predicting the network transmission control strategy based on the historical data, the accuracy of the network transmission control strategy prediction is improved, and the accuracy of the network transmission control is improved.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application as claimed.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description, serve to explain the principles of the application.

Fig. 1 is a schematic diagram illustrating a network transmission control system according to an exemplary embodiment of the present application;

fig. 2 is a flowchart illustrating a network transmission control method according to an exemplary embodiment of the present application;

fig. 3 is a flowchart illustrating a network transmission control method according to an exemplary embodiment of the present application;

Fig. 4 is a schematic diagram of a network transmission control system according to an exemplary embodiment of the present application;

fig. 5 is a timing chart illustrating a network transmission control method according to an exemplary embodiment of the present application;

fig. 6 is a block diagram showing a network transmission control apparatus according to an exemplary embodiment of the present application;

FIG. 7 is a block diagram of a computer device shown in accordance with an exemplary embodiment;

fig. 8 is a block diagram of a computer device, according to an example embodiment.

Detailed Description

Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the application. Rather, they are merely examples of apparatus and methods consistent with aspects of the application as detailed in the accompanying claims.

The embodiment of the application provides a network transmission control method, which can predict the strategy characteristic at the target time point by combining the relation of the time domain positions of the historical time point and the target time point in the time period of each time point, and the strategy parameters and the network transmission performance of each historical time point, thereby fully considering the periodical change of the network transmission condition in the process of predicting the network transmission control strategy based on the historical data, further improving the accuracy of the network transmission control strategy prediction and the accuracy of the network transmission control. For ease of understanding, the terms involved in the present application are explained below.

1) CDN (Content Delivery Network )

CDN refers to a content delivery network, also known as a content delivery network, for improving the quality of service of the internet.

The CDN network is composed of content caching equipment, a content exchanger, a content router, a CDN content management system and the like:

the content caching device is a CDN network node, is positioned at a user access point, is a content providing device for an end user, and can cache static Web content and streaming media content, so that the edge propagation and storage of the content are realized, and the user can conveniently access nearby.

The content exchanger is positioned in the user access centralized point, can balance the loads of a plurality of content caching devices at a single point, and can perform caching load balance and access control on the content.

The content router is responsible for scheduling the user's requests to the appropriate devices. The content router is usually realized through a load balancing system, dynamically balances load distribution of each content cache site, selects the best access site for the request of a user, and improves the usability of the website. The content router may route based on a variety of factors, including proximity of the site to the user, availability of content, network load, equipment conditions, and the like. The load balancing system is the core of the whole CDN. The accuracy and efficiency of load balancing directly determines the efficiency and performance of the overall CDN.

The CDN content management system is responsible for the management of the whole CDN and is an optional component, and is used for content management, such as content injection and delivery, content auditing, content service and the like.

The CDN distributes the cache servers to areas or networks with relatively concentrated user accesses by widely adopting various cache servers, when the user accesses a website, the user accesses are directed to the cache servers which work normally and are closest to the website by using a global load technology, and the cache servers directly respond to the user requests, so that the user access response speed and hit rate are improved.

CDNs are complementary approaches based on the TCP/IP architecture, and the data packets of CDNs are TCP/IP data packets.

2) TCP (Transmission Control Protocol )

TCP is a connection-oriented, reliable, byte stream based transport layer communication protocol.

TCP is connection-oriented, i.e., a client needs to establish a trusted connection with a server (or both communicating parties) before data transmission takes place. After the data transmission is finished, the connection is disconnected in an agreement mode, and the two communication parties release the resources.

TCP is reliable and defines a "time-out retransmission mechanism" for packets, i.e., each packet waits for a response after being sent out. If no response is received within the designated time, the sender performs a certain number of retransmissions to ensure reliable transmission of the data.

TCP is byte-stream based, and the application layer does not need to pay attention to the boundary of a data packet when transmitting data, and TCP automatically buffers, groups and merges data according to the network environment when transmitting data.

3) BOA (Bayesian Optimization Algorithm Bayesian optimization)

The main problems of the Bayesian optimization algorithm are:

X ^* ＝arg _x∈S maxf(x)

where S is a candidate set of x. The goal is to select one x from S such that the value of f (x) is either the smallest or the largest, and possibly the specific formulation of f (x) is not known, but if one is selected, the value of f (x) can be found experimentally or by observation. f (x) is a function of a black box.

Referring to fig. 1, a schematic diagram of a network transmission control system according to an exemplary embodiment of the present application is shown, and as shown in fig. 1, the network transmission control system 100 includes a server cluster 110 and a client 120.

The server may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDNs, basic cloud computing services such as big data and artificial intelligence platforms.

The server cluster 110 includes a transmission control server, a log monitoring server, and a policy parameter optimization server. The log monitoring server may include at least one of a client log monitoring server and a transmission control log monitoring server.

The log monitoring server comprises a log monitoring module and is mainly responsible for summarizing parameters, strategies and corresponding times corresponding to each transmission control connection and returning the summarized parameters, strategies and corresponding times to the strategy parameter optimization server as required.

And the strategy parameter optimization server performs strategy and parameter optimization according to the log reported by the log monitoring server, and returns an optimization result, namely, the optimized strategy and parameter to the corresponding transmission control server for use.

The transmission control server is mainly responsible for responding to the request of a normal client, and simultaneously, optimizing the strategy and parameters returned by the server according to the strategy parameters to carry out network transmission control. Meanwhile, the transmission control server also needs to periodically report the performance data of each transmission control connection to the log monitoring server in the last period of time, so that the log monitoring server gathers the performance data of each transmission control connection, wherein the performance data of each transmission control connection comprises bandwidth, connection time, downloaded file size and the like.

The client may be, but not limited to, a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart watch, etc. The terminal and the server may be directly or indirectly connected through wired or wireless communication, and the present application is not limited herein.

Referring to fig. 2, which is a flowchart illustrating a network transmission control method according to an exemplary embodiment of the present application, as shown in fig. 2, the method is performed by a network transmission control system, which may be the network transmission control system shown in fig. 1, and as shown in fig. 2, the network transmission control method includes the steps of:

step 210, acquiring a historical data set, wherein the historical data set comprises network transmission data at each historical time point; each historical time point corresponds to a periodic characteristic, and the periodic characteristic is used for indicating the time domain position of the corresponding time point in the belonged time period; the network transmission data is used to indicate a policy characteristic at a corresponding point in time and a network transmission performance, the policy characteristic being used to indicate at least one of a network transmission control policy and a policy parameter of the network transmission control policy.

In one possible implementation, the network transmission control policy is a network transmission control algorithm, and the policy parameter of the corresponding network transmission control policy is an algorithm parameter of the network transmission control algorithm.

In one possible implementation, after each network transmission connection completes data transmission, the corresponding protocol stack records the relevant data of the current connection and stores the relevant data as a network transmission log. For example, in text form in a log monitoring server or client or network transmission server as shown in fig. 1; and screening at least one group of network transmission data for indicating the network transmission condition at the corresponding time point according to a certain time interval based on the network transmission log, and forming a historical data set by the at least one group of network transmission data.

In at least one group of network transmission data, the network transmission data at each historical time point is used for representing the network transmission condition in a time area corresponding to each historical time point, taking a time interval as an example, the network transmission data corresponding to 6:00 yesterday is used for representing the network transmission condition in the time interval of 6:00-7:00 yesterday, and 6:00 is used for representing the starting time of the time interval and also representing the time when the strategy features corresponding to the time interval start to be used.

Step 220, obtain periodic features of the target time point.

In one possible implementation, the target time point refers to the current time point, or is a certain time point after the current time point.

That is, the network transmission control system may acquire the prediction policy feature in real time, for example, the current time is 20:00, and the network transmission control system acquires the prediction policy feature required in the time interval to which 20:00 belongs by acquiring the periodic feature of 20:00 in real time, so as to guide the network transmission control in the time interval to which 20:00 belongs.

Or, the periodic feature of a certain time point may be obtained in advance, so as to obtain the prediction policy feature required in the time interval to which the time point belongs in advance, so as to guide the network transmission control in the time interval to which the time point belongs, for example, the current time point is 20:00, and the 21:00 transmission control may be obtained according to a preset rule, so that the policy feature required in the time interval to which the 21:00 belongs is obtained in advance.

The embodiment of the application provides a network transmission control method by taking the periodic characteristic of a real-time acquisition target time point as an example.

In a possible implementation manner, the time period is preset according to practical situations, for example, the time period may be a week, a day or an hour, or a custom arbitrary time length. Taking this time period as one day (24 hours) as an example, the periodic characteristics corresponding to each historical time point appear to be a cycle from 0 to 23, i.e., 8 pm on the previous day and 8 pm today, are both shown to be at the 20 th time domain location, i.e., 20:00.

Step 230, obtaining the prediction strategy feature corresponding to the target time point according to the periodic feature of the target time point, the periodic feature of each historical time point and the historical data set.

In one possible implementation manner, the prediction policy feature is used for performing network transmission control, and for the same network transmission control, multiple transmission control policies may be corresponding, each transmission control policy may correspond to a different number and a different range of parameters, and the parameters corresponding to the same transmission control policy may be adjusted within a specified range.

In one possible implementation manner, according to the periodic feature of the target time point, the periodic feature of each historical time point, and the historical data set, the prediction policy feature corresponding to the target time point is obtained, which may be implemented as obtaining the transmission control policy corresponding to the target time point, or obtaining the parameter corresponding to the target time point, or simultaneously obtaining the transmission control policy corresponding to the target time point and the parameter corresponding to the transmission control policy.

And 240, performing network transmission control according to the prediction strategy characteristics in a time interval corresponding to the target time point.

In a possible implementation manner, the time interval corresponding to the target time point is a certain time period in the time period, for example, the time period is one day, the target time point is 8:00, the time interval corresponding to the target time point may be 8:00-9:00, and network control transmission is performed in the time interval according to the prediction policy feature.

In summary, according to the network transmission control method provided by the embodiment of the application, by combining the relation between the time domain positions of the historical time points and the target time points in the respective time periods, and the policy parameters and the network transmission performance of each historical time point, the policy characteristics at the target time points are predicted, so that the periodic variation of the network transmission condition is fully considered in the process of predicting the network transmission control policy based on the historical data, the accuracy of predicting the network transmission control policy is improved, and the accuracy of network transmission control is improved.

In a possible implementation manner, the predicted policy feature corresponding to the target time point is obtained based on a bayesian optimization algorithm, please refer to fig. 3, which shows a flowchart of a network transmission control method according to an exemplary embodiment of the present application, as shown in fig. 3, where the method is executed by a network transmission control system, and the network transmission control system may be the network transmission control system shown in fig. 1, as shown in fig. 3, and the network transmission control method includes the following steps:

Step 310, acquiring a historical data set, wherein the historical data set comprises network transmission data at each historical time point; each historical time point corresponds to a periodic characteristic, and the periodic characteristic is used for indicating the time domain position of the corresponding time point in the belonged time period; the network transmission data is used for indicating a policy feature at a corresponding point in time and network transmission performance, the policy feature is used for indicating at least one of a network transmission control policy and a policy parameter of the network transmission control policy.

In one possible implementation, prior to acquiring the historical dataset, responsive to each historical time point being represented in terms of an absolute time difference from the initial time point, modulo the first time point by a period length of the time period, obtaining a periodic characteristic of the first time point; the first time point is any one of the respective history time points.

In one possible implementation, the initial time refers to an initial running time of the network transmission control system, i.e. the initial time counted from when the network transmission control system starts to run, or the initial time is a fixed time point in the past, for example, the initial time counted from 0 point in the past day, for example, UNIX time, i.e. the total seconds from 1 month 1 day 0 minute 0 second in 1970 in the coordinated world time to the present, without consideration of leap seconds.

In one possible implementation, the network transmission data includes historical policy characteristics at corresponding points in time, and network transmission performance at corresponding points in time; referring to table 1, a historical data set provided by the present application according to an exemplary embodiment is shown, including multiple sets of network transmission data:

TABLE 1

Wherein, in each set of historical policy features, a first number represents a number of a policy used by a network transport connection under the network transport data, and subsequent values represent corresponding parameters in the policy. It should be noted that, the relevant data of the network transmission data shown in table 1 is only illustrative, and the present application does not limit the number of network transmission data in the historical data set and the number of data in the policy feature corresponding to each group of network transmission data.

In one possible implementation, to sense the periodic variation of the network transmission performance, the absolute time difference of the network transmission data is subjected to a modulo process according to a period duration, where the period duration may be hourly, daily, weekly, or the like.

The embodiment of the application takes the period duration as a daily example to describe the process of performing the modulo processing on the absolute time difference of the network transmission data according to the period duration, and the process is realized as follows:

t _i-1 ＝(t _i-1 /3600)％24

Wherein t is _i-1 Representing absolute time difference corresponding to network transmission data, t _i-1 The term "time unit of absolute time difference" is converted from seconds to hours, 24 means 24 hours a day, and the term "time unit after hour" is modulo 24 means 24 hours in terms of cycle length. And acquiring the time obtained by modulo the absolute time difference as the periodic characteristic of the corresponding time point.

In one possible implementation, the network transmission performance corresponding to each set of network transmission data is used to characterize the network transmission quality of each set of network transmission connection, where, for example, the network transmission performance may include one of a bandwidth of the TCP connection, a time required for establishing the connection, a rate of downloading a file, and a throughput.

In one possible implementation manner, the network transmission performance corresponding to each set of network transmission data is a characteristic transmission performance parameter in transmission performance parameters corresponding to a plurality of transmission connections corresponding to each set of corresponding time periods, where the network transmission performance is used to embody a working effect of policy features in each time period, for example, the network transmission performance may be a median, or an average number, among the network transmission performances corresponding to the plurality of transmission connections.

It should be noted that, for the acquisition of the network transmission performance, different acquisition methods may be selected according to the actual service requirement, which is not limited by the present application.

In one possible implementation, in response to each historical time point being represented in the form of a clock time, and the time period being periodic with a duration of 24 hours, acquiring the clock time at the first time point as a periodic characteristic of the first time point; the first time point is any one of the respective history time points.

In one possible implementation, the clock time is a time with 24 hours as a period, such as system time, local time, or UTC (Coordinated Universal Time ), etc. For example, the clock time may be 2020.06.22.00.00, which represents point number 0 of month 22 of year 2020. The clock time itself may embody the periodic characteristics of each historical time point, and thus, the clock time may be directly acquired as the periodic characteristics of the corresponding time point.

In one possible implementation, in response to the transmission state information of the first time point being the same as the transmission state information of the second time point in the respective historical time points, combining the network transmission data of the first time point and the network transmission performance of the second time point;

That is, for network transmission data with consistent transmission status information in the historical time points, which indicates that the same policy feature has been used at the same time point in the history, the process is implemented by combining the network transmission data at the first time point and the network transmission performance at the second time point:

or alternatively, the process may be performed,

based on the network transmission performance at the time point before the time point in the first time point and the second time point, the network transmission performance at the time point after the time point is adjusted.

In one possible implementation manner, the network transmission performance of the time point with the front time point is modified to the network transmission performance of the time point with the rear time point in the first time point and the second time point, and the network transmission performance of the time point with the front time point is replaced by the network transmission performance of the time point with the rear time point, so that the network transmission performance between the same time points of the transmission state information is kept consistent, and the instantaneity of the network transmission performance is ensured.

In one possible implementation, the adjusting of the network transmission performance at the time later point in time based on the network transmission performance at the time earlier point in time is performed by respectively giving different weights to the network transmission performance at the first point in time and the second point in time, or by adding or subtracting a specified step size to the network transmission at the time later point in time.

Step 320, obtain periodic characteristics of the target time point.

In one possible implementation, the network transmission performance at each historical time point is modified according to the periodic characteristic of the target time point and the periodic characteristic of each historical time point.

In one possible implementation manner, a time difference between the periodic feature of the target time point and the periodic feature of each historical time point is obtained, and for the historical time points similar to the periodic feature of the target time point, corresponding network transmission performance enhancement processing is performed; for the historical time points which are far away from the periodic characteristics of the target time points, the corresponding network transmission performance is weakened, so that the guiding effect of the network transmission data of the historical time points which are close to the periodic characteristics of the target time points on the target time points is improved.

Step 330, constructing a fitting function based on the historical data set, wherein the fitting function is used for indicating the relation between the transmission state information of each historical time point and the network transmission performance corresponding to each historical time point; the transmission status information is used to indicate the periodic characteristics of the corresponding time points, and the policy characteristics of the corresponding time points.

In one possible implementation, the fitting function is constructed based on a historical dataset in combination with a proxy model, which is a gaussian process model.

In one possible implementation, the i-th obtained historical dataset is represented as:

wherein->Representing the i-1 st updated network transmission data +.>Wherein t is _i-1 Representing the periodic characteristics corresponding to the i-1 st update,/th update>Representing policy features obtained from the i-1 st update,/->Representing the transmission performance parameters under the policy characteristics after the i-1 th update. And a fitting function constructed by combining the historical data set D with the proxy model is G (X), wherein the fitting function is used for indicating the periodic characteristics of each historical time point and the relation between the corresponding strategy characteristics and the transmission performance parameters corresponding to each historical time point.

In one possible implementation, the fitting function is constructed based on the historical dataset by the network transmission data of a portion of the historical time points in the historical dataset, implemented as:

Extracting each reference time point from each historical time point based on the periodic characteristics of the target time point and the periodic characteristics of each historical time point;

and constructing a fitting function based on the periodic characteristics of each reference time point and the network transmission data of each reference time point.

In one possible implementation, the reference time point is a historical time point that differs from the periodic characteristics of the target time point by a specified duration (e.g., within 6 hours before and after).

Step 340, extracting prediction strategy features from the fitting function based on the periodic features of the target time points.

In one possible implementation, based on the periodic features of the target time points, extracting reference evaluation points from the fitting function, wherein the reference evaluation points comprise strategy features for maximizing the network transmission performance at the target time points;

the policy features contained in the reference evaluation points are extracted as predicted policy features.

In one possible implementation, the prediction strategy features are extracted from the fitting function based on the periodic features of the target time points using an extraction model whose extraction time is the periodic features of the target time points, and the extraction function f () obtains the target time points t _i The predictive policy feature of (2) may be implemented as:

wherein, the liquid crystal display device comprises a liquid crystal display device, and the prediction strategy characteristic is obtained.

In one possible implementation, the predictive policy feature is randomly generated in response to the historical dataset being empty.

For example, when the network transmission control system acquires the prediction policy feature for the first time, the client and the server have not generated network transmission, the historical data set has not stored network transmission data, and the historical data set is empty, so as to randomly generate the prediction policy feature for realizing network transmission between the client and the server, so as to perform network transmission control. Referring to table 2, the policy features corresponding to the target time points provided by the present application according to an exemplary embodiment are shown:

TABLE 2

Start time(s)	Policy features
		1585724700	{1,0.5,0.8}

It is shown that the network transmission control is performed using the policy feature {1,0.5,0.8} in the time interval from 1585724700 s.

And 350, performing network transmission control according to the prediction strategy characteristics in a time interval corresponding to the target time point.

In one possible implementation manner, at least one of the following network transmission controls is performed according to the policy characteristics in a time interval corresponding to the target time point: transmission control protocol TCP control, audio and video transmission performance control, content delivery network CDN scheduling control, and network bandwidth scheduling control.

Taking TCP control as an example, in a time interval corresponding to a target time point, running a congestion control strategy according to strategy characteristics so as to perform network transmission control.

The application provides a network transmission control, the system includes: the system comprises a log monitoring server, a strategy parameter optimizing server and a transmission control server; referring to fig. 4, a schematic diagram of a network transmission control system according to an exemplary embodiment of the present application is shown, and as shown in fig. 4, the network transmission control system 400 includes a log monitoring server 410, a policy parameter optimization server 420, and a transmission control server 430.

Wherein, the log monitoring server 410 is configured to record a historical data set, where the historical data set includes network transmission data at each historical time point; each historical time point corresponds to a periodic characteristic, and the periodic characteristic is used for indicating the time domain position of the corresponding time point in the belonged time period; the network transmission data is used for indicating the strategy characteristics and the network transmission performance at the corresponding time points, and the strategy characteristics are used for indicating at least one of the network transmission control strategy and the strategy parameters of the network transmission control strategy;

a policy parameter optimization server 420 for recording a historical dataset comprising network transmission data at each historical point in time; each historical time point corresponds to a periodic characteristic, and the periodic characteristic is used for indicating the time domain position of the corresponding time point in the belonged time period; the network transmission data is used for indicating the strategy characteristics and the network transmission performance at the corresponding time points, and the strategy characteristics are used for indicating at least one of the network transmission control strategy and the strategy parameters of the network transmission control strategy;

and the transmission control server 430 is configured to perform network transmission control according to the prediction policy feature in a time interval corresponding to the target time point.

In one possible implementation manner, the policy parameter optimization server performs the obtaining of the predicted policy feature corresponding to the target time point according to the policy feature update period set by the policy parameter optimization server, or the policy parameter optimization server performs the obtaining of the predicted policy feature corresponding to the target time point in response to the predicted policy feature obtaining request after receiving the predicted policy feature obtaining request of the transmission control server.

In one possible implementation, the log monitoring server includes a client log server and a transmission control end log server, where the client log server is configured to monitor network transmission connection data sent by the client, and the transmission control end log server is configured to monitor network transmission connection data sent by the transmission control server. The dashed line part shown in fig. 4 represents the prediction policy feature acquisition according to the client log server, and the solid line part shown in fig. 4 represents the prediction policy feature acquisition according to the transmission control side log server.

Taking network transmission control as TCP control, the log monitoring server as a TCP log server, and the policy parameter optimization server receives a predicted policy feature acquisition request of the transmission control server (TCP server), and then responds to the predicted policy feature acquisition request to perform acquisition of a predicted policy feature corresponding to a target time point, for example, the network transmission control method provided by the present application is described, please refer to fig. 5, which shows a timing diagram of a network transmission control method provided by an exemplary embodiment of the present application, as shown in fig. 5, and the method is interactively performed by the transmission control server, the log monitoring server, and the policy parameter optimization server, as shown in fig. 5, and the network transmission control method includes the following steps:

In step 501, the TCP server sends a predicted policy feature acquisition request to the policy parameter optimization server, and the policy parameter optimization server receives the predicted policy feature acquisition request sent by the TCP server.

And acquiring the time point of the TCP server sending the predicted policy feature acquisition request to the policy parameter optimizing server as a target time point.

In one possible implementation, the TCP server sends a prediction policy feature acquisition request to the policy parameter optimization server at a first time interval, assuming that the current time is t _i I.e. the moment when the TCP server currently sends the prediction policy feature acquisition request is t _i The next time the TCP server sends the prediction policy feature acquisition request is t _i +T, where T is the first time interval.

In one possible implementation, the predicted policy feature acquisition request sent by the TCP server includes acquiring at least one of a TCP policy and a TCP parameter, the TCP policy and the TCP parameter being used to run the congestion control policy.

In step 502, the policy parameter optimizing server sends a history data set obtaining request to the log monitoring server and the TCP server, and the log monitoring server receives the history data set obtaining request sent by the policy parameter optimizing server.

In one possible implementation, the method includes that the TCP server sends a prediction policy feature acquisition request to a policy parameter optimization server in units of a first time interval, and the policy parameter optimization server requests a historical data set corresponding to the TCP server from a log monitoring server in units of the first time interval.

After receiving the predicted policy feature acquisition request sent by the TCP server, the policy parameter optimization server sends a history data set update request to the log monitoring server, the history data set update request requiring the log monitoring server to update the history data setTo the nearest state to the target point in time. For example, the first time the TCP server sends a policy feature update to the policy parameter optimizing server is t ₀ The moment when the TCP server sends the policy feature update to the policy parameter optimizing server for the second time is t ₁ ，t ₁ ＝t ₀ +T, T is the first time interval, at T ₀ To t ₁ During this time, the TCP connections that occur at the TCP server generate a corresponding TCP log that includes the relevant data for each TCP connection during the period, and the historical dataset update request requests that the network transmission data during the period be obtained from the TCP log, updating the historical dataset.

In step 503, the log monitor server extracts network transmission data in the TCP log in the last first time interval.

In one possible case, the TCP log generated by the TCP connection generated by the TCP server is sent to the log monitor server in non-real time, and is sent in units of a second time interval, wherein the second time interval is not greater than the first time interval, so as to ensure timeliness of TCP log update, that is, assuming that the moment when the TCP server sends the TCP log to the log monitor server for the first time is t ₀ The moment when the TCP server sends the TCP log to the log monitoring server for the second time is t ₁ ’，t ₁ ’＝t ₀ +T _r ，T _r T.ltoreq.t, so when the policy parameter optimization server sends a historical dataset update request to the log monitoring server, the log monitoring server will aggregate T in response to the request ₀ To t ₁ The TCP server reports the TCP log of each TCP connection to the log monitoring server in the time period to obtain network transmission data in the period; t is 5s, T _r Taking 2s as an example, assuming that the time when the TCP server sends the TCP log to the log monitoring server for the first time and the time when the TCP server sends the policy feature update to the policy parameter optimizing server for the first time are both 0s, the time when the TCP server sends the policy feature update to the policy parameter optimizing server for the second time is 5s, during which the TCP server sends the TCP log to the log monitoring server twice, namely 0-2 s and 2-4 s of TCP log respectively, the time is 0-2 s When the log monitoring server receives a historical data set updating request sent by the strategy parameter optimizing server at the 5 th s, the log monitoring server responds to the request to summarize TCP logs which can be obtained within 0-5 seconds, namely TCP logs of 0-4 seconds, and extracts network transmission data from the TCP logs to update the historical data set; when the TCP server sends a historical data set update request to the policy parameter optimization server for the third time, the TCP server sends TCP logs for the third time, namely 4-6 s, 6-8 s and 8-10 s, to the log monitoring server in the period of 5s to 10s, and at the moment, the log monitoring server receives the historical data set update request sent by the policy parameter optimization server for the third time, and the log monitoring server updates network transmission data extracted from the TCP logs which can be obtained in the period of 5-10 s on the basis of the historical data set corresponding to the second prediction policy feature acquisition request, so that a historical data set of 0-10 s is obtained.

In one possible scenario, assuming the current policy feature request is the (i+1) th request, the log monitoring server responds to the request from t _i To t _i+1 Acquiring transmission performance parameters corresponding to each TCP connection and extracting absolute time difference of network transmission data from TCP logs corresponding to the TCP connections in a time period, summarizing the transmission performance parameters corresponding to each TCP connection, acquiring network transmission performance of the historical time point, performing modulo processing on the absolute time difference of the extracted network transmission data according to cycle time length, and acquiring the network transmission data corresponding to the latest first time interval after periodic characteristics of the historical time point are acquired Wherein t is _i Representing the periodic characteristics corresponding to the ith update,/->Representing policy features resulting from the ith update, < + >>Indicating the network transmission performance in the time interval under the policy feature updated at the ith time.

In one possible implementation manner, the network transmission performance is throughput, and the summarizing the transmission performance parameters corresponding to each TCP connection may be to obtain an average value of the throughput in the last first time interval, or obtain a median of the throughput in the first time interval, or the like, that is, obtain, from the transmission performance parameters of all the TCP connections in the last first time interval, a characteristic transmission performance parameter capable of reflecting the working effect of the policy feature in the time zone, as the network transmission performance of the time zone.

In one possible implementation, the policy characteristics used by the TCP connections within the same first time interval are the same, and the TCP connection data is deleted in response to at least one TCP connection being different from the policy characteristics used by other TCP connections in each of the TCP connections within the same first time interval.

That is, when a singular point exists in the same period, the singular point is removed to ensure the accuracy of the obtained data.

In step 504, the log monitoring server adds the network transmission data in the last first time interval to the candidate historical data set to obtain the historical data set.

In one possible implementation, before the log monitoring server receives the current (i+1th) request for acquiring the network transmission data set, the log monitoring server has received other requests for acquiring the network transmission data set, that is, before the network transmission data in the TCP log in the last first time interval is extracted, the log monitoring server has extracted and updated the corresponding network transmission data in the TCP in response to the other requests for acquiring the network transmission data set (i-th request), so as to form a candidate historical data set, which is expressed as:

wherein D is ₁ A set of candidate historical data is represented,representing the network transmission data after the i-1 th update.

Adding network transmission data in the last first time interval into a candidate historical data set to obtain a historical transmission record D ₂ ：

In one possible implementation, in response to the policy feature in the network transmission data in the most recent first time interval being the same as at least one policy feature in the candidate historical data set, the same policy feature is described as having been used before, the past record is replaced with network transmission performance in the network transmission data in the most recent first time interval, e.g., And->The same will->Corresponding->Replaced by->Thereby ensuring the instantaneity of the network transmission performance.

And adding the network transmission data in the latest first time interval into the updated candidate historical data set to obtain a historical characteristic data set.

That is, when the same policy feature has been used in the same time historically and there is a difference between the network transmission performance in the time historically and the network transmission performance in the last first time interval, the network transmission performance in the time historically is replaced with the network transmission performance in the last first time interval, that is, the network transmission performance in the time historically is considered to have failed, based on the latest network transmission performance.

Alternatively, different weights may be set for the network transmission performance at the time historically and the network transmission performance in the last first time interval to correct the effect of the policy feature in the last first time interval in combination with the network transmission performance historically, where the weight of the network transmission performance corresponding to the historical time closer to the target time point is higher.

In step 505, the log monitoring server returns the historical data set to the policy parameter optimizing server, and the policy parameter optimizing server receives the historical data set correspondingly.

In step 506, the policy parameter optimization server obtains the periodic characteristics of the target time point.

In step 507, the policy parameter optimization server obtains the predicted policy feature corresponding to the target time point according to the periodic feature of the target time point, the periodic feature of each historical time point, and the historical dataset.

The implementation process of steps 505 to 507 may refer to the relevant content in the embodiments shown in fig. 2 and 3, and will not be described herein.

In one possible implementation, the policy parameter optimization server randomly obtains policy features in response to the historical dataset being empty.

In step 508, the policy parameter optimizing server sends the predicted policy feature to the TCP server, and correspondingly, the TCP server receives the predicted policy feature.

In step 509, the tcp server performs network transmission control according to the prediction policy feature in a time interval corresponding to the target time point.

It should be noted that, when the log monitoring server is the client log server, the network transmission control method is similar to that when the log monitoring server is the transmission control end server, the step of acquiring the historical data set from the transmission control end server by the policy parameter optimization server is changed to that of acquiring the historical data set from the client log server, and the process of generating the historical data set by the client log server is the same as that of generating the historical data set by the transmission control end server, which is not repeated herein.

Referring to fig. 6, which is a block diagram illustrating a network transmission control apparatus according to an exemplary embodiment of the present application, as shown in fig. 6, the apparatus may be applied to a network transmission control system, which may be the network transmission control system shown in fig. 1 or 4, to perform all or part of the steps of the embodiment shown in fig. 2, 3 or 5, as shown in fig. 6, and the apparatus includes:

a data set acquisition module 601, configured to acquire a historical data set, where the historical data set includes network transmission data at each historical time point; each historical time point corresponds to a periodic characteristic, and the periodic characteristic is used for indicating the time domain position of the corresponding time point in the belonged time period; the network transmission data is used for indicating the strategy characteristics and the network transmission performance at the corresponding time points, and the strategy characteristics are used for indicating at least one of the network transmission control strategy and the strategy parameters of the network transmission control strategy;

A first feature acquisition module 602, configured to acquire a periodic feature of a target time point;

the policy feature obtaining module 603 is configured to obtain a prediction policy feature corresponding to the target time point according to the periodic feature of the target time point, the periodic feature of each historical time point, and the historical dataset;

and the transmission control module 604 is configured to perform network transmission control according to the prediction policy feature in a time interval corresponding to the target time point.

In one possible implementation, the policy feature acquisition module 603 includes:

the fitting function construction submodule is used for constructing a fitting function based on the historical data set, and the fitting function is used for indicating the relation between the transmission state information of each historical time point and the network transmission performance corresponding to each historical time point; the transmission state information is used for indicating the periodic characteristics of the corresponding time points and the strategy characteristics of the corresponding time points;

extracting a reference evaluation point from the fitting function based on the periodic characteristics of the target time point, wherein the reference evaluation point comprises strategy characteristics which enable the network transmission performance to be maximized at the target time point;

In one possible implementation, the apparatus further includes:

a second feature acquisition module for obtaining a periodic feature of the first time point by modulo the period length of the time period in response to each historical time point expressed in the form of an absolute time difference from the initial time point before the data set acquisition module 601 acquires the historical data set; the first time point is any one of the respective history time points.

In one possible implementation, the apparatus further includes:

a third feature acquiring module, configured to, before the data set acquiring module 601 acquires the historical data sets, respond to each historical time point by representing in the form of clock time, and the time period is a period taking the duration of 24 hours as a period, and acquire the clock time of the first time point as a periodic feature of the first time point; the first time point is any one of the respective history time points.

In one possible implementation, the apparatus further includes:

a merging module, configured to merge the network transmission performance at the first time point and the network transmission performance at the second time point in response to the transmission state information at the first time point being the same as the transmission state information at the second time point in each historical time point;

In one possible implementation, the combining module is configured to, in use,

or alternatively, the process may be performed,

In one possible implementation, the apparatus further includes:

and the strategy characteristic generation module is used for randomly generating the prediction strategy characteristic in response to the historical data set being empty.

In one possible implementation, the apparatus further includes:

the correction module is configured to correct, before the policy feature obtaining module 603 obtains the prediction policy feature corresponding to the target time point according to the periodic feature of the target time point, the periodic feature of each historical time point, and the historical dataset, the network transmission performance of each historical time point according to the periodic feature of the target time point and the periodic feature of each historical time point.

In a possible implementation manner, the transmission control module 604 is configured to perform at least one of the following network transmission control according to the policy feature in a time interval corresponding to the target time point: transmission control protocol TCP control, audio and video transmission performance control, content delivery network CDN scheduling control, and network bandwidth scheduling control.

Fig. 7 is a block diagram illustrating a computer device 700, according to an example embodiment. The computer device 700 may be a client, such as a smart phone, tablet, MP3 player (Moving Picture Experts Group Audio Layer III, mpeg 3), MP4 (Moving Picture Experts Group Audio Layer IV, mpeg 4) player, notebook, or desktop. The computer device 700 may also be referred to by other names of user devices, portable terminals, laptop terminals, desktop terminals, and the like.

In general, the computer device 700 includes: a processor 701 and a memory 702.

Processor 701 may include one or more processing cores, such as a 4-core processor, a 7-core processor, and the like. The processor 701 may be implemented in at least one hardware form of DSP (Digital Signal Processing ), FPGA (Field-Programmable Gate Array, field programmable gate array), PLA (Programmable Logic Array ). The processor 701 may also include a main processor, which is a processor for processing data in an awake state, also referred to as a CPU (Central Processing Unit ); a coprocessor is a low-power processor for processing data in a standby state. In some embodiments, the processor 701 may integrate a GPU (Graphics Processing Unit, image processor) for rendering and drawing of content required to be displayed by the display screen. In some embodiments, the processor 701 may also include an AI (Artificial Intelligence ) processor for processing computing operations related to machine learning.

Memory 702 may include one or more computer-readable storage media, which may be non-transitory. The memory 702 may also include high-speed random access memory, as well as non-volatile memory, such as one or more magnetic disk storage devices, flash memory storage devices. In some embodiments, a non-transitory computer readable storage medium in memory 702 is used to store at least one instruction for execution by processor 701 to implement the network transmission control method provided by the method embodiments of the present application.

In some embodiments, the computer device 700 may further optionally include: a peripheral interface 703 and at least one peripheral. The processor 701, the memory 702, and the peripheral interface 703 may be connected by a bus or signal lines. The individual peripheral devices may be connected to the peripheral device interface 703 via buses, signal lines or a circuit board. Specifically, the peripheral device includes: at least one of radio frequency circuitry 704, touch display 705, camera 706, audio circuitry 707, positioning component 708, and power supply 709.

In some embodiments, the computer device 700 also includes one or more sensors 710. The one or more sensors 710 include, but are not limited to: acceleration sensor 711, gyroscope sensor 712, pressure sensor 713, fingerprint sensor 714, optical sensor 715, and proximity sensor 716.

Those skilled in the art will appreciate that the architecture shown in fig. 7 is not limiting of the computer device 700, and may include more or fewer components than shown, or may combine certain components, or employ a different arrangement of components.

Fig. 8 is a block diagram illustrating a computer device 800, according to an example embodiment. The computer device may be implemented as a server in the above-described aspects of the present application. The computer apparatus 800 includes a central processing unit (Central Processing Unit, CPU) 801, a system Memory 804 including a random access Memory (Random Access Memory, RAM) 802 and a Read-Only Memory (ROM) 803, and a system bus 805 connecting the system Memory 804 and the central processing unit 801. The computer device 800 also includes a basic Input/Output system (I/O) 806 for facilitating the transfer of information between the various devices within the computer, and a mass storage device 807 for storing an operating system 813, application programs 814, and other program modules 815.

The basic input/output system 806 includes a display 806 for displaying information and an input device 809, such as a mouse, keyboard, or the like, for user input of information. Wherein the display 806 and the input device 809 are connected to the central processing unit 801 via an input output controller 810 connected to a system bus 805. The basic input/output system 806 can also include an input/output controller 810 for receiving and processing input from a number of other devices, such as a keyboard, mouse, or electronic stylus. Similarly, the input output controller 810 also provides output to a display screen, a printer, or other type of output device.

The mass storage device 807 is connected to the central processing unit 801 through a mass storage controller (not shown) connected to the system bus 805. The mass storage device 807 and its associated computer-readable media provide non-volatile storage for the computer device 800. That is, the mass storage device 807 may include a computer readable medium (not shown) such as a hard disk or a compact disk-Only (CD-ROM) drive.

The computer readable medium may include computer storage media and communication media without loss of generality. Computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes RAM, ROM, erasable programmable read-Only register (Erasable Programmable Read Only Memory, EPROM), electrically erasable programmable read-Only Memory (EEPROM) flash Memory or other solid state Memory technology, CD-ROM, digital versatile disks (Digital Versatile Disc, DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices. Of course, those skilled in the art will recognize that the computer storage medium is not limited to the one described above. The system memory 804 and mass storage device 807 described above may be collectively referred to as memory.

According to various embodiments of the disclosure, the computer device 800 may also operate by being connected to a remote computer on a network, such as the Internet. I.e., the computer device 800 may be connected to a network 812 through a network interface unit 811 connected to the system bus 805, or other types of networks or remote computer systems (not shown) may be connected to the system using the network interface unit 811.

The memory further includes at least one instruction, at least one program, a code set, or an instruction set, where the at least one instruction, the at least one program, the code set, or the instruction set is stored in the memory, and the central processor 801 implements all or part of the steps in the network transmission control method shown in the above embodiments by executing the at least one instruction, the at least one program, the code set, or the instruction set.

In an exemplary embodiment, a non-transitory computer readable storage medium is also provided, such as a memory including at least one instruction, at least one program, code set, or instruction set executable by a processor to perform all or part of the steps of the methods illustrated in any of the embodiments of fig. 2, 3, or 5 described above. For example, the non-transitory computer readable storage medium may be ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.

In an exemplary embodiment, a computer program product or computer program is provided, the computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions so that the computer device performs all or part of the steps of the network transmission control method described above.

Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. This application is intended to cover any variations, uses, or adaptations of the application following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the application pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.

It is to be understood that the application is not limited to the precise arrangements and instrumentalities shown in the drawings, which have been described above, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the application is limited only by the appended claims.

Claims

1. A network transmission control method, the method comprising:

acquiring periodic characteristics of a target time point;

2. The method according to claim 1, wherein the obtaining the prediction strategy feature corresponding to the target time point according to the periodicity feature of the target time point, the periodicity feature of each historical time point, and the historical dataset includes:

Constructing a fitting function based on the historical data set, wherein the fitting function is used for indicating the relation between the transmission state information of each historical time point and the network transmission performance corresponding to each historical time point; the transmission state information is used for indicating the periodic characteristics of the corresponding time points and the strategy characteristics of the corresponding time points;

and extracting the prediction strategy characteristic from the fitting function based on the periodic characteristic of the target time point.

3. The method of claim 2, wherein the extracting the predictive strategy feature from the fitting function based on the periodic feature of the target point in time comprises:

4. The method of claim 2, wherein the constructing a fitting function based on the historical dataset comprises:

5. The method of claim 1, wherein prior to the acquiring the historical dataset, further comprising:

in response to the respective historical time points being represented in the form of absolute time differences from an initial time point, modulo a first time point according to a period length of the time period, obtaining a periodic characteristic of the first time point; the first time point is any one of the respective history time points.

6. The method of claim 1, wherein prior to the acquiring the historical dataset, further comprising:

responsive to the respective historical time points being represented in the form of clock times and the time period being periodic with a duration of 24 hours, obtaining a clock time at a first time point as a periodic characteristic of the first time point; the first time point is any one of the respective history time points.

7. The method according to claim 5 or 6, characterized in that the method further comprises:

Combining the network transmission performance at the first time point and the network transmission performance at the second time point in response to the transmission state information at the first time point being the same as the transmission state information at the second time point in the respective historical time points;

8. The method of claim 7, wherein the combining the network transmission data at the first point in time and the network transmission performance at the second point in time comprises:

or alternatively, the process may be performed,

9. The method according to claim 1, wherein the method further comprises:

the predictive policy feature is randomly generated in response to the historical dataset being empty.

10. The method according to claim 1, wherein before the obtaining the prediction policy feature corresponding to the target time point according to the periodic feature of the target time point, the periodic feature of each historical time point, and the historical dataset, further comprises:

and correcting the network transmission performance of each historical time point according to the periodic characteristics of the target time point and the periodic characteristics of each historical time point.

11. The method according to claim 1, wherein the performing network transmission control according to the policy feature in the time interval corresponding to the target time point includes:

and in a time interval corresponding to the target time point, at least one of the following network transmission control is performed according to the strategy characteristics: transmission control protocol TCP control, audio and video transmission performance control, content delivery network CDN scheduling control, and network bandwidth scheduling control.

12. A network transmission control system, the system comprising: the system comprises a log monitoring server, a strategy parameter optimizing server and a transmission control server;

13. A network transmission control apparatus, the apparatus comprising:

14. A computer device comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, a set of codes, or a set of instructions, the at least one instruction, the at least one program, the set of codes, or the set of instructions being loaded and executed by the processor to implement the network transmission control method of any one of claims 1 to 11.

15. A computer-readable storage medium, characterized in that at least one instruction, at least one program, a set of codes, or a set of instructions is stored in the storage medium, the at least one instruction, the at least one program, the set of codes, or the set of instructions being loaded and executed by a processor to implement the network transmission control method according to any one of claims 1 to 11.