CN113079045B

CN113079045B - Bandwidth allocation method, device, server and storage medium

Info

Publication number: CN113079045B
Application number: CN202110328426.XA
Authority: CN
Inventors: 连博; 赵凯传
Original assignee: Beijing Dajia Internet Information Technology Co Ltd
Current assignee: Beijing Dajia Internet Information Technology Co Ltd
Priority date: 2021-03-26
Filing date: 2021-03-26
Publication date: 2022-03-25
Anticipated expiration: 2041-03-26
Also published as: CN113079045A

Abstract

The disclosure relates to a bandwidth allocation method, a bandwidth allocation device, a server and a storage medium, and belongs to the technical field of computers. The bandwidth allocation method comprises the following steps: acquiring a first bandwidth corresponding to a first time period, wherein the first time period is any time period after the current time; acquiring first performance parameter information of each content distribution server, wherein the first performance parameter information of each content distribution server represents the content distribution performance of each content distribution server; determining the bandwidth allocation proportion of each content distribution server based on the first performance parameter information of each content distribution server, wherein the bandwidth allocation proportion of each content distribution server is positively correlated with the content distribution performance of the content distribution server; and determining the target bandwidth of each content distribution server based on the first bandwidth and the bandwidth distribution proportion of each content distribution server. The present disclosure improves the bandwidth utilization by enabling the bandwidth of each content distribution server to be effectively utilized according to its content distribution performance.

Description

Bandwidth allocation method, device, server and storage medium

Technical Field

The present disclosure relates to the field of computer technologies, and in particular, to a bandwidth allocation method, apparatus, server, and storage medium.

Background

In recent years, mass multimedia contents are spread through networks, and a huge content transmission demand is generated. To meet the network load challenges caused by the huge content transmission demands, the scheduling server of the content service provider typically schedules the content distribution server for content distribution.

In the content distribution process, the bandwidth of the content distribution server needs to be consumed, usually, the scheduling server will schedule any one content distribution server to distribute the content first, and when the bandwidth required by the content distribution exceeds the upper limit of the bandwidth of the currently scheduled content distribution server, another content distribution server is scheduled according to the exceeded bandwidth, and the content distribution is performed by the other content distribution server. However, in such a manner of sequentially scheduling the content distribution servers, some content distribution servers are not scheduled, and the bandwidth of each content distribution server cannot be effectively utilized, which results in a low bandwidth utilization rate.

Disclosure of Invention

The present disclosure provides a bandwidth allocation method, apparatus, server, and storage medium, which can improve bandwidth utilization.

According to a first aspect of the embodiments of the present disclosure, there is provided a bandwidth allocation method, including:

acquiring a first bandwidth corresponding to a first time period, wherein the first time period is any time period after the current time, and the first bandwidth is the total bandwidth consumed for content distribution in the first time period;

acquiring first performance parameter information of each content distribution server, wherein each content distribution server is used for content distribution, and the first performance parameter information of each content distribution server represents the content distribution performance of each content distribution server;

determining a bandwidth allocation proportion of each content distribution server based on first performance parameter information of each content distribution server, wherein the bandwidth allocation proportion of each content distribution server is positively correlated with the content distribution performance of the content distribution server;

and determining the target bandwidth of each content distribution server based on the first bandwidth and the bandwidth distribution proportion of each content distribution server.

According to the technical scheme provided by the embodiment of the disclosure, the bandwidth consumed by content distribution in the future time period is predicted, the bandwidth distribution proportion positively correlated with the content distribution performance of each content distribution server is determined for each content distribution server, and the target bandwidth of each content distribution server required to be consumed in the future time period is determined according to the bandwidth distribution proportion, so that the target bandwidth of each content distribution server is matched with the content distribution performance of each content distribution server, the bandwidth of each content distribution server can be effectively utilized according to the content distribution performance of each content distribution server, and the bandwidth utilization rate is improved.

In some embodiments, the first performance parameter information includes a plurality of performance parameters belonging to different dimensions, the plurality of performance parameters characterizing content distribution performance of the content distribution server from the different dimensions, respectively;

the determining the bandwidth allocation proportion of each content distribution server based on the first performance parameter information of each content distribution server includes:

fusing the multiple performance parameters of each content distribution server to obtain the fusion performance parameters of each content distribution server;

determining the sum of the fusion performance parameters of the content distribution servers as a total performance parameter;

and determining the proportion between the fusion performance parameter of each content distribution server and the total performance parameter as the bandwidth allocation proportion of each content distribution server.

According to the technical scheme, the multiple performance parameters of the content distribution server are fused to obtain the fusion performance parameters, so that the content distribution performance of the content distribution server is comprehensively represented, the bandwidth distribution proportion is determined according to the fusion performance parameters, the bandwidth distribution proportion is matched with the comprehensive content distribution performance of the content distribution server, and the accuracy of determining the bandwidth distribution proportion is improved.

In some embodiments, the fusing the multiple performance parameters of each content distribution server to obtain a fused performance parameter of each content distribution server includes:

acquiring various performance parameters of each content distribution server and weights corresponding to the various performance parameters, wherein the weights represent the importance of the corresponding performance parameters to the content distribution performance of each content distribution server;

and performing weighted fusion on the multiple performance parameters according to the weights corresponding to the multiple performance parameters to obtain the fusion performance parameters of the content distribution servers.

According to the technical scheme, different weights can be given to each performance parameter according to the requirements of the service scene on the content distribution performance in different aspects, and the emphasis is placed on the process of evaluating the content distribution performance of the content distribution server, so that the content distribution performance of the content distribution server can be represented more accurately by fusing the performance parameters.

In some embodiments, the performing weighted fusion on the multiple performance parameters according to the weights corresponding to the multiple performance parameters to obtain the fusion performance parameters of each content distribution server includes:

carrying out normalization processing on the multiple performance parameters;

and performing weighted fusion on the multiple performance parameters after the normalization processing according to the weights corresponding to the multiple performance parameters to obtain the fusion performance parameters of the content distribution servers.

According to the technical scheme, the performance parameters are normalized and converted into the numerical values between 0 and 1, and the difference between the content distribution performances of the content distribution servers is represented more accurately by the numerical values between 0 and 1, so that the accuracy of the performance parameters after normalization on the content distribution performance representation of the content distribution servers is improved.

In some embodiments, the determining the target bandwidth of each content distribution server based on the first bandwidth and the bandwidth allocation ratio of each content distribution server includes:

and determining the bandwidth distribution proportion of each content distribution server as the product of the first bandwidth and the target bandwidth of each content distribution server.

According to the technical scheme, the product of the bandwidth allocation proportion of each content distribution server and the first bandwidth is determined as the target bandwidth of each content distribution server, so that the allocated bandwidth is matched with the content distribution performance of the corresponding content distribution server, the bandwidth allocated to the content distribution server with better content distribution performance is improved, and the overall content distribution performance of the plurality of content distribution servers is guaranteed.

determining a first target bandwidth of each content distribution server by multiplying the bandwidth allocation proportion of each content distribution server by the first bandwidth;

and adjusting the first target bandwidth of each content distribution server to obtain a second target bandwidth of each content distribution server.

In some embodiments, the adjusting the first target bandwidth of each content distribution server to obtain the second target bandwidth of each content distribution server includes:

determining a plurality of bandwidth adjustment strategies, wherein each bandwidth adjustment strategy comprises the adjustment bandwidth of each content distribution server;

according to the multiple bandwidth adjustment strategies, adjusting the first target bandwidth of each content distribution server respectively to obtain an adjusted third target bandwidth, wherein the sum of the third target bandwidth of each content distribution server is the first bandwidth;

under various bandwidth adjustment strategies, determining a target quantity based on a third target bandwidth of each content distribution server, wherein the target quantity represents the total quantity of resources needing to be transferred out when each content distribution server is scheduled by the third target bandwidth of each content distribution server;

and determining the third target bandwidth of each content distribution server in the bandwidth adjustment strategy with the minimum target number as the second target bandwidth of each content distribution server.

According to the technical scheme, after the initial first target bandwidth of each content distribution server is determined based on the bandwidth distribution proportion of each content distribution server, the first target bandwidth is adjusted, and the second target bandwidth which enables the total amount of resources required to be transferred out by a plurality of content distribution servers in the first time period to be the minimum is determined, so that the cost required by scheduling the plurality of content distribution servers in the first time period is reduced, the bandwidth requirement in the first time period can be met by scheduling the content distribution servers under the condition that the total amount of the transferred-out resources is the minimum, and the effective utilization rate of the bandwidth is improved.

determining a plurality of bandwidth adjustment strategies, wherein each bandwidth adjustment strategy comprises adjustment bandwidths of a plurality of first time periods of the plurality of content distribution servers in the same period;

according to the multiple bandwidth adjustment strategies, respectively adjusting the first target bandwidth of each content distribution server in each first time period to obtain an adjusted third target bandwidth, wherein the sum of the third target bandwidth of each content distribution server in each first time period is the first bandwidth corresponding to each first time period;

under various bandwidth adjustment strategies, determining a target quantity based on a third target bandwidth of each content distribution server in the multiple first time periods, wherein the target quantity represents the total quantity of resources which need to be transferred out when each content distribution server is scheduled in the multiple first time periods by the third target bandwidth of each content distribution server in each first time period;

and in the bandwidth adjustment strategy with the minimum target quantity, determining a third target bandwidth of each content distribution server in each first time period as a second target bandwidth of each content distribution server in each first time period.

According to the technical scheme, after the initial first target bandwidth of each content distribution server is determined based on the bandwidth distribution proportion of each content distribution server, the first target bandwidth is adjusted, and the second target bandwidth which enables the total amount of resources required to be transferred out by a plurality of content distribution servers in the first period to be the minimum is determined, so that the cost required to be consumed by scheduling the plurality of content distribution servers in the first period is reduced, the scheduling of the content distribution servers can meet the bandwidth requirement under the condition that the total amount of the transferred resources is the minimum, and the effective utilization rate of the bandwidth is improved.

dividing content distribution servers with the same quantity calculation mode into a group, wherein the quantity calculation mode represents a mode of calculating the quantity of resources to be transferred according to the target bandwidth of the content distribution servers;

and adjusting the first target bandwidth of each content distribution server in each group to obtain the second target bandwidth of each content distribution server.

According to the technical scheme, the content distribution servers are divided into different groups according to different quantity calculation modes, and in the process of adjusting the bandwidth of each content distribution server, the process of determining the quantity of resources to be transferred by the content distribution servers is clearer and clearer, so that the adjustment efficiency of the first target bandwidth is improved.

In some embodiments, after determining the target bandwidth of each of the content distribution servers based on the first bandwidth and the bandwidth allocation ratio of each of the content distribution servers, the bandwidth allocation method further includes:

and scheduling each content distribution server according to the target bandwidth of each content distribution server in the first time period.

In some embodiments, after scheduling each of the content distribution servers according to the target bandwidth of each of the content distribution servers in the first time period, the bandwidth allocation method further includes:

acquiring second performance parameter information of each content distribution server in the first time period, wherein the second performance parameter information of each content distribution server represents the content distribution performance of the content distribution server in the first time period;

and in response to that the second performance parameter information of any content distribution server in the first time period does not meet the performance condition, deducting the target bandwidth of any content distribution server in the next first time period of the first time period from the target adjustment bandwidth.

In some embodiments, after, in response to that the second performance parameter information of any content distribution server in the first time period does not satisfy the performance condition, deducting the target bandwidth of any content distribution server in a first time period next to the first time period by the target adjustment bandwidth, the bandwidth allocation method further includes:

and increasing the target bandwidth of any content distribution server meeting the performance condition in the next first time period of the first time period by the target adjustment bandwidth.

According to the technical scheme, the target bandwidth of the content distribution server in the next time period is dynamically adjusted based on the performance parameter information of the content distribution server in a certain time period in real time, the real-time content distribution performance of the scheduled content distribution server is effectively guaranteed to be at a higher level on the basis of effectively reducing the quantity of resources required to be transferred out, the content distribution server with better content distribution performance is effectively scheduled by consuming lower cost, and the effective utilization rate of the bandwidth is improved.

and adjusting the first target bandwidth of each content distribution server to obtain a second target bandwidth of each content distribution server in response to the fact that the first time period belongs to a time period in a peak period.

According to the technical scheme, the first target bandwidth of each content distribution server corresponding to the first time period belonging to the peak period is adjusted, so that the quantity of resources required to be transferred out by scheduling the content distribution servers in the peak period can be effectively reduced, namely, the cost required to be consumed by scheduling the content distribution servers in the peak period is effectively reduced, and the effective utilization rate of the bandwidth is improved.

In some embodiments, the obtaining the first bandwidth corresponding to the first period includes:

determining second bandwidths corresponding to a plurality of second time periods in the second period based on the downloading behavior information acquired in the second period, wherein the downloading behavior information represents the content downloading condition in the second period, and the second bandwidth corresponding to any one second time period represents the bandwidth consumed for content distribution in any one second time period;

and acquiring first bandwidths corresponding to the plurality of first time periods based on second bandwidths corresponding to the plurality of second time periods.

According to the technical scheme, the bandwidth consumption situation in the historical period is determined through analysis of the historical content downloading situation, the bandwidth consumption situation in the future period is reasonably predicted, the accuracy of bandwidth prediction is improved, further more accurate bandwidth allocation can be achieved based on the predicted bandwidth with higher accuracy, and the availability of bandwidth allocation is further improved.

In some embodiments, the download behavior information includes a location of a terminal that downloads the content and a data amount of the content;

determining second bandwidths corresponding to a plurality of second time periods in the second period based on the download behavior information acquired in the second period includes:

for any second time period in the second period, determining a plurality of contents of which the corresponding terminal is located in any area range from the downloading behavior information collected in the second time period;

and determining a second bandwidth corresponding to the second time period and the area range based on the data volume of the plurality of contents and the duration of the second time period, wherein the second bandwidth represents the bandwidth consumed for content distribution of the terminal located in the area range in the second time period.

According to the technical scheme, the consumed bandwidth in each second time period is counted according to different region ranges, the process of determining the bandwidth is further refined, and the fineness of determining the bandwidth is improved.

In some embodiments, the obtaining the first bandwidths corresponding to the plurality of first time periods based on the second bandwidths corresponding to the plurality of second time periods includes:

and for any area range and any second time period, predicting based on a second bandwidth corresponding to the area range and the second time period to obtain a first bandwidth, wherein the first bandwidth represents the bandwidth required to be consumed for content distribution for terminals located in the area range in the first period corresponding to the second time period.

According to the technical scheme, the bandwidth prediction is carried out based on the finer second bandwidth counted according to the region range, and the accuracy of the bandwidth prediction is improved.

for a first bandwidth corresponding to any area range and any first time period, determining a target bandwidth of each content distribution server with a service range including the area range based on the first bandwidth and a bandwidth allocation proportion of each content distribution server.

According to the technical scheme, the content distribution server with the service range including the area range can be scheduled in a certain area range to meet the bandwidth requirement, so that the scheduled content distribution server is closer to the terminal with the bandwidth requirement, and the content distribution performance of the scheduled content distribution server can be improved.

In some embodiments, the download behavior information includes an operator to which a terminal downloading the content belongs and a data amount of the content;

for any second time period in the second period, determining a plurality of contents of any operator to which the corresponding terminal belongs from the downloading behavior information collected in the second time period;

and determining a second time period and a second bandwidth corresponding to the operator based on the data volume of the plurality of contents and the duration of the second time period, wherein the second bandwidth represents the bandwidth consumed by content distribution through a content distribution server built on the network of the operator in the second time period.

According to the technical scheme, the bandwidth consumed in each second time period is counted according to different operators, the process of determining the bandwidth is further refined, and the fineness of determining the bandwidth is improved.

for any operator and any second time period, predicting based on a second bandwidth corresponding to the operator and the second time period to obtain a first bandwidth, wherein the first bandwidth represents a bandwidth consumed by content distribution through a content distribution server built on a network of the operator in a first time period corresponding to the second time period in the first period.

According to the technical scheme, the bandwidth prediction is carried out based on the more refined second bandwidth counted by the operator, and the accuracy of the bandwidth prediction is improved.

and for a first bandwidth corresponding to any operator and any first time period, determining a target bandwidth of each content distribution server belonging to the operator based on the first bandwidth and a bandwidth allocation proportion of each content distribution server belonging to the operator.

According to the technical scheme, the bandwidth is distributed according to different operators, so that the bandwidth requirements corresponding to different operators are matched with the bandwidth provided by the scheduled content distribution server, and the accuracy of bandwidth scheduling is improved.

According to a second aspect of the embodiments of the present disclosure, there is provided a bandwidth allocation apparatus, including:

a first bandwidth obtaining module configured to perform obtaining of a first bandwidth corresponding to a first time period, where the first time period is any time period after a current time, and the first bandwidth is a total bandwidth consumed for content distribution within the first time period;

a first parameter obtaining module configured to perform obtaining first performance parameter information of each content distribution server, each content distribution server being a server for content distribution, the first performance parameter information of each content distribution server representing content distribution performance of each content distribution server;

a bandwidth distribution ratio determination module configured to perform determination of a bandwidth distribution ratio of each of the content distribution servers based on first performance parameter information of each of the content distribution servers, the bandwidth distribution ratio of each of the content distribution servers being positively correlated with content distribution performance of the content distribution server;

a target bandwidth determination module configured to perform determination of a target bandwidth of each of the content distribution servers based on the first bandwidth and a bandwidth allocation ratio of each of the content distribution servers.

In some embodiments, the first performance parameter information includes a plurality of performance parameters belonging to different dimensions, the plurality of performance parameters characterizing content distribution performance of the content distribution server from the different dimensions, respectively; the bandwidth allocation proportion determining module comprises:

a performance parameter fusion unit configured to perform fusion of a plurality of performance parameters of each of the content distribution servers to obtain a fusion performance parameter of each of the content distribution servers;

a bandwidth allocation proportion determination unit configured to perform determination of a sum value of the fusion performance parameters of the content distribution servers as a total performance parameter; and determining the proportion between the fusion performance parameter of each content distribution server and the total performance parameter as the bandwidth allocation proportion of each content distribution server.

In some embodiments, the performance parameter fusion unit includes:

the weight acquisition subunit is configured to perform acquisition of multiple performance parameters of each content distribution server and a weight corresponding to the multiple performance parameters, where the weight represents an importance of the corresponding performance parameter to content distribution performance of each content distribution server;

and the performance parameter fusion subunit is configured to perform weighted fusion on the multiple performance parameters according to the weights corresponding to the multiple performance parameters to obtain fusion performance parameters of the content distribution servers.

In some embodiments, the performance parameter fusion subunit is configured to perform:

carrying out normalization processing on the multiple performance parameters;

In some embodiments, the target bandwidth determination module is configured to perform:

In some embodiments, the target bandwidth determination module comprises:

a target bandwidth determining unit configured to perform determination of a first target bandwidth of each of the content distribution servers as a product of a bandwidth allocation proportion of each of the content distribution servers and the first bandwidth;

and the bandwidth adjusting unit is configured to adjust the first target bandwidth of each content distribution server to obtain a second target bandwidth of each content distribution server.

In some embodiments, the bandwidth adjusting unit is configured to perform:

In some embodiments, the bandwidth allocation apparatus further comprises:

and the content distribution server scheduling module is configured to execute scheduling of each content distribution server according to the target bandwidth of each content distribution server in the first time period.

In some embodiments, the bandwidth allocation apparatus further comprises:

a second parameter obtaining module configured to perform obtaining of second performance parameter information of each of the content distribution servers in the first time period, the second performance parameter information of each of the content distribution servers representing content distribution performance of the content distribution server in the first time period;

a bandwidth adjustment module configured to perform, in response to second performance parameter information of any content distribution server in the first time period not satisfying a performance condition, deducting a target bandwidth of the any content distribution server in a first time period next to the first time period by a target adjustment bandwidth.

In some embodiments, the bandwidth adjustment module is further configured to perform increasing the target bandwidth of any content distribution server that satisfies the performance condition by the target adjustment bandwidth in a first time period next to the first time period.

In some embodiments, the bandwidth adjusting unit is configured to perform adjusting the first target bandwidth of each of the content distribution servers to obtain the second target bandwidth of each of the content distribution servers in response to the first time period belonging to a time period in a peak period.

In some embodiments, the first cycle includes a plurality of first time periods, the second cycle includes a plurality of second time periods, the second cycle being a cycle before the first cycle, the first bandwidth acquisition module includes:

a second bandwidth determining unit, configured to determine, based on the download behavior information acquired in the second period, second bandwidths corresponding to a plurality of second time periods in the second period, where the download behavior information represents a content download situation in the second period, and a second bandwidth corresponding to any one of the second time periods represents a bandwidth consumed for content distribution in the any one of the second time periods;

a first bandwidth acquiring unit configured to acquire first bandwidths corresponding to the plurality of first time periods based on second bandwidths corresponding to the plurality of second time periods.

the second bandwidth determination unit configured to perform:

In some embodiments, the first bandwidth obtaining unit is configured to perform:

the second bandwidth determination unit configured to perform:

According to a third aspect of embodiments of the present disclosure, there is provided a server, including: one or more processors; a memory for storing instructions executable by the processor; wherein the processor is configured to execute the instructions to implement the bandwidth allocation method of the first aspect.

According to a fourth aspect of embodiments of the present disclosure, there is provided a computer-readable storage medium, wherein instructions, when executed by a processor of a server, enable the server to perform the bandwidth allocation method of the first aspect.

According to yet another aspect of embodiments of the present disclosure, there is provided a computer program product comprising a computer program which, when executed by a processor, implements the bandwidth allocation method of the first aspect.

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 disclosure.

Drawings

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

FIG. 1 is a schematic diagram illustrating one implementation environment in accordance with an example embodiment.

Fig. 2 is a flow diagram illustrating a method of bandwidth allocation in accordance with an example embodiment.

Fig. 3 is a flow diagram illustrating a method of bandwidth allocation in accordance with an example embodiment.

Fig. 4 is a flow diagram illustrating a method of bandwidth allocation in accordance with an example embodiment.

Fig. 5 is a block diagram illustrating a bandwidth allocation apparatus according to an example embodiment.

FIG. 6 is a block diagram illustrating a server in accordance with an example embodiment.

Detailed Description

In order to make the technical solutions of the present disclosure better understood by those of ordinary skill in the art, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.

It should be noted that the terms "first," "second," and the like in the description and claims of the present disclosure and in the description of the above-described figures are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the disclosure described herein are capable of operation in sequences other than those illustrated or otherwise described herein. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.

It should be noted that, the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to behavior data of the user, etc.) referred to in the present disclosure are information and data authorized by the user or sufficiently authorized by each party.

FIG. 1 is a schematic diagram illustrating one implementation environment in accordance with an example embodiment. Referring to fig. 1, the implementation environment includes a terminal 110, an application server 120, and a content distribution server 130.

The terminal 110 may be a smart phone, a smart watch, a desktop computer, a laptop computer, a smart television, a vehicle terminal, or other electronic devices. The terminal 110 is provided with an application client, for example, the application client is an online video client, a short video client, a live broadcast client, an online education client, an audio client, a social client, a shopping client, or a game client.

The application server 120 is a background server for the application client. The application server 120 may be a server, a plurality of servers, a cloud server, a cloud computing platform, or a virtualization center, which is not limited in this disclosure. The content distribution server 130 has functions of content storage and content distribution, includes a plurality of edge servers deployed in different area ranges, and provides a nearby content download service for the terminal 110, so as to improve content distribution efficiency, accelerate the speed of acquiring content by the terminal, improve instantaneity of content display, and improve user experience. In some embodiments, the Content Delivery server 130 is a CDN (Content Delivery Network).

The terminal 110 is directly or indirectly connected to the application server 120 and the content distribution server 130 by wired or wireless communication, and the content distribution server 130 is directly or indirectly connected to the application server 120 by wired or wireless communication.

Where the content distribution server 130 generally refers to one of a plurality of content distribution servers, the disclosed embodiments are illustrated with the content distribution server 130. When a content distribution demand is generated, that is, when the application server 120 receives a content acquisition request transmitted by the terminal 110, address information of the content distribution server 130 is transmitted to the terminal 110; the terminal 110 transmits a content acquisition request to the content distribution server 130 based on the address information of the content distribution server 130; the content distribution server 130 transmits the content requested to be acquired by the content acquisition request to the terminal 110 in response to locally storing the content; the content distribution server 130 acquires the content requested to be acquired by the content acquisition request from the application server 120 in response to the content requested to be acquired by the content acquisition request not being locally stored, and transmits the content to the terminal 110. The content distribution server 130 also stores the content acquired from the application server 120 locally to acquire the content directly from the local storage and transmit the content to the terminal that transmitted the content acquisition request, when the content acquisition request requesting to acquire the content is received next time.

In the above process, the application server 120 assigns the task of transmitting the content to the terminal 110 to the content distribution server 130, and realizes scheduling for the content distribution server 130. In the process of transmitting the content to the terminal 110 by the content distribution server 130, the data amount of the content transmitted to the terminal 110 per unit time is the bandwidth consumed by the content distribution server 130 for content distribution to the terminal 110.

The bandwidth allocation method provided by the embodiment of the disclosure can be applied to any content distribution scene.

For example, in a short video distribution scene, a terminal displays video covers of a plurality of short videos in a short video recommendation interface, and sends a content acquisition request to an application server in response to a click operation of a user on any video cover; the application server receives the content acquisition request and sends address information of a content distribution server to the terminal so as to consume the bandwidth of the content distribution server for content distribution by scheduling the content distribution server, namely, the application server consumes the bandwidth of the content distribution server to send the short video requested to be acquired by the content acquisition request to the terminal; the terminal sends a content acquisition request to the content distribution server based on the address information of the content distribution server; the content distribution server receives the content acquisition request and sends the short video requested to be acquired by the content acquisition request to the terminal; and the terminal receives the short video sent by the content distribution server and plays the short video.

For another example, in a game resource package distribution scene, the terminal responds to a resource package downloading operation and sends a content acquisition request to the application server; the application server receives the content acquisition request, and sends address information of a content distribution server to the terminal, so as to consume the bandwidth of the content distribution server for content distribution by scheduling the content distribution server, that is, the bandwidth of the content distribution server is consumed to send the resource packet requested to be acquired by the content acquisition request to the terminal; the terminal sends a content acquisition request to the content distribution server based on the address information of the content distribution server; the content distribution server receives the content acquisition request and sends a resource packet requested to be acquired by the content acquisition request to the terminal; and the terminal receives the resource package sent by the content distribution server and installs the resource package.

Fig. 2 is a flow diagram illustrating a method of bandwidth allocation in accordance with an example embodiment. Referring to fig. 2, the bandwidth allocation method is performed by an application server and includes the following steps.

201. And acquiring a first bandwidth corresponding to the first time period.

The first time period is any time period after the current time, that is, the first time period is a certain time period in the future. The starting time of the first time period is any time after the current time. The duration of the first time period is flexibly configurable, for example, the duration of the first time period is 1 minute, 5 minutes or 10 minutes, which is not limited by the embodiment of the disclosure. The first bandwidth is a total bandwidth consumed by the application server for content distribution during the first time period.

It should be noted that, the application server receives a plurality of content acquisition requests in the first time period, and the application server schedules a plurality of content distribution servers to consume bandwidths of the plurality of content distribution servers for content distribution. Wherein, the content distribution is directed to a terminal sending a content acquisition request, and the content requested to be acquired by the content acquisition request is sent. In the process of sending the content to the terminal by the content distribution server, the content distribution server sends a certain amount of content to the terminal in each unit time until all the content requested to be acquired by the terminal is completely transmitted.

The content distribution server transmitting the content to the terminal consumes data transmission resources of the content distribution server, that is, bandwidth of the content distribution server. The total bandwidth consumed by the application server to perform content distribution in the first time period is also the total bandwidth of the plurality of content distribution servers that is required to be consumed by the application server to schedule the plurality of content distribution servers to perform content distribution in the first time period.

202. First performance parameter information of each content distribution server is acquired.

Each content distribution server is a server for distributing content, and each content distribution server is each of a plurality of content distribution servers that can be scheduled by an application server. The first performance parameter information of each content distribution server represents the content distribution performance of each content distribution server, that is, one content distribution server corresponds to one first performance parameter information, and the first performance parameter information of one content distribution server represents the content distribution performance of the content distribution server. The content distribution performance is performance that the content distribution server exhibits in the process of transmitting the content that each terminal requests to acquire to each terminal.

203. And determining the bandwidth allocation proportion of each content distribution server based on the first performance parameter information of each content distribution server.

The bandwidth allocation ratio of each content distribution server is positively correlated with the content distribution performance of each content distribution server, that is, for one content distribution server, the bandwidth allocation ratio of the content distribution server is positively correlated with the content distribution performance of the content distribution server. The higher the bandwidth allocation ratio of the content distribution server is, the better the content distribution performance of the content distribution server is as compared with the other content distribution servers in the plurality of content distribution servers. Wherein the sum of the bandwidth allocation proportions of the plurality of content distribution servers is 1.

204. And determining the target bandwidth of each content distribution server based on the first bandwidth and the bandwidth distribution proportion of each content distribution server.

And allocating the bandwidth according to the bandwidth allocation proportion of each content distribution server, and allocating the first bandwidth to each content distribution server. The bandwidth allocated to one content distribution server is the target bandwidth of the content distribution server, and the sum of the target bandwidths of the plurality of content distribution servers is the first bandwidth. The target bandwidth of one content distribution server is the bandwidth of the content distribution server, which is consumed by the application server through scheduling the content distribution server in the first time period.

Fig. 3 is a flow diagram illustrating a method of bandwidth allocation in accordance with an example embodiment. In the embodiment of the present disclosure, taking an example that an execution subject of the bandwidth allocation method is an application server as an example, referring to fig. 3, the bandwidth allocation method includes the following steps.

301. And the application server acquires the downloading behavior information acquired in the second period.

In some embodiments, the trend of the bandwidth consumed by content distribution over time has periodicity, so that the bandwidth is predicted according to the period, that is, the bandwidth consumed by content distribution in the future period is predicted based on the bandwidth consumed by content distribution in the historical period. The second period is a historical period, and the duration of the second period may be flexibly configured, for example, the duration of the second period is one day or one week, and the like.

The download behavior information characterizes a content download during the second period. In some embodiments, the download behavior information includes a data amount of the downloaded content. Wherein the content is downloaded through a content distribution server. The content includes at least one of multimedia content such as video, audio, image, text, etc., which is not limited in the embodiments of the present disclosure. In some embodiments, the download behavior information includes, in addition to the data amount of the downloaded content, at least one of a location where a terminal that downloads the content is located, an operator to which the terminal that downloads the content belongs, and a download time of the content.

In some embodiments, the application server includes a log collection module, where the log collection module is configured to collect and record the downloading behavior information, and the application server obtains the downloading behavior information collected in the second period from the downloading behavior information recorded by the log collection module. The download behavior information recorded by the log collection module is uploaded from a terminal downloading the content from the content distribution server.

302. The application server determines second bandwidths corresponding to a plurality of second time periods in a second period based on the download behavior information acquired in the second period.

In order to realize more accurate bandwidth allocation, the second period is divided into a plurality of second time periods, and the determination and prediction of the bandwidth are carried out according to smaller time granularity. And the duration of the second time period is equal to the duration of the first time period.

And the second bandwidth corresponding to one second time period represents the bandwidth consumed by the application server for scheduling the content distribution servers to distribute the content in the second time period. And the bandwidth generally represents the data amount transmitted in the unit time, so that a second bandwidth corresponding to the second time period is determined based on the total data amount of the content downloaded in the second time period and the duration of the second time period, and accordingly, the step 302 includes: for any second time period in a plurality of second time periods in a second period, the application server determines the downloading behavior information collected in the second time period from the downloading behavior information collected in the second period; determining the sum of the data volumes of the plurality of contents included in the download behavior information collected in the second time period as the total data volume downloaded in the second time period; and determining the ratio of the total data volume to the duration of the second time period as a second bandwidth corresponding to the second time period.

For example, the sum of the data amount of the plurality of contents downloaded in a certain second time period is 15000 Megabits, the duration of the second time period is 300 seconds, and the second bandwidth corresponding to the second time period is 50 Megabits per second (Megabits per second), that is, 50 Mbps.

In some embodiments, the download behavior information includes a data amount of the downloaded content and a location of a terminal that downloads the content, and the application server counts bandwidths consumed in different area ranges according to the location of the terminal that downloads the content, and accordingly, step 302 includes: for any second time period in the second period, the application server determines a plurality of contents of which the corresponding terminal is located in any area range from the downloading behavior information collected in the second time period; and determining a second bandwidth corresponding to a second time period and the area range based on the data volume of the plurality of contents and the duration of the second time period, wherein the second bandwidth represents the bandwidth consumed by content distribution in the area range in the second time period.

For example, the download behavior information collected in the second time period includes data amounts of 300 pieces of content downloaded by a plurality of terminals, and each of 100 pieces of content is downloaded by a terminal located in the area range Q, and if the sum of the data amounts of the 100 pieces of content is 6000 mbit and the duration of the second time period is 300 seconds, the second bandwidth corresponding to the second time period and the area range Q is 20 mbit/s, that is, the bandwidth consumed for content distribution for the terminals located in the area range Q in the second time period is 20 mbit/s.

In some embodiments, the download behavior information includes a data amount of the downloaded content and an operator to which a terminal downloading the content belongs, and the application server counts bandwidths consumed by different operators according to the operator to which the terminal downloading the content belongs, and accordingly, the step 302 includes: for any second time period in the second period, the application server determines a plurality of contents of any operator to which the corresponding terminal belongs from the downloading behavior information collected in the second time period; and determining a second bandwidth corresponding to the second time period and the operator based on the data volume of the plurality of contents and the duration of the second time period, wherein the second bandwidth represents a bandwidth consumed by content distribution through a content distribution server constructed on the network of the operator in the second time period, and the content distribution server constructed on the network of the operator distributes the contents for the terminal through the network of the operator.

For example, the download behavior information collected in the second time period includes data amounts of 300 pieces of content downloaded by the plurality of terminals, and each of 100 pieces of content is downloaded from a content distribution server built on the network of the operator Y through the network of the operator Y, and if the sum of the data amounts of the 100 pieces of content is 6000 mbit and the duration of the second time period is 300 seconds, the second bandwidth corresponding to the second time period and the operator Y is 20 mbit/s, that is, the bandwidth consumed for content distribution through the content distribution server built on the network of the operator Y in the second time period is 20 mbit/s.

In some embodiments, the download behavior information includes a data amount of the downloaded content, a location of a terminal that downloads the content, and an operator to which the terminal that downloads the content belongs, and the application server counts the bandwidth consumed in the second time period according to a region and a difference between the operators. Accordingly, the step 302 includes: for any second time period in the second period, the application server determines a plurality of contents of which the corresponding terminals are located in the same area range and belong to the same operator from the downloading behavior information collected in the second time period; and determining a second time period, the area range and a second bandwidth corresponding to the operator based on the data volume of the plurality of contents and the duration of the second time period, wherein the second bandwidth represents the bandwidth consumed by content distribution for the terminals located in the area range through a content distribution server constructed on the network of the operator in the second time period.

For example, the download behavior information collected in the second time period includes data amounts of 300 contents downloaded by a plurality of terminals, and 50 contents are all downloaded by a terminal located in the area range Q through a content distribution server configured on the network of the operator Y, and if the sum of the data amounts of the 50 contents is 3000 mbit and the duration of the second time period is 300 seconds, the second bandwidth corresponding to the second time period, the area range Q and the operator Y is 10 mbit/s, that is, the bandwidth consumed for content distribution by the content distribution server configured on the network of the operator Y for the terminal located in the area range Q in the second time period is 10 mbit/s.

It should be noted that the download behavior information includes a plurality of pieces of data, where one piece of data includes content, a location, and an operator that are stored correspondingly, the location in any piece of data is a location of a terminal that downloads the content in the piece of data, the operator in any piece of data is an operator to which the terminal belongs, and a content distribution server through which the terminal downloads the content is also established on a network of the operator.

According to the technical scheme, the bandwidth consumed in the second time period is counted according to the difference of the area range and the operators, the process of determining the bandwidth is further refined, and the fineness of determining the bandwidth is improved.

In addition, in some embodiments, the application server may also perform statistics according to a bandwidth consumption curve of each content distribution server in the second period to obtain a second bandwidth corresponding to each second time period. Accordingly, the above steps 301 to 302 may be replaced by the following steps: the application server acquires a bandwidth consumption curve of each content distribution server in a second period; for any second time period and any content distribution server, extracting the bandwidth corresponding to any time point in the second time period from the bandwidth consumption curve of the content distribution server as the consumed bandwidth of the content distribution server in the second time period; and determining the sum of the bandwidths consumed by the plurality of content distribution servers in the second time period as a second bandwidth corresponding to the second time period. The bandwidth consumption curve of any content distribution server represents the trend of the consumed bandwidth of the content distribution server for content distribution of the application server along with time.

303. The application server obtains first bandwidths corresponding to a plurality of first time periods in the first period based on second bandwidths corresponding to a plurality of second time periods in the second period.

The second period is a period before the first period, the second period is a history period based on which bandwidth prediction is based, and the first period is a future period requiring bandwidth prediction. The duration of the second period is the same as the duration of the first period, the duration of the second time period in the second period is equal to the duration of the first time period in the first period, and correspondingly, the number of the plurality of second time periods included in the second period is also equal to the number of the plurality of first time periods included in the first period.

The second bandwidths corresponding to the second time periods form a time sequence, and the first bandwidths corresponding to the first time periods also form a time sequence. In some embodiments, the application server predicts, by a time series prediction method, a first bandwidth corresponding to a plurality of first time periods included in the first cycle based on a second bandwidth corresponding to a plurality of second time periods.

For example, the application server inputs the second bandwidths corresponding to the plurality of second time periods into the time series prediction model to obtain the first bandwidths corresponding to the plurality of first time periods output by the time series prediction model.

The time series prediction model is obtained by performing model training by using bandwidths corresponding to a plurality of time periods included in a historical cycle as training samples. The training process of the time series prediction model comprises the following steps: inputting bandwidths corresponding to a plurality of time periods included in the first historical period into a time series prediction model to obtain the bandwidths corresponding to the plurality of time periods included in the prediction period and output by the time series prediction model; comparing bandwidths corresponding to a plurality of time periods included in the prediction period with bandwidths corresponding to a plurality of time periods included in the second history period, and updating model parameters of the time series prediction model based on a comparison result; if the updated time series prediction model meets the training stopping condition, finishing the training of the time series prediction model, and outputting the trained time series prediction model; and if the updated time series prediction model does not meet the training stopping condition, continuing the next iteration process until the training stopping condition is met. Wherein the second history period is the next period after the first history period.

In some embodiments, the training stopping condition is that, in the process of multiple consecutive iterations, the difference values corresponding to multiple time periods included in the prediction period are all smaller than a threshold value. The threshold value may be flexibly configured, for example, the threshold value is 10, 20, or 30, which is not limited in this disclosure. In addition, the plurality of time periods included in the prediction period respectively correspond to the plurality of time periods included in the second history period in a one-to-one correspondence manner, and the difference value corresponding to one time period included in the prediction period is a difference value between the bandwidth corresponding to the time period and belonging to the second history period.

For example, the prediction cycle includes one of the time periods from 18 hours 00 to 18 hours 05, where the time period from 18 hours 00 to 18 hours 05 corresponds to 18 hours 00 to 18 hours 05 in the second cycle. The difference corresponding to the time period from 18 hours 00 to 18 hours 05 in the prediction period refers to the difference between the bandwidth corresponding to 18 hours 00 to 18 hours 05 in the prediction period and the bandwidth corresponding to 18 hours 00 to 18 hours 05 in the second history period.

It should be noted that, in the above embodiment, the application server obtains, based on the second bandwidths corresponding to the plurality of second time periods included in one second cycle, the first bandwidths corresponding to the plurality of first time periods included in the first cycle. In some embodiments, the application server may also predict the time series corresponding to the first period with reference to a plurality of time series corresponding to the second period, so as to improve the accuracy of bandwidth prediction by referring to a greater number of historical time series.

According to the technical scheme provided by the embodiment of the disclosure, the bandwidth consumption condition in the historical period is determined by analyzing the historical content downloading condition, so that the bandwidth consumption condition in the future period is reasonably predicted, the accuracy of bandwidth prediction is improved, and further more accurate bandwidth allocation can be realized based on the predicted bandwidth with higher accuracy, and the availability of bandwidth allocation is further improved.

The above embodiment is described by taking as an example that the second bandwidth represents the total bandwidth consumed in the second time period, and the first bandwidth represents the total bandwidth required to be consumed in the first time period. In some embodiments, the second bandwidth is a bandwidth corresponding to a second time period and a certain area range, and the second bandwidth represents a bandwidth consumed for content distribution for a terminal located in the certain area range in the second time period, and accordingly, for any area range and any second time period, the application server performs prediction based on the area range and the second bandwidth corresponding to the second time period to obtain a first bandwidth, which represents a bandwidth required to be consumed for content distribution for the terminal located in the area range in a first time period corresponding to the second time period in the first cycle.

For example, if the second bandwidth is a bandwidth consumed for content distribution for the terminal located in the area range Q during the second period of time 18 hours 00 to 18 hours 05 of the second cycle, the application server predicts a first bandwidth, which is a bandwidth required to be consumed for content distribution for the terminal located in the area range Q during the first period of time 18 hours 00 to 18 hours 05 of the first cycle, based on the second bandwidth.

In some embodiments, the second bandwidth is a second time period and a bandwidth corresponding to a certain operator, the second bandwidth represents a bandwidth consumed for content distribution by a content distribution server established on a network of the certain operator in the second time period, and accordingly, for any operator and any second time period, the application server performs prediction based on the second bandwidth corresponding to the operator and the second time period to obtain a first bandwidth, and the first bandwidth represents a bandwidth consumed for content distribution by the content distribution server established on the network of the operator in a first time period corresponding to the second time period in the first period.

For example, if the second bandwidth is a bandwidth consumed for content distribution by the content distribution server established on the network of the carrier Y during the second period of time 18 hours 00 to 18 hours 05 of the second cycle, the application server predicts, based on the second bandwidth, a first bandwidth that is a bandwidth required to be consumed for content distribution by the content distribution server established on the network of the carrier Y during the first period of time 18 hours 00 to 18 hours 05 of the first cycle.

In some embodiments, the second bandwidth is a second time period, a certain area range and a bandwidth corresponding to a certain operator, the second bandwidth represents a bandwidth consumed for content distribution for a terminal located in the area range by a content distribution server built on a network of the operator in the second time period, and accordingly, for any operator, any second time period and any area range, the application server predicts a first bandwidth based on a second bandwidth corresponding to the operator, the second time period and the area range, and the first bandwidth represents a bandwidth consumed for content distribution for a terminal located in the area range by a content distribution server built on a network of the operator in a first time period corresponding to the second time period in the first period.

For example, if the second bandwidth is a bandwidth consumed for content distribution by the content distribution server built on the network of the carrier Y for the terminals located in the area range Q in the second period of 18 hours 00 to 18 hours 05 of the second cycle, the application server predicts a first bandwidth, which is a bandwidth required to be consumed for content distribution by the content distribution server built on the network of the carrier Y for the terminals located in the area range Q in the first period of 18 hours 00 to 18 hours 05 of the first cycle, based on the second bandwidth.

304. The application server obtains first performance parameter information of each content distribution server, wherein the first performance parameter information comprises multiple performance parameters belonging to different dimensions.

The multiple performance parameters respectively represent the content distribution performance of the content distribution server from different dimensions. In some embodiments, the plurality of performance parameters includes at least two of bandwidth capacity, download failure rate, katton rate, download speed, and download error rate. For any content distribution server, the bandwidth capacity refers to the upper limit of bandwidth consumption of the content distribution server, that is, the maximum bandwidth of the content distribution server that can be consumed; the download failure rate refers to the probability of the download failure phenomenon generated when the content distribution server is scheduled to download the content; the pause rate is the probability of content playing pause phenomenon when the content distribution server is scheduled to download and play the content under the condition that the content is audio or video; the download error rate is a probability of a download error phenomenon occurring when the content distribution server is scheduled to download the content.

In some embodiments, the download failure rate of any one content distribution server is a ratio of the total number of times that content has failed to be downloaded by that content distribution server to the total number of times that content has been downloaded by that content distribution server. The jam-on rate of any one content distribution server is the ratio of the total number of times of jam-on occurring when a terminal downloading content through the content distribution server plays the downloaded content to the total number of times of downloading and playing the content through the content distribution server. The download error rate of any one of the content distribution servers is a ratio of the number of times of errors occurring in downloading the content by the content distribution server to the total number of times of downloading the content by the content distribution server. In some embodiments, the download failure rate, the pause rate, and the download error rate are counted over a time period of one week of the history or one day of the history. For example, the download failure rate of any content distribution server is the ratio of the total number of times of content download failure by the content distribution server in the previous week of the current time to the total number of times of content download by the content distribution server; alternatively, the download failure rate is a ratio of the total number of times of failed downloading of the content by the content distribution server on the day before the current time to the total number of times of downloading the content by the content distribution server, which is not limited in the embodiment of the present disclosure.

In some embodiments, the application server includes a domain name management module, the domain name management module has a function of recording first performance parameter information of a plurality of content distribution servers, and the application server obtains the first performance parameter information of each content distribution server from the domain name management module. In some embodiments, the first performance parameter information of any one of the content distribution servers is determined based on download feedback information uploaded by a plurality of terminals that download content through the content distribution server. Wherein, a piece of download feedback information represents an actual download situation of a piece of content, for example, the download feedback information includes at least one of download success, download failure, download error and occurrence of pause.

305. The application server fuses the multiple performance parameters of each content distribution server to obtain the fusion performance parameters of each content distribution server.

For any content distribution server, the server fuses multiple performance parameters of the content distribution server to obtain a fusion performance parameter, and comprehensively expresses the content distribution performance of the content distribution server through the fusion performance parameter.

In some embodiments, the application server respectively gives a certain weight to the multiple performance parameters, and performs weighted fusion on the multiple performance parameters to obtain a fused performance parameter, that is, the foregoing step 305 includes: the application server acquires various performance parameters of each content distribution server and weights corresponding to the various performance parameters; and performing weighted fusion on the multiple performance parameters according to the weights corresponding to the multiple performance parameters to obtain the fusion performance parameters of each content distribution server. That is, for any content distribution server, the application server performs weighted fusion on the multiple performance parameters of the content distribution server according to the weights corresponding to the multiple performance parameters of the content distribution server to obtain the fusion performance parameters.

One performance parameter corresponds to a weight, and the weight represents the importance of the corresponding performance parameter to the content distribution performance of the content distribution server, which can also be understood as that the weight corresponding to one performance parameter represents the attention degree of the performance parameter in the process of evaluating the content distribution performance of the content distribution server. The larger the weight corresponding to a certain performance parameter is, the higher the importance of the performance parameter to the content distribution performance is, and the higher the performance parameter is, the higher the content distribution performance of the content distribution server is. For example, the short video distribution scenario has a high requirement on the stuck rate, that is, in the process of evaluating the content distribution performance of the content distribution server, the attention degree on the performance parameter of the stuck rate is high, and the performance parameter of the stuck rate corresponds to a high weight. It should be noted that the sum of the weights corresponding to the multiple performance parameters is 1, and the weight corresponding to each performance parameter can be flexibly configured, which is not limited in the embodiment of the present disclosure.

In some embodiments, for any content distribution server in a plurality of content distribution servers, the application server obtains a plurality of performance parameters of the content distribution server; and determining the average value of the multiple performance parameters as the fusion performance parameter of the content distribution server.

It should be noted that, the application server performs weighted fusion on the multiple performance parameters according to weights corresponding to the multiple performance parameters, and performs normalization processing on the multiple performance parameters before obtaining the fusion performance parameters of the content distribution server; and then, according to the weights corresponding to the multiple performance parameters, performing weighted fusion on the multiple performance parameters after the normalization processing to obtain the fusion performance parameters of the content distribution server.

The normalization processing is to convert the performance parameter into a value between 0 and 1, and to express the content distribution performance of the content distribution server by a value between 0 and 1. In some embodiments, for any one of the multiple performance parameters, the application server performs normalization processing based on multiple performance parameters of the multiple content distribution servers, so as to obtain multiple performance parameters after normalization processing. In the embodiment of the present disclosure, the larger the performance parameter after normalization processing is, the better the content distribution performance is; the larger the fusion performance parameter is, the better the content distribution performance is.

In some embodiments, the process of normalizing any one of the performance parameters of the plurality of content distribution servers by the application server comprises: the application server sequences a plurality of performance parameters of a plurality of content distribution servers, and determines the maximum performance parameter and the minimum performance parameter in the performance parameters; for any one of the plurality of performance parameters, the application server determines a first difference between a maximum performance parameter and the performance parameter; determining a second difference between the maximum performance parameter and the minimum performance parameter; and determining the ratio of the first difference to the second difference as the performance parameter after the normalization processing.

For example, one performance parameter is the stuck rate, and the performance parameters of the content distribution servers are 5%, 4%, 3%, and 2%, respectively, where the maximum performance parameter is 5%, the minimum performance parameter is 2%, and for the performance parameter of 5%, the performance parameter after the normalization processing is (5% -5%)/(5% -2%) -0; for performance parameter 4%, the performance parameter after normalization was (5% -4%)/(5% -2%) -1/3; for performance parameter 3%, the performance parameter after normalization is (5% -3%)/(5% -2%) -2/3; for a performance parameter of 2%, the performance parameter after normalization is (5% -2%)/(5% -2%) -1.

306. The application server determines the sum of the fusion performance parameters of the content distribution servers as a total performance parameter, and determines the proportion between the fusion performance parameters of the content distribution servers and the total performance parameter as the bandwidth allocation proportion of the content distribution servers.

After the fusion performance parameters of the plurality of content distribution servers are obtained, the proportion of the fusion performance parameter of any content distribution server in the fusion performance parameters of the plurality of content distribution servers is determined as the bandwidth allocation proportion of the content distribution server.

For example, the plurality of content distribution servers include a content distribution server a, a content distribution server B, and a content distribution server C, where the fusion performance parameter of the content distribution server a is 0.7, the fusion performance parameter of the content distribution server B is 0.6, and the fusion performance parameter of the content distribution server C is 0.8, and then the sum of the fusion performance parameters of the content distribution servers, that is, the total performance parameter is 0.6+0.7+0.8 — 2.1, where the bandwidth allocation example of the content distribution server a is 0.7/2.1 — 1/3; the bandwidth allocation ratio of the content distribution server a is 0.6/2.1-2/7; the bandwidth allocation ratio of the content distribution server a is 0.8/2.1-8/21.

According to the technical scheme provided by the embodiment of the disclosure, the multiple performance parameters of the content distribution server are fused to obtain the fusion performance parameters, so that the content distribution performance of the content distribution server is comprehensively represented, and the bandwidth distribution proportion is determined according to the fusion performance parameters, so that the bandwidth distribution proportion is matched with the comprehensive content distribution performance of the content distribution server, and the accuracy of determining the bandwidth distribution proportion is improved.

It should be noted that, in the above embodiments, the order of obtaining the first bandwidth and determining the bandwidth allocation ratio is described by taking the first bandwidth obtaining first and then determining the bandwidth allocation ratio as an example, and in some embodiments, the step of obtaining the first bandwidth and the step of determining the bandwidth allocation ratio may also be performed according to other timings. For example, the first bandwidth is obtained and the bandwidth allocation ratio is determined at the same time, that is, step 301 to step 303 and step 304 to step 306 are performed at the same time; alternatively, the bandwidth allocation ratio is determined first, and then the first bandwidth is acquired, that is, step 304 to step 306 are performed first, and then step 301 to step 303 are performed, which is not limited in the embodiment of the present disclosure.

307. The application server determines the bandwidth allocation proportion of each content distribution server as a first target bandwidth of each content distribution server by multiplying the first bandwidth by the bandwidth allocation proportion of each content distribution server.

After the first bandwidths corresponding to the plurality of first time periods and the bandwidth allocation proportions of the content distribution servers are obtained, for any first time period and any content distribution server, the application server determines the first target bandwidth of the content distribution server by multiplying the bandwidth allocation proportion of the content distribution server by the first bandwidth corresponding to the first time period.

For example, the first bandwidth corresponding to a certain first time period is 100 mbits/sec, and the plurality of content distribution servers include a content distribution server a, a content distribution server B, and a content distribution server C, where the bandwidth allocation ratio of the content distribution server a is 0.3, the bandwidth allocation ratio of the content distribution server B is 0.4, the bandwidth allocation ratio of the content distribution server B is 0.3, and accordingly, the first target bandwidth of the content distribution server a is 0.3 × 100 ═ 30, the first target bandwidth of the content distribution server B is 0.4 × 100 ═ 30, and the first target bandwidth of the content distribution server C is 0.3 × 100 ═ 30.

308. The application server adjusts the first target bandwidth of each content distribution server to obtain the second target bandwidth of each content distribution server.

In some embodiments, after determining the first target bandwidth of each content distribution server, the application server further adjusts the determined first target bandwidth to reduce the amount of the transferred resources required for scheduling the plurality of content distribution servers, that is, to reduce the cost consumed for scheduling the content distribution servers. Accordingly, the step 308 includes the following steps 3081 to 3084.

3081. The application server determines a plurality of bandwidth adjustment policies.

The various bandwidth adjustment strategies include adjustment bandwidths of the content distribution servers in a plurality of first time periods in the same period. For example, the plurality of content distribution servers include a content distribution server a and a content distribution server B, the plurality of first time periods in the same period include a first time period T1 and a first time period T2, and one bandwidth adjustment policy includes an adjusted bandwidth of the content distribution server a in the first time period T1, an adjusted bandwidth of the content distribution server a in the first time period T2, an adjusted bandwidth of the content distribution server B in the first time period T1, and an adjusted bandwidth of the content distribution server B in the first time period T2. The various bandwidth adjustment strategies differ in the individual adjustment bandwidths.

Wherein the adjustment bandwidth is positive, negative or 0. If the adjustment bandwidth is a positive number, the absolute value of the adjustment bandwidth is increased on the basis of the first target bandwidth; if the adjustment bandwidth is negative, the absolute value of the adjustment bandwidth is deducted on the basis of the first target bandwidth. It should be noted that, in each bandwidth adjustment policy, the sum of the adjusted bandwidths in the same first time period is 0, that is, in one first time period, the sum of the adjusted bandwidths of the content servers is 0, so as to ensure that the sum of the bandwidths of the content distribution servers is still the first bandwidth after adjustment.

In some embodiments, the application server further limits the adjusted bandwidth of each content distribution server within a limited range, so as to ensure that the bandwidth adjusted by the content distribution server is still positively correlated with the content distribution performance of the content distribution server, and ensure the content distribution performance of the scheduled content distribution server. The value range of the adjustment bandwidth can be flexibly configured, for example, the value range of the adjustment bandwidth is [ -10,10], which means that the adjustment bandwidth takes a value between-10 and 10; alternatively, the adjustment bandwidth may be in a range of [ -15,15], [ -20,20], and the like, which is not limited in the embodiment of the present disclosure.

In some embodiments, the application server further determines an adjusted bandwidth for each content distribution server with reference to the bandwidth capacity of each content distribution server. For any content distribution server, the sum of the adjusted bandwidth of the content distribution server and the first target bandwidth is less than or equal to the bandwidth capacity of the content distribution server, so as to avoid the adjusted bandwidth from exceeding the maximum bandwidth which can be provided by the content distribution server, and reduce the bandwidth overrun or extra cost consumed by additionally scheduling other content distribution servers on the basis of ensuring the content distribution performance of the content distribution server.

3082. And the application server respectively adjusts the first target bandwidth of each content distribution server in each first time period according to a plurality of bandwidth adjustment strategies to obtain an adjusted third target bandwidth.

For any one of the multiple bandwidth adjustment policies, the bandwidth adjustment policy includes adjustment bandwidths of the multiple content distribution servers in multiple first time periods, and correspondingly, the application server adjusts the first target bandwidth of the content distribution server in any first time period according to the adjustment bandwidth of the content distribution server in any first time period, so as to obtain an adjusted third target bandwidth.

For example, the plurality of content distribution servers include a content distribution server a and a content distribution server B, the plurality of first time periods in the same period include a first time period T1 and a first time period T2, and one bandwidth adjustment policy includes an adjusted bandwidth of the content distribution server a in the first time period T1, an adjusted bandwidth of the content distribution server a in the first time period T2, an adjusted bandwidth of the content distribution server B in the first time period T1, and an adjusted bandwidth of the content distribution server B in the first time period T2, so that the application server adjusts the first target bandwidth of the content distribution server a in the first time period T1 according to the adjusted bandwidth of the content distribution server a in the first time period T1, and obtains a third target bandwidth of the content distribution server a in the first time period T1; and according to the adjusted bandwidth of the content distribution server A in the first time period T2, adjusting the first target bandwidth of the content distribution server A in the first time period T2 to obtain a third target bandwidth of the content distribution server A in the first time period T2, and so on.

3083. And the application server determines the target quantity based on the third target bandwidth of each content distribution server in a plurality of first time periods under various bandwidth adjustment strategies.

And the target quantity represents the total quantity of resources which need to be transferred out when each content distribution server is scheduled by a third target bandwidth of each content distribution server in each first time period in a plurality of first time periods. In the embodiment of the present disclosure, the target number is also used to represent a cost required to schedule the plurality of content distribution servers with the third target bandwidths of the plurality of content distribution servers in each first time period in the first period.

In some embodiments, the application server determines, based on the third target bandwidths of the content distribution servers in the plurality of first time periods, the first number of the target numbers in a calculation manner of: for any content distribution server in the plurality of content distribution servers, sorting third target bandwidths of the content distribution servers in a plurality of first time periods in descending order, and removing the third target bandwidth which is sorted in the top 5% from the third target bandwidths of the content distribution servers in the plurality of first time periods; determining a maximum third target bandwidth among the third target bandwidths of the plurality of first time periods from the remaining content distribution servers; determining a first target number corresponding to the maximum third target bandwidth; and determining the sum of the first target quantity corresponding to the plurality of content distribution servers as the target quantity. The first target quantity refers to the quantity of resources which need to be transferred out by the content distribution server and are scheduled in the first period.

In some embodiments, the application server determines the second number calculation manner of the target number based on the third target bandwidth of each content distribution server in the plurality of first time periods as follows: for any content distribution server in the plurality of content distribution servers, sorting third target bandwidths of the content distribution servers in a plurality of first time periods in descending order, and determining the largest third target bandwidth from the third target bandwidths of the content distribution servers in the plurality of first time periods; determining a second target number corresponding to the maximum third target bandwidth; and determining the sum of the second target number corresponding to the plurality of content distribution servers as the target number.

3084. And the application server determines the third target bandwidth of each content distribution server in each first time period as the second target bandwidth of each content distribution server in each first time period in the bandwidth adjustment strategy with the minimum target number.

The application server determines a bandwidth adjustment strategy with the minimum target number from the plurality of bandwidth adjustment strategies, and determines a third target bandwidth of each content distribution server in each first time period in the bandwidth adjustment strategy with the minimum target number as a second target bandwidth of each content distribution server in each first time period.

For example, the bandwidth adjustment policy with the smallest target number includes the third target bandwidth D1 of the content distribution server a in the first time period T1, the third target bandwidth D2 of the content distribution server a in the first time period T2, the third target bandwidth D3 of the content distribution server B in the first time period T1, and the third target bandwidth D4 of the content distribution server B in the first time period T2, the application server determines D1 as the second target bandwidth of the content distribution server a in the first time period T1; determining D2 as a second target bandwidth for the content distribution server a for the first time period T2; determining D3 as a second target bandwidth for the content distribution server B for the first time period T1; d4 is determined as a second target bandwidth for the content distribution server B for the first time period T2.

According to the technical scheme provided by the embodiment of the disclosure, after the initial first target bandwidth of each content distribution server is determined based on the bandwidth allocation proportion of each content distribution server, the first target bandwidth is also adjusted, and the second target bandwidth which minimizes the total amount of resources required to be transferred out by a plurality of content distribution servers in the first period is determined, so that the cost required to be consumed by scheduling the plurality of content distribution servers in the first period is reduced, and under the condition that the total amount of the transferred-out resources is minimized, the scheduling of the content distribution servers can meet the bandwidth requirement, and the effective utilization rate of the bandwidth is improved.

In some embodiments, after the application server determines the second target bandwidth of each content distribution server in each first time period, the application server further outputs the scheduling policy based on the second target bandwidth of each content distribution server in each first time period. The scheduling strategy comprises a scheduling sequence, a scheduling time length, a scheduling interval and a scheduling bandwidth. The scheduling sequence represents the scheduling sequence of the content distribution servers in a first period; the scheduling duration of any content delivery server represents the scheduled continuous duration of the content delivery server; the scheduling interval of any two content distribution servers represents the difference of the starting scheduling time of the two content distribution servers; the scheduling bandwidth of any content distribution server is the second target bandwidth corresponding to the content distribution server.

For example, the plurality of first time periods are T1 and T2 in order, the second target bandwidth of the content distribution server a in the first time period T1 is D1, the second target bandwidth of the content distribution server a in the first time period T2 is D2, the second target bandwidth of the content distribution server B in the first time period T2 is D3, and none of D1, D2, and D3 is 0, so that the content distribution server a is scheduled first in the first time period T1, the content distribution server B is scheduled only in the first time period T2, and the scheduling order is the content distribution server a and the content distribution server B; the scheduling interval between the content distribution server a and the content distribution server B is the duration of the first time period T1; the scheduling duration of the content distribution server a is the sum of the duration of the first time period T1 and the duration of the second time period T2; the scheduling duration of the content distribution server B is the duration of the second time period T2; the scheduled bandwidth of the content distribution server a in the first time period T1 is D1, the scheduled bandwidth in the second time period T2 is D2, and the scheduled bandwidth of the content distribution server B in the second time period T2 is D3.

In some embodiments, multiple quantity calculation modes exist among the multiple content distribution servers, the processes for determining the target quantity are different for the content distribution servers with different quantity calculation modes, in order to make the process for determining the target quantity corresponding to the multiple content distribution servers clearer and improve the adjustment efficiency of the first target bandwidth, the application server divides the content distribution servers with the same quantity calculation mode into a group, wherein the quantity calculation mode represents a mode for calculating the quantity of resources to be transferred according to the target bandwidth of the content distribution server; the application server adjusts the first target bandwidth of each content distribution server in each group to obtain the second target bandwidth of each content distribution server. The bandwidth adjustment modes between different groups may be the same or different. For any group, the bandwidth adjustment modes of the content distribution servers in the group are different, for example, the first target bandwidth of part of the content distribution servers in the group is increased, and the first target bandwidth of other content distribution servers in the group is decreased, so that the sum of the second target bandwidths obtained after the content distribution servers in the group are adjusted is not changed.

In some embodiments, different content distribution servers correspond to different numbers of units, the number of units of any content distribution server characterizing the amount of destaged resources required to consume the bandwidth of that content distribution server over a period of time. For any group, the application server divides the content distribution servers in the group into a first group and a second group according to the unit number corresponding to the content distribution servers in the group, wherein the number of the content distribution servers in the first group is equal to that of the content distribution servers in the second group, and the unit number corresponding to the content distribution servers in the first group is smaller than that of the content distribution servers in the second group; the application server determines an adjustment amplitude, wherein the adjustment amplitude is a positive number; the first target bandwidth of the content distribution servers in the first group is increased and the first target bandwidth of the content distribution servers in the second group is decreased according to the adjustment magnitude. Or the application server randomly generates a plurality of adjustment amplitudes, wherein any adjustment amplitude is a positive number or a negative number, the sum of the adjustment amplitudes is 0, the number of the adjustment amplitudes is equal to the number of the content distribution servers in the group, and the application server sorts the content distribution servers according to the order of the unit number corresponding to the content distribution servers from small to large and sorts the content distribution servers according to the order of the adjustment amplitudes from large to small; and adjusting the first target bandwidth of the content distribution server with the minimum unit quantity according to the maximum adjustment amplitude, adjusting the first target bandwidth of the content distribution server with the second minimum unit quantity according to the second maximum adjustment amplitude, and so on. Therefore, on the basis of considering the content distribution performance of the content distribution servers, the bandwidth of the content distribution servers in any group is further adjusted based on the amount of the resources needing to be transferred out by scheduling the content distribution servers, and the amount of the resources needing to be transferred out by scheduling each content distribution server is reduced.

In some embodiments, the process of adjusting, by the application server, the first target bandwidth of each content distribution server in each group to obtain the second target bandwidth of each content distribution server is similar to that in steps 3081 to 3084, and includes the following steps: for any group, the application server determines a plurality of bandwidth adjustment strategies, wherein each bandwidth adjustment strategy comprises the adjustment bandwidth of each content distribution server in the group in a plurality of first time periods in the same period; according to a plurality of bandwidth adjustment strategies, respectively adjusting the first target bandwidth of each content distribution server in the group in each first time period to obtain an adjusted third target bandwidth, wherein the sum of the third target bandwidth of each content distribution server in the group in each first time period is equal to the sum of the first target bandwidth of each content distribution server in the group in each first time period; under various bandwidth adjustment strategies, determining a target quantity based on a third target bandwidth of each content distribution server in the group in a plurality of first time periods, wherein the target quantity represents the total quantity of resources which need to be transferred out when each content distribution server is scheduled by the third target bandwidth of each content distribution server in the group in each first time period in the plurality of first time periods; and respectively determining the third target bandwidth of each content distribution server in the group in each first time period as the second target bandwidth of each content distribution server in the group in the bandwidth adjustment strategy with the minimum target number in each first time period.

In the above embodiment, the application server is exemplified to adjust the first target bandwidth of each first time slot of each content distribution server in the first period. In some embodiments, the above-mentioned process of adjusting the first target bandwidth to obtain the second target bandwidth may also be applied only in a peak period of the first period, that is, the application server adjusts the first target bandwidth of each content distribution server to obtain the second target bandwidth of each content distribution server in response to the first time period belonging to a time period within the peak period. The peak period refers to a time period with a large bandwidth demand, and the peak period can be flexibly configured, for example, the peak period is from 11 hours to 14 hours per day or from 19 hours to 22 hours per day, which is not limited by the embodiment of the present disclosure. In some embodiments, the bandwidth consumed for content distribution during each time period in the peak period is greater than the bandwidth threshold, e.g., the bandwidth consumed for content distribution during each time period in the peak period is greater than 100 megabits per second. The process of adjusting the first target bandwidth of each content distribution server by the application server to obtain the second target bandwidth of each content distribution server is the same as that in step 308, and is not described herein again. And for any first time period which does not belong to the peak period, the application server determines the target bandwidth of each content distribution server as the product of the first bandwidth corresponding to the first time period and the bandwidth allocation proportion of each content distribution server.

In some embodiments, the application server may also determine the target bandwidth of each content distribution server by directly multiplying the bandwidth allocation proportion of each content distribution server by the first bandwidth without adjusting the first target bandwidth determined based on the bandwidth allocation proportion, that is, for any first time period, the application server determines the target bandwidth of each content distribution server by multiplying the bandwidth allocation proportion of each content distribution server by the first bandwidth corresponding to the first time period.

In some embodiments, the first bandwidth represents a bandwidth that is required to be consumed for content distribution for a terminal located in a certain area range in a first time period, the application server should schedule a content distribution server whose service range includes the area range to distribute content for the terminal located in the area range, and accordingly, for a first bandwidth corresponding to any area range and any first time period, the application server determines a target bandwidth of each content distribution server whose service range includes the area range based on the first bandwidth and a bandwidth allocation ratio of each content distribution server. The step 307 to step 308 are the same as the above step, and the details are not repeated here, where the application server determines, based on the first bandwidth and the bandwidth allocation ratio of each content distribution server, the target bandwidth of each content distribution server whose service range includes the area range.

In some embodiments, the first bandwidth represents a bandwidth consumed by content distribution servers built on a network of an operator during a first time period, and the application server should schedule the content distribution servers built on the network of the operator to meet the bandwidth requirement during the first time period. The process of the application server determining the target bandwidth of each content distribution server established on the network of the operator based on the first bandwidth and the bandwidth allocation ratio of each content distribution server is the same as the above steps 307 to 308, and is not described herein again.

In some embodiments, the first bandwidth represents a bandwidth consumed by a content distribution server built on a network of an operator for content distribution for a terminal located in a certain area range in a first time period, the application server should schedule the content distribution server built on the network of the operator and having a service range including the area range to meet a bandwidth requirement in the first time period, and accordingly, for a first bandwidth corresponding to any operator, any area range and any first time period, the application server determines a target bandwidth of each content distribution server built on the network of the operator and having a service range including the area range based on the first bandwidth and a bandwidth allocation ratio of each content distribution server. The process of the application server determining the target bandwidth of each content distribution server that is established on the network of the operator and whose service range includes the area range based on the first bandwidth and the bandwidth allocation ratio of each content distribution server is the same as the process from step 307 to step 308, and is not described herein again.

It should be noted that the above-mentioned embodiment is described as an example in which the process of determining the target bandwidth of each content distribution server is performed by an application server, and in some embodiments, the process of determining the target bandwidth of each content distribution server may be performed by a scheduling server, and after determining the target bandwidth of each content distribution server through the steps similar to the steps 301 to 308, the scheduling server transmits the target bandwidth of each content distribution server to the application server, and the application server schedules each content distribution server according to the target bandwidth of each content distribution server.

309. For any first time period in the first period, the application server schedules each content distribution server according to the target bandwidth of each content distribution server in the first time period.

When the first time period is reached, in the case that the bandwidth of the content distribution server consumed by the application server in the first time period does not exceed the target bandwidth of the content distribution server for any content distribution server, in response to receiving a content acquisition request sent by a terminal, sending address information of the content distribution server to the terminal to schedule the content distribution server by causing the terminal to send the content acquisition request to the content distribution server; the application server does not transmit the address information of the content distribution server to the terminal to schedule the content distribution server in response to receiving the content acquisition request transmitted by the terminal in the case where the bandwidth of the content distribution server consumed during the first period exceeds the target bandwidth of the content distribution server. The content distribution server responds to the received content acquisition request sent by the terminal, consumes the bandwidth of the content distribution server, and sends the content requested to be acquired by the content acquisition request to the terminal.

In some embodiments, the application server performs step 309 after performing step 307 and step 308, where the target bandwidth of the content distribution server is a second target bandwidth of the content distribution server, and the application server schedules the content distribution server according to the second target bandwidth of the content distribution server.

In some embodiments, the above steps 307 to 308 can be replaced by the following steps: the application server determines the product of the bandwidth allocation proportion of each content distribution server and the first bandwidth as the target bandwidth of each content distribution server; the application server schedules each content distribution server according to the target bandwidth of each content distribution server.

310. The application server acquires second performance parameter information of each content distribution server in a first time period.

In some embodiments, the application server obtains, at the end of the first time period, second performance parameter information of each content distribution server in the first time period, where the second performance parameter information of each content distribution server characterizes content distribution performance of each content distribution server in the first time period.

In some embodiments, the second performance parameter information of any one of the content distribution servers includes a plurality of performance parameters including at least two of a download failure rate, a card-ton rate, a download speed, and a download error rate. The download failure rate of the content distribution server is a ratio of the total number of times of content download failure through the content distribution server in the first time period to the total number of times of content download through the content distribution server in the first time period. The card pause rate of the content distribution server is the ratio of the total times of the terminal downloading the content through the content distribution server in the first time period to the total times of downloading and playing the content through the content distribution server in the first time period when playing the downloaded content. The download error rate of the content distribution server is the ratio of the number of times of errors generated by downloading the content through the content distribution server in the first time period to the total number of times of downloading the content through the content distribution server in the first time period.

311. And the application server subtracts the target bandwidth of any content distribution server in the next first time period of the first time period from the target adjustment bandwidth in response to the second performance parameter information of the content distribution server in the first time period not meeting the performance condition.

Wherein, the performance condition refers to a preset performance limiting condition. In some embodiments, the performance condition includes at least one of a download failure rate less than a failure rate threshold, a download error rate less than an error rate threshold, a stuck rate less than a stuck rate threshold, and a download speed greater than a minimum speed threshold. The failure rate threshold, the error rate threshold, the stuck rate threshold, and the minimum speed threshold may be flexibly configured, for example, the failure rate threshold is 1% or 5%, the error rate threshold is 3% or 6%, and the stuck rate threshold is 4% or 7%. The minimum speed threshold is 10 kbits per second or 20 kbits per second, which is not limited by the embodiments of the present disclosure.

In some embodiments, the application server further fuses the multiple performance parameters included in the second performance parameter information to obtain a fused performance parameter. The application server fuses the plurality of performance parameters included in the second performance parameter information to obtain a fused performance parameter, and fuses the plurality of performance parameters included in the first performance parameter information in step 305 to obtain a fused performance parameter, which is similar to the process of obtaining the fused performance parameter, and accordingly, the performance condition is that the fused performance parameter is greater than the parameter threshold. The parameter threshold may be flexibly configured, for example, the parameter threshold is 0.8 or 0.9, and the like, which is not limited in this disclosure. The target adjustment bandwidth is a positive number, and the target adjustment bandwidth can be flexibly configured, for example, the target adjustment bandwidth is 10, 20, or 30, and the like, which is not limited by the embodiment of the disclosure.

In some embodiments, the application server further increases the target bandwidth of any content distribution server meeting the performance condition in the next first time period of the first time period by the target adjustment bandwidth on the basis of deducting the target bandwidth of the content distribution server in the next first time period of the first time period, so as to meet the bandwidth requirement in the first time period on the basis of ensuring the content distribution performance of the scheduled content distribution server.

In some embodiments, if the application server further determines the fusion performance parameter of each content distribution server, and the performance condition is that the fusion performance parameter is greater than the parameter threshold, the application server increases the target bandwidth of the content distribution server with the largest fusion performance parameter in the remaining content distribution servers in the next first time period of the first time period by the target adjustment bandwidth. If the target bandwidth of any content distribution server exceeds the bandwidth capacity of the content distribution server after the target bandwidth is increased by the target adjustment bandwidth, the bandwidth exceeding the bandwidth capacity is increased on the target bandwidth of other content distribution servers meeting the performance condition in the next first time period of the first time period.

In the above embodiment, the application server determines the bandwidth adjustment policy with the minimum total amount of resources that need to be transferred out by scheduling each content distribution server in the first period by adjusting the first target bandwidth of each content distribution server in each first time period in the first period. In addition, in some embodiments, the application server may also determine a bandwidth adjustment policy that minimizes the total amount of resources that need to be transferred out by each content distribution server during the first time period for a first target bandwidth adjustment of each content distribution server during the first time period. Correspondingly, for a first target bandwidth of a plurality of content distribution servers in any first time period, the step of adjusting the first target bandwidth of each content distribution server by the application server to obtain a second target bandwidth of each content distribution server includes: the application server determines a plurality of bandwidth adjustment strategies, wherein each bandwidth adjustment strategy comprises the adjustment bandwidth of each content distribution server; respectively adjusting the first target bandwidth of each content distribution server according to a plurality of bandwidth adjustment strategies to obtain an adjusted third target bandwidth, wherein the sum of the third target bandwidth of each content distribution server is the first bandwidth; under various bandwidth adjustment strategies, determining a target quantity based on a third target bandwidth of each content distribution server, wherein the target quantity represents the total quantity of resources to be transferred out when each content distribution server is scheduled by the third target bandwidth of each content distribution server; and determining the third target bandwidth of each content distribution server in the bandwidth adjustment strategy with the minimum target number as the second target bandwidth of each content distribution server.

In order to make the process of bandwidth allocation clearer, the following description is made with reference to fig. 4, on one hand, downloading behavior information is collected through step 301, and then bandwidth is predicted through steps 302 to 303; on the other hand, the first performance parameter information is collected through step 304, and the bandwidth allocation proportion is determined through steps 305 to 306; further, based on the first bandwidth predicted in step 303 and the bandwidth allocation ratio obtained in step 306, the first target bandwidth of each content distribution server in each first time period is obtained in step 307, that is, the initial bandwidth sequence is obtained; adjusting the first target bandwidth through step 308 to obtain a scheduling policy; the scheduling policy is executed, via step 309; in the course of executing the scheduling policy, download behavior information is generated, which can be used in the course of the next bandwidth allocation.

All the above optional technical solutions may be combined arbitrarily to form the optional embodiments of the present disclosure, and are not described herein again.

Fig. 5 is a block diagram illustrating a bandwidth allocation apparatus according to an example embodiment. Referring to fig. 5, the bandwidth allocation apparatus includes:

a first bandwidth obtaining module 501, configured to perform obtaining of a first bandwidth corresponding to a first time period, where the first time period is any time period after the current time, and the first bandwidth is a total bandwidth consumed for content distribution in the first time period;

a first parameter obtaining module 502 configured to perform obtaining first performance parameter information of each content distribution server, where each content distribution server is a server for content distribution, and the first performance parameter information of each content distribution server represents content distribution performance of each content distribution server;

a bandwidth allocation ratio determination module 503 configured to perform determining a bandwidth allocation ratio of each content distribution server based on the first performance parameter information of each content distribution server, the bandwidth allocation ratio of each content distribution server being positively correlated with the content distribution performance of the content distribution server;

a target bandwidth determining module 504 configured to perform determining a target bandwidth of each content distribution server based on the first bandwidth and the bandwidth distribution ratio of each content distribution server.

The bandwidth allocation device provided by the embodiment of the disclosure predicts the bandwidth required to be consumed for content distribution in a future time period, determines the bandwidth allocation proportion positively correlated to the content distribution performance of each content distribution server for each content distribution server, and further determines the target bandwidth required to be consumed for each content distribution server in the future time period according to the bandwidth allocation proportion, so that the target bandwidth of each content distribution server is matched with the content distribution performance of each content distribution server, and therefore, the bandwidth of each content distribution server can be effectively utilized according to the content distribution performance of each content distribution server, and the bandwidth utilization rate is improved.

In some embodiments, the first performance parameter information includes a plurality of performance parameters belonging to different dimensions, the plurality of performance parameters respectively characterizing content distribution performance of the content distribution server from the different dimensions; the bandwidth allocation ratio determining module 503 includes:

the performance parameter fusion unit is configured to perform fusion on multiple performance parameters of each content distribution server to obtain fusion performance parameters of each content distribution server;

a bandwidth allocation proportion determination unit configured to perform determination of a sum value of the fusion performance parameters of the content distribution servers as a total performance parameter; and determining the proportion between the fusion performance parameter and the total performance parameter of each content distribution server as the bandwidth allocation proportion of each content distribution server.

In some embodiments, the performance parameter fusion unit comprises:

the weight acquisition subunit is configured to perform acquisition of multiple performance parameters of each content distribution server and weights corresponding to the multiple performance parameters, and the weights represent importance of the corresponding performance parameters to the content distribution performance of each content distribution server;

and the performance parameter fusion subunit is configured to perform weighted fusion on the multiple performance parameters according to the weights corresponding to the multiple performance parameters to obtain fusion performance parameters of each content distribution server.

In some embodiments, a performance parameter fusion subunit configured to perform:

carrying out normalization processing on various performance parameters;

and performing weighted fusion on the multiple performance parameters after the normalization processing according to the weights corresponding to the multiple performance parameters to obtain the fusion performance parameters of each content distribution server.

In some embodiments, the target bandwidth determination module 504 is configured to perform:

In some embodiments, the target bandwidth determination module 504 includes:

a target bandwidth determining unit configured to perform determination of a first target bandwidth of each content distribution server as a product of a bandwidth allocation proportion of each content distribution server and the first bandwidth;

In some embodiments, the bandwidth adjusting unit is configured to perform:

according to various bandwidth adjustment strategies, the first target bandwidth of each content distribution server is adjusted respectively to obtain an adjusted third target bandwidth, and the sum of the third target bandwidth of each content distribution server is the first bandwidth;

In some embodiments, the bandwidth adjusting unit is configured to perform:

determining a plurality of bandwidth adjustment strategies, wherein each bandwidth adjustment strategy comprises adjustment bandwidths of a plurality of content distribution servers in a plurality of first time periods in the same period;

according to a plurality of bandwidth adjustment strategies, respectively adjusting the first target bandwidth of each content distribution server in each first time period to obtain an adjusted third target bandwidth, wherein the sum of the third target bandwidth of each content distribution server in each first time period is the first bandwidth corresponding to each first time period;

under various bandwidth adjustment strategies, determining a target quantity based on a third target bandwidth of each content distribution server in a plurality of first time periods, wherein the target quantity represents the total quantity of resources which need to be transferred out when each content distribution server is scheduled by the third target bandwidth of each content distribution server in each first time period in the plurality of first time periods;

and respectively determining the third target bandwidth of each content distribution server in each first time period as the second target bandwidth of each content distribution server in each first time period in the bandwidth adjustment strategy with the minimum target number.

In some embodiments, the bandwidth adjusting unit is configured to perform:

dividing the content distribution servers with the same quantity calculation mode into a group, wherein the quantity calculation mode represents a mode of calculating the quantity of resources to be transferred according to the target bandwidth of the content distribution servers;

In some embodiments, the bandwidth allocation apparatus further comprises:

and the content distribution server scheduling module is configured to execute scheduling of each content distribution server according to the target bandwidth of each content distribution server in a first time period.

In some embodiments, the bandwidth allocation apparatus further comprises:

the second parameter acquisition module is configured to perform acquisition of second performance parameter information of each content distribution server in a first time period, wherein the second performance parameter information of each content distribution server represents the content distribution performance of the content distribution server in the first time period;

and the bandwidth adjusting module is configured to execute deducting the target bandwidth of any content distribution server in the next first time period of the first time period by the target adjusting bandwidth in response to that the second performance parameter information of any content distribution server in the first time period does not meet the performance condition.

In some embodiments, the bandwidth adjustment module is further configured to perform increasing the target bandwidth of any content distribution server that meets the performance condition by the target adjustment bandwidth in a next first time period of the first time period.

In some embodiments, the bandwidth adjusting unit is configured to perform adjusting the first target bandwidth of each content distribution server to obtain the second target bandwidth of each content distribution server in response to the first time period belonging to the time period in the peak period.

In some embodiments, the first cycle includes a plurality of first time periods, the second cycle includes a plurality of second time periods, and the second cycle is a cycle before the first cycle, and the first bandwidth obtaining module 501 includes:

the second bandwidth determining unit is configured to determine second bandwidths corresponding to a plurality of second time periods in a second period based on the download behavior information acquired in the second period, wherein the download behavior information represents a content download situation in the second period, and the second bandwidth corresponding to any one of the second time periods represents a bandwidth consumed for content distribution in any one of the second time periods;

and the first bandwidth acquisition unit is configured to acquire first bandwidths corresponding to the plurality of first time periods based on second bandwidths corresponding to the plurality of second time periods.

In some embodiments, the download behavior information includes a location where a terminal that downloads the content is located and a data amount of the content;

a second bandwidth determination unit configured to perform:

for any second time period in the second period, determining a plurality of contents of which the corresponding terminal is positioned in any area range from the downloading behavior information collected in the second time period;

and determining a second bandwidth corresponding to the second time period and the area range based on the data volume of the plurality of contents and the duration of the second time period, wherein the second bandwidth represents the bandwidth consumed for content distribution for the terminals located in the area range in the second time period.

and for any area range and any second time period, predicting based on the area range and a second bandwidth corresponding to the second time period to obtain a first bandwidth, wherein the first bandwidth represents the bandwidth required to be consumed by content distribution for the terminals located in the area range in the first period corresponding to the second time period.

and for a first bandwidth corresponding to any area range and any first time period, determining the target bandwidth of each content distribution server with the service range including the area range based on the first bandwidth and the bandwidth allocation proportion of each content distribution server.

a second bandwidth determination unit configured to perform:

and determining a second time period and a second bandwidth corresponding to the operator based on the data volume of the plurality of contents and the duration of the second time period, wherein the second bandwidth represents the bandwidth consumed by content distribution through a content distribution server constructed on the network of the operator in the second time period.

and for any operator and any second time period, predicting based on second bandwidths corresponding to the operator and the second time period to obtain a first bandwidth, wherein the first bandwidth represents the bandwidth required to be consumed by content distribution through a content distribution server constructed on the network of the operator in the first period corresponding to the second time period.

and for a first bandwidth corresponding to any operator and any first time period, determining a target bandwidth of each content distribution server belonging to the operator based on the first bandwidth and a bandwidth distribution proportion of each content distribution server belonging to the operator.

With regard to the apparatus in the above-described embodiments, the specific manner in which each unit performs operations has been described in detail in the embodiments of the related method, and will not be described in detail here.

Fig. 6 is a block diagram of a server according to an exemplary embodiment, where the server 600 may generate a relatively large difference due to different configurations or performances, and may include one or more processors (CPUs) 601 and one or more memories 602, where the memories 602 are used for storing executable instructions, and the processors 601 are configured to execute the executable instructions to implement the bandwidth allocation methods provided by the method embodiments. Of course, the server may also have components such as a wired or wireless network interface, a keyboard, and an input/output interface, so as to perform input/output, and the server may also include other components for implementing the functions of the device, which are not described herein again.

In an exemplary embodiment, a computer-readable storage medium comprising instructions, such as the memory 602 comprising instructions, executable by the processor 601 of the server 600 to perform the bandwidth allocation method described above is also provided. Alternatively, the computer-readable storage medium may be a ROM (Read-Only Memory), a RAM (Random Access Memory), a CD-ROM (Compact Disc Read-Only Memory), a magnetic tape, a floppy disk, an optical data storage device, and the like.

In an exemplary embodiment, a computer program product is also provided, which comprises a computer program that, when being executed by a processor, implements the bandwidth allocation method in the above-mentioned respective method embodiments.

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

It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for bandwidth allocation, the method comprising:

determining a target bandwidth of each content distribution server based on the first bandwidth and a bandwidth allocation proportion of each content distribution server;

the determining the target bandwidth of each content distribution server based on the first bandwidth and the bandwidth allocation ratio of each content distribution server includes:

and adjusting the first target bandwidth of each content distribution server according to a plurality of bandwidth adjustment strategies to obtain a second target bandwidth of each content distribution server, so that the total amount of resources required to be transferred out when each content distribution server is scheduled by the second target bandwidth of each content distribution server is minimum.

2. The bandwidth allocation method according to claim 1, wherein the first performance parameter information includes a plurality of performance parameters belonging to different dimensions, and the plurality of performance parameters respectively characterize content distribution performance of the content distribution server from different dimensions;

3. The method of claim 2, wherein the fusing the multiple performance parameters of each content distribution server to obtain the fused performance parameters of each content distribution server comprises:

4. The method for allocating bandwidth according to claim 3, wherein the performing weighted fusion on the multiple performance parameters according to the weights corresponding to the multiple performance parameters to obtain the fusion performance parameters of each content distribution server includes:

carrying out normalization processing on the multiple performance parameters;

5. The method of claim 1, wherein the adjusting the first target bandwidth of each content distribution server according to the plurality of bandwidth adjustment policies to obtain the second target bandwidth of each content distribution server, so as to minimize the total amount of resources to be transferred out when scheduling each content distribution server with the second target bandwidth of each content distribution server, comprises:

determining the plurality of bandwidth adjustment strategies, wherein each bandwidth adjustment strategy comprises the adjustment bandwidth of each content distribution server;

6. The method of claim 1, wherein the adjusting the first target bandwidth of each content distribution server according to the plurality of bandwidth adjustment policies to obtain the second target bandwidth of each content distribution server, so as to minimize the total amount of resources to be transferred out when scheduling each content distribution server with the second target bandwidth of each content distribution server, comprises:

determining the plurality of bandwidth adjustment strategies, wherein each bandwidth adjustment strategy comprises adjustment bandwidths of the plurality of content distribution servers in a plurality of first time periods in the same period;

7. The method of claim 1, wherein the adjusting the first target bandwidth of each content distribution server according to the plurality of bandwidth adjustment policies to obtain the second target bandwidth of each content distribution server, so as to minimize the total amount of resources to be transferred out when scheduling each content distribution server with the second target bandwidth of each content distribution server, comprises:

and adjusting the first target bandwidth of each content distribution server in each group according to the multiple bandwidth adjustment strategies to obtain a second target bandwidth of each content distribution server, so that the total amount of resources required to be transferred out when each content distribution server is scheduled by the second target bandwidth of each content distribution server is minimum.

8. The bandwidth allocation method according to claim 1, wherein after determining the target bandwidth of each of the content distribution servers based on the first bandwidth and the bandwidth allocation ratio of each of the content distribution servers, the bandwidth allocation method further comprises:

9. The method of allocating bandwidth according to claim 8, wherein after scheduling each of the content distribution servers according to a target bandwidth of each of the content distribution servers in the first time period, the method further comprises:

10. The bandwidth allocation method according to claim 9, wherein in response to that the second performance parameter information of any content distribution server in the first time period does not satisfy the performance condition, after subtracting the target bandwidth of any content distribution server in a first time period next to the first time period from the target adjustment bandwidth, the bandwidth allocation method further comprises:

11. The method of claim 1, wherein the adjusting the first target bandwidth of each content distribution server according to the plurality of bandwidth adjustment policies to obtain the second target bandwidth of each content distribution server, so as to minimize the total amount of resources to be transferred out when scheduling each content distribution server with the second target bandwidth of each content distribution server, comprises:

and in response to the time period that the first time period belongs to the peak period, adjusting the first target bandwidth of each content distribution server according to the multiple bandwidth adjustment strategies to obtain a second target bandwidth of each content distribution server, so that the total amount of resources required to be transferred out when each content distribution server is scheduled by the second target bandwidth of each content distribution server is minimum.

12. The method of claim 1, wherein the first cycle includes a plurality of first time periods, the second cycle includes a plurality of second time periods, the second cycle is a cycle before the first cycle, and the obtaining the first bandwidth corresponding to the first time period includes:

13. The bandwidth allocation method according to claim 12, wherein the download behavior information includes a location of a terminal that downloads the content and a data amount of the content;

14. The method of claim 13, wherein the obtaining the first bandwidths corresponding to the first time periods based on the second bandwidths corresponding to the second time periods comprises:

15. The method of allocating bandwidth according to claim 14, wherein the determining a target bandwidth for each of the content distribution servers based on the first bandwidth and a bandwidth allocation ratio of each of the content distribution servers includes:

16. The bandwidth allocation method according to claim 12, wherein the download behavior information includes an operator to which a terminal that downloads the content belongs and a data amount of the content;

17. The method of claim 16, wherein the obtaining the first bandwidths corresponding to the first time periods based on the second bandwidths corresponding to the second time periods comprises:

18. The method of allocating bandwidth according to claim 17, wherein the determining a target bandwidth for each of the content distribution servers based on the first bandwidth and a bandwidth allocation ratio of each of the content distribution servers includes:

19. A bandwidth allocation apparatus, comprising:

a target bandwidth determination module configured to perform determination of a target bandwidth of each of the content distribution servers based on the first bandwidth and a bandwidth allocation ratio of each of the content distribution servers;

the target bandwidth determination module includes:

and the bandwidth adjusting unit is configured to perform adjustment on the first target bandwidth of each content distribution server according to a plurality of bandwidth adjusting strategies to obtain a second target bandwidth of each content distribution server, so that the total amount of resources required to be transferred out when each content distribution server is scheduled by the second target bandwidth of each content distribution server is minimum.

20. The bandwidth allocation apparatus according to claim 19, wherein the first performance parameter information includes a plurality of performance parameters belonging to different dimensions, and the plurality of performance parameters respectively characterize content distribution performance of the content distribution server from different dimensions; the bandwidth allocation proportion determining module comprises:

21. The apparatus for allocating bandwidth as defined in claim 20, wherein the performance parameter fusion unit comprises:

22. The apparatus according to claim 21, wherein the performance parameter fusion subunit is configured to perform:

carrying out normalization processing on the multiple performance parameters;

23. The apparatus according to claim 19, wherein the bandwidth adjusting unit is configured to perform:

24. The apparatus according to claim 19, wherein the bandwidth adjusting unit is configured to perform:

25. The apparatus according to claim 19, wherein the bandwidth adjusting unit is configured to perform:

26. The apparatus of claim 19, further comprising:

27. The apparatus for allocating bandwidth as defined in claim 26, wherein the apparatus for allocating bandwidth further comprises:

28. The apparatus according to claim 27, wherein the bandwidth adjustment module is further configured to perform increasing the target bandwidth of any content distribution server satisfying the performance condition by the target adjustment bandwidth in a next first time period of the first time period.

29. The apparatus according to claim 19, wherein the bandwidth adjusting unit is configured to perform, in response to the first time period belonging to a time period in a peak period, adjusting the first target bandwidth of each of the content distribution servers according to the plurality of bandwidth adjustment policies to obtain the second target bandwidth of each of the content distribution servers, so as to minimize a total amount of resources to be transferred out when each of the content distribution servers is scheduled at the second target bandwidth of each of the content distribution servers.

30. The apparatus of claim 19, wherein the first cycle comprises a plurality of first time periods, wherein the second cycle comprises a plurality of second time periods, and wherein the second cycle is a cycle prior to the first cycle, wherein the first bandwidth obtaining module comprises:

31. The apparatus according to claim 30, wherein the download behavior information comprises a location of a terminal downloading the content and a data amount of the content;

the second bandwidth determination unit configured to perform:

32. The apparatus according to claim 31, wherein the first bandwidth acquiring unit is configured to perform:

33. The apparatus according to claim 32, wherein the target bandwidth determining module is configured to perform:

34. The apparatus according to claim 30, wherein the download behavior information includes an operator to which a terminal downloading the content belongs and a data amount of the content;

the second bandwidth determination unit configured to perform:

35. The apparatus according to claim 34, wherein the first bandwidth obtaining unit is configured to perform:

36. The apparatus according to claim 35, wherein the target bandwidth determining module is configured to perform:

37. A server, characterized in that the server comprises:

one or more processors;

a memory for storing the processor-executable instructions;

wherein the processor is configured to execute the instructions to implement the bandwidth allocation method of any of claims 1 to 18.

38. A computer-readable storage medium, wherein instructions in the computer-readable storage medium, when executed by a processor of a server, enable the server to perform the bandwidth allocation method of any one of claims 1 to 18.