WO2018112877A1

WO2018112877A1 - Path calculating and access request distributing methods, devices and systems

Info

Publication number: WO2018112877A1
Application number: PCT/CN2016/111692
Authority: WO
Inventors: 朱磊
Original assignee: 深圳前海达闼云端智能科技有限公司
Priority date: 2016-12-23
Filing date: 2016-12-23
Publication date: 2018-06-28
Also published as: CN106716937A

Abstract

Path calculating and access request distributing methods, devices and systems. The path calculating method comprises: calculating distances among server nodes according to obtained delay and packet loss probability information among the server nodes, and calculating the distance between an edge server and a source station server according to obtained delay and packet loss probability information between the edge server and the source station server (101); and calculating path information according to the distances among the server nodes and the distance between the edge server and the source station server, and sending the path information to a plurality of portal servers (102). According to the methods, devices and systems, transmission path load balancing is achieved based on information a plurality of optimum paths calculated by a central controller, thereby increasing the user access speed.

Description

Path calculation and access request distribution method, device and system

Technical field

The present application relates to the field of Internet technologies, and in particular, to a path calculation and access request distribution method, apparatus, and system.

Background technique

The Application Delivery Network (ADN) is usually based on the network layer or transport layer and provides protocols based on Internet Protocol (IP), Transmission Control Protocol (TCP), or User Datagram Protocol ( UDP: User Datagram Protocol) accelerates the application transfer to improve user access speed. The ADN network access usually uses the CNAME record mode of the Domain Name System (DNS) to implement traffic guidance. In the ADN network transmission process, TCP (or UDP) proxy or IP routing is usually used for transmission.

The disadvantages of the prior art are:

In the ADN network transmission process, the pre-configured static path transmission mode cannot be dynamically adjusted according to the network status, and does not have the ability to cope with the sudden interruption or deterioration of the network; the pre-configured dynamic path transmission mode is adopted. The calculated shortest path is usually used to implement the transmission of the access request, but when the bulk access request floods into the ADN network, the path is easily congested and the user experience is poor.

Summary of the invention

The embodiment of the present application proposes a path calculation and access request distribution method, device and system, so that bulk access requests from users can be multi-path and parallel transmission, so as to solve the technical problem that the optimal path congestion leads to slow transmission speed.

In one aspect, an embodiment of the present application provides a path calculation method, which is applied to a central controller, and includes:

Calculate the distance between the server nodes according to the obtained delay and packet loss rate information between the server nodes, and calculate the edge server according to the obtained delay and packet loss rate information of the edge server to the source server. The distance to the source server;

Calculating the path information according to the distance between the server nodes and the edge server to the source server, and sending the path information to multiple entry servers;

The path information includes a plurality of pieces of optimal path information and path indication information, where the path indication information is used to indicate a transmission capability of each of the plurality of optimal paths to an access request.

In another aspect, an embodiment of the present application provides an access request distribution method, which is applied to an ingress server, and includes:

Obtaining, by the application, the path information corresponding to the application, where the path information includes multiple pieces of optimal path information and path indication information, where the path indication information is used to indicate the multiple optimal The ability of each optimal path in the path to transmit access requests;

And according to the path information, the access request is sent to the source station server in parallel via the plurality of optimal paths.

In another aspect, an embodiment of the present application provides a path calculation apparatus, which is applied to a central controller, and includes:

The distance module is configured to calculate a distance between the server nodes according to the obtained delay and packet loss rate information between the server nodes, and according to the obtained delay time and packet loss rate information of the edge server to the source server , calculate the distance from the edge server to the source server;

a path module, configured to calculate path information according to a distance between server nodes and an edge server to a source server, and send the path information to multiple entry servers;

In another aspect, an embodiment of the present application provides an access request distribution apparatus, which is applied to Port server, including:

a receiving module, configured to acquire path information corresponding to the application when the application initiates an access request, where the path information includes multiple pieces of optimal path information and path indication information, where the path indication information is used to indicate Decoding the transmission capability of each of the plurality of optimal paths to the access request;

And a sending module, configured to send the access request to the source station server in parallel according to the path information, by using the multiple optimal paths.

In another aspect, an embodiment of the present application provides an access request distribution system, including:

The central controller is configured to calculate a distance between the server nodes according to the obtained delay and packet loss rate information between the server nodes, and according to the obtained delay time and packet loss rate of the edge server to the source server Information, calculating the distance from the edge server to the source server; and

Calculating path information including multiple optimal paths according to the distance between the server nodes and the edge server to the source server, and transmitting the path information to multiple entry servers;

An access server, configured to acquire path information corresponding to the application when receiving an access request initiated by the application;

And sending, according to the path information, the access request to the source station server in parallel via the multiple optimal paths;

In another aspect, an embodiment of the present application provides an electronic device, which is applied to a central controller, where the electronic device includes:

Transceiver, memory, one or more processors;

One or more modules, the one or more modules being stored in the memory and configured to be executed by the one or more processors, the one or more modules comprising Instructions for each step.

In another aspect, an embodiment of the present application provides an electronic device, which is applied to an access server, where the electronic device includes:

Transceiver, memory, one or more processors;

In another aspect, embodiments of the present application provide a computer program product for use with an electronic device, the computer program product comprising a computer readable storage medium and a computer program mechanism embedded therein, the computer program mechanism Instructions are included for performing the various steps in the above methods.

The benefits are as follows:

In this embodiment, for a scenario in which the bulk access request is implemented by the ADN network, the central controller calculates the distance between the server nodes according to the obtained delay and packet loss rate information between the server nodes, and according to the obtained Calculate the distance from the edge server to the source server by the delay and packet loss rate information of the edge server to the source server; calculate the path information according to the distance between the server nodes and the edge server to the source server, and Sent to multiple portal servers. The path information includes a plurality of pieces of optimal path information and path indication information, where the path indication information is used to indicate a transmission capability of each of the plurality of optimal paths to an access request. In this way, it can be avoided that the bulk access request is transmitted through only one shortest path in the prior art, and the shortest path is congested. The other paths are not as the shortest path, the path load utilization is low, and the resource allocation is unreasonable, thereby achieving The technical effect of improving user access speed.

DRAWINGS

Specific embodiments of the present application will be described below with reference to the accompanying drawings, in which:

1 is a schematic diagram of a method for calculating a path of a central controller side in the first embodiment of the present application;

2 is a schematic diagram of a method for accessing an access server side access request in the first embodiment of the present application;

3 is a flowchart of a method for path calculation and access request distribution in Embodiment 2 of the present application;

4 is a schematic flowchart of path calculation and access request distribution in Embodiment 2 of the present application;

5 is a structural diagram of a device for calculating a path of a central controller side in Embodiment 3 of the present application;

6 is a structural diagram of an apparatus for accessing an access server side access request in Embodiment 3 of the present application;

7 is a system architecture diagram of access request distribution in Embodiment 4 of the present application;

FIG. 8 is a schematic structural diagram of an electronic device according to Embodiment 5 of the present application.

detailed description

The exemplary embodiments of the present application are further described in detail below with reference to the accompanying drawings, in which the embodiments described are only a part of the embodiments of the present application, but not all embodiments. An exhaustive example. And in the case of no conflict, the features in the embodiments and the embodiments in the description can be combined with each other.

The inventor noticed in the process of studying the prior art:

When the ADN network receives the access request, the shortest path is calculated as the optimal path according to the access request, so that the access request accelerates the transmission of the access request via the optimal path, but when the bulk access request is connected to the ADN network, A single optimal path transmission leads to a steep increase in optimal path load utilization, other paths are not fully utilized, and load utilization is low, resulting in unreasonable resource allocation.

For the above-mentioned deficiencies, the embodiment of the present application proposes that the central controller calculates the distance between the server nodes according to the obtained delay and packet loss rate information between the server nodes, and according to the obtained edge server to the source station server. The delay and packet loss rate information is used to calculate the distance from the edge server to the source server; according to the distance between the server nodes and the edge server to the source server, the path information is calculated and sent to multiple portal servers. The path information includes a plurality of pieces of optimal path information and path indication information, where the path indication information is used to indicate a transmission capability of each of the plurality of optimal paths to an access request. Embodiments of the present application utilize central control The device calculates a plurality of optimal path information, so that the ingress server implements load balancing of the transmission path according to the plurality of optimal path information, and achieves the technical effect of improving user access speed.

In order to facilitate the implementation of the present application, the following examples are described.

Embodiment 1

FIG. 1 is a schematic diagram of a method for calculating a path of a central controller side in the first embodiment of the present application, as shown in FIG. 1 , including:

Step 101: Calculate the distance between the server nodes according to the obtained delay and packet loss rate information between the server nodes, and calculate the delay and packet loss rate information of the obtained edge server to the source server. The distance from the edge server to the source server.

Step 102: Calculate the path information according to the distance between the server nodes and the edge server to the source server, and send the path information to multiple portal servers. The path information includes a plurality of pieces of optimal path information and path indication information, where the path indication information is used to indicate a transmission capability of each of the plurality of optimal paths to an access request.

In step 101, the central controller periodically receives the delay and packet loss rate of each server node in the network through mutual detection between the server nodes in the network, and the intra-edge edge server passes through the intra-network edge server to the source station server. Delay and packet loss rate obtained by one-way detection. According to the detected data between the server nodes in the network and the edge server to the source server in the network, the distance between each server node and the edge server to the source server is calculated. In addition, the central controller periodically receives all the server nodes in the network to detect the bandwidth utilization calculated by the actual NIC flow meter of the device through self-detection. The delay in the ADN network may be a TCP delay or an Internet Control Message Protocol (ICMP) delay, which is not further limited herein.

In step 102, the central controller further calculates the N optimal paths by using the shortest path algorithm YEN according to the calculated distance values between the server nodes and the intra-edge edge server to the source station server. The value of N can be set according to the actual situation. When the calculated optimal path number does not reach N, the calculated optimal path number is taken as the standard.

The path information sent by the central controller to the multiple ingress servers includes multiple pieces of optimal path information and path indication information, where the path indication information may be a weight attribute set for each optimal path to represent each optimal path pair. The transmission capability of the access request. The optimal path weight is the ratio of the distance value of an optimal path in the optimal path to the sum of the distance values of all the optimal paths. The specific calculation method is:

Where N is the optimal number of paths for the same ingress server to reach the same source server. If there is only one optimal path in the optimal path, the weight of the optimal path is set to 1; if there are multiple optimal paths in the optimal path, the optimal path weight algorithm is used to calculate each of the most The weight of the excellent path.

In this embodiment, the central controller sends the multiple optimal path information and the weight attribute of each of the plurality of optimal paths to the ingress server, so that the ingress server receives the bulk access request. The parallel transmission of the batch access request is implemented according to the path information from the central controller, thereby achieving the technical effect of balancing the optimal path load utilization and rational resource allocation.

Further, the server node is a server node in the ADN network, the edge server is an ADN network inner edge server, and the ADN network is an application distribution network.

Further, the calculating the distance between the server nodes according to the obtained delay and packet loss rate information between the server nodes, and the delay and packet loss rate information according to the obtained edge server to the source server The steps of calculating the distance from the edge server to the source server include:

Set the sensitivity of the packet loss rate according to the type of data to be accessed by the access request;

Calculate the distance between server nodes based on the sensitivity of the packet loss rate and the delay and packet loss rate between the server nodes, and the sensitivity based on the packet loss rate and the delay from the edge server to the source server. And the packet loss rate information, calculate the distance between the server nodes and the edge server to the source server.

The central controller calculates the delay and packet loss rate information between the obtained server nodes. The distance between the server nodes and the delay and packet loss rate information of the obtained edge server to the source server determine the distance from the edge server to the source server. The specific calculation method is as follows:

Where N is [0, X], where N represents the sensitivity of the packet loss rate. When X is 0, it means that only the round-trip delay (RTT: Round-Trip Time) is considered regardless of the packet loss rate Loss, when X is At 3 o'clock, it indicates that the packet loss rate is higher. The higher the value of X, the higher the loss sensitivity of the packet loss rate, and is not limited to 3.

Specifically, the value of the N is determined for different application scenarios. The data type is parsed by the received access request. If the access request is a request to access a video conference class, the request for the packet loss rate is lower. High, in the calculation method of such an application scenario, the value of N can be defined as 3; similarly, if the access request is a request to access a file download class, the request for the packet loss rate is lower, then In the calculation method of such an application scenario, the value of N can be defined as 0.

Further, it also includes:

Obtain load utilization of all server nodes except the ingress server;

When the load utilization of a certain server node other than the ingress server reaches the threshold in the optimal path, the optimal path to which the certain server node belongs is deleted from the multiple optimal paths.

For example, the server node load utilization is the server node bandwidth utilization. The central controller uses the shortest path algorithm YEN to calculate the N shortest paths from the same ingress server to the same source server as N. If the calculated optimal path number is less than N, the calculated optimal path number shall prevail; if the calculated optimal path has a bandwidth of a server node other than the ingress server If the utilization reaches the threshold (the threshold is 100% by default), the optimal path is deleted from the optimal path; if there is only one optimal path in the optimal path, even if the optimal path has one other than the ingress server The bandwidth utilization of the server node reaches the threshold and the optimal path is also retained.

Each server node presets the rated bandwidth and obtains the actual bandwidth by collecting the uplink and downlink bandwidths. The specific calculation method of the bandwidth utilization is:

2 is a schematic diagram of a method for accessing an access server side access request in the first embodiment of the present application. As shown in FIG. 2, the method includes:

In step 201, when the application-initiated access request is received, the path information corresponding to the application is obtained, where the path information includes multiple pieces of optimal path information and path indication information, where the path indication information is used to indicate the multiple The ability of each optimal path to transmit access requests in the optimal path.

Step 202: Send the access request to the source station server in parallel via the plurality of optimal paths according to the path information.

In step 201, the ingress server receives the access request, determines an application that initiates the access request according to the application identifier in the access request, and acquires corresponding path information according to the application category.

Further, the step of transmitting the access request to the source station server in parallel via the multiple optimal paths according to the path information includes:

Calculating, according to the path indication information, a parallel transmission ratio corresponding to each of the plurality of optimal paths;

The access request is sent to the source station server in parallel according to the parallel transmission ratio.

The parallel transmission ratio of each optimal path in the multiple optimal paths is the weight ratio of each optimal path in the multiple optimal paths. For example, the portal server calculates the weight of the optimal path based on the path information. The ratio of the weights corresponding to each optimal path in the multiple optimal paths, that is, the ratio of the weights of a certain optimal path to the sum of the weights of all the optimal paths. The specific calculation method is:

Where N is [1, X], and N represents the number of optimal paths. Usually, the default N is 3.

In this embodiment, the portal server classifies the batch access request according to the application category, obtains path information including multiple optimal paths corresponding to the application category, and selects the classified access request according to the path information according to the path information. The weight ratio of the path is sent to the source station server in parallel via multiple optimal paths, so that the portal server implements the load of the transmission path based on the multiple optimal paths. Balance, while achieving the technical effect of improving user access speed.

The second embodiment of the present application provides a detailed description of the first embodiment of the present application by taking a specific scenario as an example.

Embodiment 2

FIG. 3 is a flowchart of a path calculation and an access request distribution method according to Embodiment 2 of the present application. As shown in FIG. 3, the path calculation and the specific implementation method of the access request distribution are:

Step 301: The server node periodically reports the detection result to the central controller, and the detection result includes at least a TCP delay, a packet loss rate, and a bandwidth utilization. 4 is a schematic flowchart of path calculation and access request distribution in Embodiment 2 of the present application. As shown in FIG. 4, ping detection is performed between server nodes in an ADN network, and TCP delay and packet loss rate between server nodes are obtained. And the server node calculates the bandwidth utilization by detecting the actual NIC traffic of the server, and sends the TCP delay, the packet loss rate, and the bandwidth utilization between the server nodes to the central controller. The edge server of the ADN network pings the source server to obtain the TCP delay and packet loss rate of the edge server to the source server, and the edge server calculates the bandwidth utilization by detecting its actual network card traffic, and the edge server is used. The TCP delay, packet loss rate, and bandwidth utilization to the source server are sent to the central controller.

Step 302: The central controller calculates the distance between the server nodes according to the received TCP delay and the packet loss rate information between the server nodes, and the TCP delay and the loss according to the obtained edge server to the source server. The packet rate information calculates the distance from the edge server to the source server.

Step 303: The central controller obtains N optimal paths from each ingress server to all source server according to the distance between the server nodes and the distance from the edge server to the source server (N is set according to actual conditions, and the default is 3) .

Step 304: The central controller performs weight calculation on the N optimal paths, and sets weights for the weight attributes of each of the N optimal paths.

Step 305: The central controller sends the N optimal path information including each shortest path weight to multiple ingress servers.

Step 306: When the portal server receives the bulk access request from the user, according to the initiating batch The category of the application that accesses the request obtains the corresponding N optimal paths.

Step 307: The ingress server implements parallel transmission of the bulk access request according to the weight ratio of each of the N optimal paths.

Embodiment 3

Based on the same inventive concept, a path calculation device is also provided in the embodiment of the present application. Since the principle of solving the problem of these devices is similar to a path calculation method, the implementation of these devices can refer to the implementation of the method, and the repetition is no longer Narration.

5 is a structural diagram of a device for calculating a central controller side path according to Embodiment 3 of the present application. As shown in FIG. 5, the path computing device 500 may include:

The distance module 501 is configured to calculate, according to the obtained delay and packet loss rate information between the server nodes, a distance between the server nodes, and a delay and a packet loss rate according to the obtained edge server to the source server. Information, calculating the distance between server nodes and the edge server to the source server.

The path module 502 is configured to calculate path information according to the distance between the server nodes and the edge server to the source server, and send the path information to the multiple portal servers. The path information includes a plurality of pieces of optimal path information and path indication information, where the path indication information is used to indicate a transmission capability of each of the plurality of optimal paths to an access request.

Further, the distance module 501 includes:

The setting unit 5011 is configured to set the sensitivity of the packet loss rate according to the type of data to be accessed by the access request.

The calculating unit 5012 is configured to calculate a distance between the server nodes according to the sensitivity of the packet loss rate and the delay and packet loss rate information between the server nodes, and the sensitivity according to the packet loss rate, and the edge server to The delay and packet loss rate information of the source station server calculates the distance between the server nodes and the edge server to the source server.

Further, the path module 502 includes:

The obtaining unit 5021 is configured to obtain load utilization of all server nodes except the ingress server.

a deleting unit 5022, configured to delete an optimality of the server node from the multiple optimal paths when a load utilization of a server node other than the ingress server reaches a threshold in the optimal path. path.

FIG. 6 is a structural diagram of an apparatus for accessing an access server side access request in the third embodiment of the present application. As shown in FIG. 6, the access request distribution apparatus 600 may include:

The receiving module 601 is configured to: when receiving an access request initiated by the application, acquire path information corresponding to the application, where the path information includes multiple pieces of optimal path information and path indication information, where the path indication information is used to indicate The transmission capability of each of the plurality of optimal paths to the access request.

The sending module 602 is configured to send the access request to the source station server in parallel according to the path information by using the multiple optimal paths.

Further, the sending module 602 includes:

The ratio unit 6021 is configured to calculate, according to the path indication information, a parallel transmission ratio corresponding to each of the plurality of optimal paths.

The parallel unit 6022 is configured to send the access request to the source station server in parallel according to the parallel transmission ratio.

Embodiment 4

Based on the same inventive concept, an access request distribution system is also provided in the embodiment of the present application. Since the principle of solving the problem of these devices is similar to the path calculation and access request distribution method, a path calculation and the access request distribution device, these devices are similar. For the implementation of the method, refer to the implementation of the method, and the repetition will not be repeated.

FIG. 7 is a system architecture diagram of the access request distribution in the fourth embodiment of the present application. As shown in FIG. 7, the access request distribution system 700 may include:

The central controller 701 is configured to: according to the obtained delay and packet loss rate between server nodes Calculate the distance between the server nodes, and calculate the distance between the server nodes and the edge server to the source server according to the obtained delay and packet loss rate information of the edge server to the source server;

According to the distance between the server nodes and the edge server to the source server, path information including multiple optimal paths is calculated and sent to multiple portal servers.

The portal server 702 is configured to acquire path information corresponding to the application when receiving an access request initiated by the application;

Embodiment 5

Based on the same inventive concept, an electronic device is also provided in the embodiment of the present application. Since the principle is similar to a path calculation and access request distribution method, the implementation of the method may refer to the implementation of the method, and the repeated description is not repeated.

8 is a schematic structural diagram of an electronic device in Embodiment 5 of the present application. As shown in FIG. 8, the electronic device includes: a transceiver device 801, a memory 802, one or more processors 803, and one or more modules. The one or more modules are stored in the memory and configured to be executed by the one or more processors, the one or more modules including steps for performing the steps of any of the above methods instruction.

Embodiment 6

Based on the same inventive concept, the embodiment of the present application further provides a computer program product for use in combination with an electronic device. Since the principle is similar to a path calculation and access request distribution method, the implementation may refer to the implementation of the method, and the method is repeated. I won't go into details here. The computer program product comprises a computer readable storage medium and a computer program mechanism embedded therein, the computer program The mechanism includes instructions for performing the various steps in any of the above methods.

For the convenience of description, the various parts of the above-described apparatus are separately described by functions into various modules. Of course, the functions of each module or unit may be implemented in the same software or hardware in the implementation of the present application.

Those skilled in the art will appreciate that embodiments of the present application can be provided as a method, system, or computer program product. Thus, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment in combination of software and hardware. Moreover, the application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (system), and computer program products according to embodiments of the present application. It will be understood that each flow and/or block of the flowchart illustrations and/or FIG. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine for the execution of instructions for execution by a processor of a computer or other programmable data processing device. Means for implementing the functions specified in one or more of the flow or in a block or blocks of the flow chart.

The computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device. The apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.

These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device. The instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.

While the preferred embodiment of the present application has been described, it will be apparent that those skilled in the art can make further changes and modifications to the embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and the modifications and

Claims

A path calculation method is applied to a central controller, which includes:

Calculate the distance between the server nodes according to the obtained delay and packet loss rate information between the server nodes, and calculate the edge server according to the obtained delay and packet loss rate information of the edge server to the source server. The distance to the source server;

Calculating the path information according to the distance between the server nodes and the edge server to the source server, and sending the path information to multiple entry servers;

The path information includes a plurality of pieces of optimal path information and path indication information, where the path indication information is used to indicate a transmission capability of each of the plurality of optimal paths to an access request.
The method of claim 1, wherein the server node is a server node within an ADN network, the edge server is an ADN network inner edge server, and the ADN network is an application distribution network.
The method according to claim 1, wherein the calculating the distance between the server nodes according to the obtained delay and packet loss rate information between the server nodes, and according to the obtained edge server to the source The delay and packet loss rate information of the station server, and the steps of calculating the distance from the edge server to the source server include:

Set the sensitivity of the packet loss rate according to the type of data to be accessed by the access request;

Calculate the distance between server nodes based on the sensitivity of the packet loss rate and the delay and packet loss rate between the server nodes, and the sensitivity based on the packet loss rate and the delay from the edge server to the source server. And the packet loss rate information, calculate the distance from the edge server to the source server.
The method of claim 1 further comprising:

Obtain load utilization of all server nodes except the ingress server;

When the load utilization of a certain server node other than the ingress server reaches the threshold in the optimal path, the optimal path to which the certain server node belongs is deleted from the multiple optimal paths.
An access request distribution method is applied to an ingress server, and includes:

Obtaining, by the application, the path information corresponding to the application, where the path information includes multiple pieces of optimal path information and path indication information, where the path indication information is used to indicate the multiple optimal The ability of each optimal path in the path to transmit access requests;

And according to the path information, the access request is sent to the source station server in parallel via the plurality of optimal paths.
The method according to claim 5, wherein the step of transmitting the access request to the source station server in parallel via the plurality of optimal paths according to the path information comprises:

Calculating, according to the path indication information, a parallel transmission ratio corresponding to each of the plurality of optimal paths;

The access request is sent to the source station server in parallel according to the parallel transmission ratio.
A path computing device is applied to a central controller, and includes:

The distance module is configured to calculate a distance between the server nodes according to the obtained delay and packet loss rate information between the server nodes, and according to the obtained delay time and packet loss rate information of the edge server to the source server , calculate the distance from the edge server to the source server;

a path module, configured to calculate path information according to a distance between server nodes and an edge server to a source server, and send the path information to multiple entry servers;

The path information includes a plurality of pieces of optimal path information and path indication information, where the path indication information is used to indicate a transmission capability of each of the plurality of optimal paths to an access request.
The apparatus according to claim 7, wherein the server node is a server node in an ADN network, the edge server is an ADN network inner edge server, and the ADN network is an application distribution network.
The device of claim 7 wherein said distance module further comprises:

a setting unit, configured to set a sensitivity of a packet loss rate according to a data type to be accessed by the access request;

A calculation unit for sensitivity based on packet loss rate and delay and loss between server nodes The packet rate information is calculated, and the distance between the server nodes is calculated, and the distance from the edge server to the source server is calculated according to the sensitivity of the packet loss rate and the delay and packet loss rate information of the edge server to the source server.
The device of claim 7, wherein the path module further comprises:

An obtaining unit, configured to acquire load utilization of all server nodes except the portal server;

a deleting unit, configured to delete, when the load utilization of a server node other than the ingress server reaches a threshold, the optimal path to the server node from the multiple optimal paths .
An access request distribution device is applied to an portal server, and includes:

a receiving module, configured to acquire path information corresponding to the application when the application initiates an access request, where the path information includes multiple pieces of optimal path information and path indication information, where the path indication information is used to indicate Decoding the transmission capability of each of the plurality of optimal paths to the access request;

And a sending module, configured to send the access request to the source station server in parallel according to the path information, by using the multiple optimal paths.
The device of claim 11, wherein the sending module further comprises:

And a ratio unit, configured to calculate, according to the path indication information, a parallel transmission ratio corresponding to each of the plurality of optimal paths;

And a parallel unit, configured to send the access request to the source station server in parallel according to the parallel transmission ratio.
An access request distribution system, comprising:

The central controller is configured to calculate a distance between the server nodes according to the obtained delay and packet loss rate information between the server nodes, and according to the obtained delay time and packet loss rate of the edge server to the source server Information, calculating the distance from the edge server to the source server; and

Calculate the package based on the distance between the server nodes and the edge server to the source server Path information containing multiple optimal paths and sent to multiple ingress servers;

An access server, configured to acquire path information corresponding to the application when receiving an access request initiated by the application;

And sending, according to the path information, the access request to the source station server in parallel via the multiple optimal paths;

The path information includes a plurality of pieces of optimal path information and path indication information, where the path indication information is used to indicate a transmission capability of each of the plurality of optimal paths to an access request.
An electronic device, which is applied to a central controller, the electronic device comprising:

Transceiver, memory, one or more processors;

One or more modules stored in the memory and configured to be executed by the one or more processors, the one or more modules including for performing claim 1 The instructions of the various steps in any of the methods described in 4.
An electronic device, which is applied to an access server, the electronic device comprising:

Transceiver, memory, one or more processors;

One or more modules stored in the memory and configured to be executed by the one or more processors, the one or more modules including for performing claim 5 The instructions of the various steps in any of the methods described in -6.
A computer program product for use in conjunction with an electronic device, the computer program product comprising a computer readable storage medium and a computer program mechanism embedded therein, the computer program mechanism comprising means for performing any of claims 1-3 An instruction for each step in the method.
A computer program product for use in conjunction with an electronic device, the computer program product comprising a computer readable storage medium and a computer program mechanism embedded therein, the computer program mechanism comprising means for performing any of claims 5-6 An instruction for each step in the method.