CN108243234B

CN108243234B - Access scheduling method and device

Info

Publication number: CN108243234B
Application number: CN201611225420.5A
Authority: CN
Inventors: 李智; 熊刚; 杨辉鱼
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2016-12-27
Filing date: 2016-12-27
Publication date: 2020-04-24
Anticipated expiration: 2036-12-27
Also published as: CN108243234A

Abstract

The disclosure discloses an access scheduling method and device. The method comprises the following steps: receiving user reported data to obtain network detection data from each user, wherein the network detection data comprises network connection information of the user and a network quality detection value of a path; aggregating users according to the network connection information of the users according to a preset user aggregation granularity to obtain a user group, wherein the paths accessed by the aggregated users form a plurality of paths of the user group, and the user aggregation granularity is fine granularity; acquiring a reference path in the user group according to the network quality detection value, and identifying a fluctuation path according to the difference value of other paths in the user group relative to the reference path in the network quality detection value; and scheduling users on the fluctuation path to avoid accessing the fluctuation path. In the process, the fluctuation in the path is accurately and quickly identified, the fluctuation is responded in time, effective access scheduling is realized for path access performed by a user, and operation and maintenance cost is not brought.

Description

Access scheduling method and device

Technical Field

The present disclosure relates to the field of internet application technologies, and in particular, to an access scheduling method and apparatus.

Background

The access scheduling is used for realizing scheduling for server access of the user, so that the user can access the network through accessing the server to realize the service configured by the server for the user. In addition to the capacity and load of the server, the access scheduling needs to consider the network quality of the link, but in the implementation of the existing access scheduling, the network is not easy to perceive when fluctuation occurs, and even if the fluctuation is perceived, operation and maintenance cost is brought.

Existing access scheduling can be broadly divided into two categories: one is scheduling based on static configuration, that is, configuring according to the granularity of regional operators, and scheduling a path accessed by a user to access a corresponding server; the other is dynamic scheduling based on network probing.

For scheduling based on static configuration, due to the lack of network quality data as reference, it is difficult to sense fluctuation in time, and when fluctuation occurs, the configuration can only be modified manually, which has the defects of untimely time and high labor cost.

For dynamic scheduling based on network probing, the network fluctuation is determined according to data obtained by probing, such as packet loss rate.

Specifically, there are two main ways to determine the packet loss rate, one way is to set a fixed threshold, for example, 5%. For a path, the packet loss rate exceeds 5%, and the network fluctuation is considered to occur; the other method is to compare with the historical packet loss rate, for example, if the historical packet loss rate is always about 3%, and the packet loss rate of the path suddenly rises to 10%, it is considered that the path has network fluctuation.

However, for the network fluctuation judgment realized by the packet loss rate, on one hand, the normal packet loss rate ranges are different at different times and different paths, and it is very difficult to set a fixed value in terms of accuracy, for example, the packet loss rate is very different between the peak time and the low time of the network; on the other hand, for the judgment process compared with the historical packet loss rate, since the judgment process is easily influenced by seasonal changes of the network quality, a scene with slowly-degraded network quality is difficult to identify, and the realization cost is large due to the need of storing the historical packet loss rate.

In summary, the existing access scheduling is not able to accurately identify the fluctuation of the access path, and then respond to the fluctuation in time.

Disclosure of Invention

In order to solve the technical problem that the fluctuation in the path cannot be accurately identified in the implementation of access scheduling in the related art, and then the fluctuation is responded in time, the disclosure provides an access scheduling method and device.

A method of access scheduling, the method comprising:

receiving user reported data to obtain network detection data from each user, wherein the network detection data comprises network connection information of the user and a network quality detection value of a path;

aggregating the users according to the network connection information of the users and preset user aggregation granularity to obtain a user group, wherein the aggregated paths accessed by the users form a plurality of paths of the user group, and the user aggregation granularity is fine granularity;

acquiring a reference path in the user group according to the network quality detection value, and identifying a fluctuation path according to the difference value of other paths in the user group relative to the reference path in the network quality detection value;

and scheduling users on the fluctuation path to avoid accessing the fluctuation path.

A method of access scheduling, the method comprising:

performing network detection on all connectable paths to obtain network detection data of a user;

and reporting the network detection data of the user to a server, wherein the network detection data is used for identifying the fluctuation path of the user group where the user is located and the path access scheduling on the fluctuation path.

The technical scheme provided by the embodiment of the disclosure can have the following beneficial effects:

in the network access realized for users through path access, network detection data from each user is obtained by receiving reported data of the users, the network detection data comprises network connection information of the users and network quality detection values of access paths, the users are aggregated according to the network connection information of the users according to preset user aggregation granularity to obtain user groups, the aggregated user access paths form a plurality of paths of the user groups, wherein the user aggregation granularity is fine granularity, a reference path in the user groups is obtained according to the network quality detection values, a fluctuation path is identified according to the difference value of other paths in the user groups relative to the reference path on the network quality detection values, the users on the fluctuation path are scheduled to avoid accessing the fluctuation path, and in the process, the user groups are obtained through the realization of user aggregation based on the fine granularity, according to network detection data reported by users in a user group, an accurate data basis is provided for the identification of path fluctuation in a plurality of paths of the user group, a reference path for identifying the fluctuation path of the user group is determined according to a network quality detection value in the network detection data, the fluctuation path is finally identified according to the difference value of other paths relative to the reference path on the network quality detection value, the identified fluctuation path is separated to carry out wave motion response, the occurrence of fluctuation in the path is accurately and rapidly identified, the fluctuation is timely responded, effective access scheduling is realized for the path access of the users, and the operation and maintenance cost is not brought.

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 invention and together with the description, serve to explain the principles of the invention.

FIG. 1 is a schematic illustration of an implementation environment according to the present disclosure;

FIG. 2 is a block diagram illustrating an apparatus in accordance with an exemplary embodiment;

fig. 3 is a flow diagram illustrating a method of access scheduling in accordance with an exemplary embodiment;

fig. 4 is a flowchart illustrating details of a plurality of path steps for aggregating users according to preset user aggregation granularity to obtain a user group according to network connection information of the users and forming the user group by paths accessed by the aggregated users according to the corresponding embodiment of fig. 3;

fig. 5 is a flowchart illustrating details of a step of obtaining a reference path in a user group according to a network quality detection value and identifying a fluctuation path according to a difference between other paths in the user group and the reference path in the network quality detection value according to the corresponding embodiment of fig. 3;

FIG. 6 is a diagram illustrating a multi-path topology and network quality for a group of users in accordance with an exemplary embodiment;

fig. 7 is a flow chart illustrating a method of access scheduling in accordance with another exemplary embodiment;

fig. 8 is a flow chart illustrating a method of scheduling access in accordance with another exemplary embodiment;

fig. 9 is a flow diagram illustrating a method of access scheduling on a user side in accordance with an exemplary embodiment;

fig. 10 is a block diagram illustrating an access scheduling apparatus in accordance with an example embodiment;

FIG. 11 is a block diagram illustrating details of an aggregation module according to the corresponding embodiment of FIG. 10;

FIG. 12 is a block diagram illustrating details of an identification module according to a corresponding embodiment of FIG. 10;

fig. 13 is a block diagram illustrating an access scheduling apparatus according to another example embodiment;

fig. 14 is a block diagram illustrating an access scheduling apparatus according to another example embodiment;

fig. 15 is a block diagram of an access scheduling apparatus shown on a user side according to an example embodiment.

Detailed Description

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

FIG. 1 is a schematic illustration of an implementation environment according to the present disclosure. The implementation environment includes: the terminal 110 and the scheduling server 130, and the terminal 110 is used as a user side for implementing the service, and accesses a path under the scheduling of the scheduling server 130 to implement the service. Specifically, as shown in fig. 1, the terminal 110 accesses the service server 170 through an access server 150, and in the process, the access server 150 and the service server 170 implement a path accessed by the user side.

Fig. 2 is a block diagram illustrating an apparatus 200 according to an example embodiment. The apparatus 200 may be a database-equipped machine, such as the dispatch server 130 shown in fig. 1, that implements user-side network access scheduling.

Referring to fig. 2, the apparatus 200 may have a relatively large difference due to different configurations or performances, and may include one or more Central Processing Units (CPUs) 222 (e.g., one or more processors) and a memory 232, one or more storage media 230 (e.g., one or more mass storage devices) storing applications 242 or data 244. Memory 232 and storage medium 230 may be, among other things, transient or persistent storage. The program stored in the storage medium 230 may include one or more modules (not shown), each of which may include a series of instruction operations in the apparatus 200. Still further, the central processor 222 may be configured to communicate with the storage medium 230 to execute a series of instruction operations in the storage medium 230 on the device 200. The device 200 may also include one or more power supplies 226, one or more wired or wireless network interfaces 250, one or more input-output interfaces 258, and/or one or more operating systems 241, such as Windows Server, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, and so forth. The steps performed in the embodiments shown in fig. 2, 3 and 4 described above may be based on the device structure shown in fig. 2.

Fig. 3 is a flow chart illustrating a method of access scheduling in accordance with an example embodiment. The access scheduling method is suitable for the scheduling server 130 of the implementation environment shown in fig. 1, and the scheduling server 130 may be the apparatus shown in fig. 2 in an exemplary embodiment. As shown in fig. 3, the access scheduling method, which may be performed by the scheduling server 130, may include the following steps.

In step 310, network probe data from each user is obtained by receiving the user reported data, where the network probe data includes network connection information of the user and a network quality probe value of a path.

Wherein, the user can realize the network access initiated by the user by performing the path access. The requested service is implemented on the subscriber side, for example, by means of a path access made by the subscriber. The user performs network quality detection on all paths that the user can connect to according to a preset time period to obtain corresponding network detection data, and reports the network detection data to the scheduling server 130.

Correspondingly, as the user reports data, the scheduling server 130 receives the data reported by the user. It can be understood that all users capable of working normally report data to the scheduling server 130, and therefore, the scheduling server 130 receives network probe data reported by each user.

For example, a user that continuously reports data to the scheduling server 130 forms a user side of a service, and a service party deploys the service server to provide service for the user, and in addition, the scheduling server 130 is also deployed to access the service server for the user side to implement path scheduling. Therefore, the scheduling server 130 receives the network probe data reported by all users on the user side.

The network connection information of the user is used for indicating attribute information corresponding to the user for realizing the network access, namely, the network connection information describes detailed information of attributes related to the user for realizing the network access. For example, the network connection information of the user includes a gateway address, a physical address, a server access address configured by an operator, and the like.

The network detection data reported by the user to the scheduling server includes the network connection information and also includes the network detection data of the path. As described above, any user will perform network quality detection on all paths that the user can connect to, and therefore, the network quality detection value reported by the user to the scheduling server indicates the network quality of one or more paths. In a specific implementation of an exemplary embodiment, the network quality detection value may be a packet loss rate, a delay, or the like.

It can be understood that, through the continuous network probe data reception in the scheduling server 130, a large amount of network probe data related to the paths reported by the users will be obtained, and thus network quality probe values of a large number of paths are obtained.

In step 330, the users are aggregated according to the network connection information of the users and a preset user aggregation granularity to obtain a user group, and the aggregated user access paths form a plurality of paths of the user group, wherein the user aggregation granularity is a fine granularity.

When reporting the network quality detection value of the path for the scheduling server 130, each user also reports its own network connection information. Different users have different network connection information, so the network connection information can be used as the attribute of the user, and further the user aggregation is realized by taking the network connection information as the basis.

In order to realize user aggregation, user aggregation granularity is configured in advance. The user aggregation granularity is used to indicate the degree of refinement of user aggregation. As can be appreciated, user aggregation is essentially the process of aggregating users having the same or similar attributes together. The finer the user aggregation granularity is, the finer the aggregation for the user is, and conversely, the coarser the user aggregation granularity is, the lower the refinement degree of the aggregation for the user is.

The network connection information comprises a plurality of attribute information corresponding to the network access realized by the user, different attribute information used for user aggregation in the network connection information correspond to different user aggregation granularities, and the refinement degree of the user aggregation granularity is relatively speaking.

For example, for the gateway address and the server access address configured by the operator included in the network connection information, the gateway address is used as a fine granularity, and the server access address configured by the operator is used as a coarse granularity.

In the user aggregation realized according to the network connection information of the users, the users reporting the network detection data are aggregated according to a preset fine granularity, so that a plurality of user groups are obtained from a plurality of users reporting the network detection data, and at the moment, paths accessed by the users included in the user groups form a plurality of paths of the user groups.

Through the realization of the user aggregation of the fine granularity, a plurality of paths conforming to the fine granularity are obtained, and then the subsequent fluctuation path identification and scheduling are realized based on the fine granularity, so that the access scheduling can be realized on the basis of keeping the fine granularity, and the optimal scheduling capability is ensured.

In step 350, a reference path in the user group is obtained according to the network quality detection value, and a fluctuation path is identified according to a difference between other paths in the user group and the reference path in the network quality detection value.

Through the steps, a plurality of user groups are obtained, the fluctuation path is identified for each user group, and then the fluctuation path, namely the fluctuation path, is identified from the paths of each user group.

From the network probing data from each user, network quality probing values obtained by probing the network quality of the user can be obtained, and the network quality probing values correspond to paths, and each path in a plurality of paths of the user group has a corresponding network quality probing value from a plurality of users. Combining the network quality detection values of the paths by a plurality of users, the network quality detection value corresponding to the path can be obtained.

Therefore, for each user group, the identification of the fluctuation paths in the multiple paths can be performed according to the network quality detection value corresponding to each path. The reference path refers to a path in which the network quality detection value indicates the optimal network quality among the multiple paths of the user group. And calculating the difference value of each path and the reference path in the network quality detection value in a plurality of paths of the user group, and identifying the path as a fluctuation path when the difference value exceeds a preset threshold value.

Therefore, the fluctuation in the path is not required to be identified through a fixed threshold value or according to the historical network quality detection value, so that the identification of the fluctuation in the path is suitable for the current overall network condition, the accuracy is effectively improved, and the timeliness of the identification of the fluctuation in the path is also ensured because the identification is realized based on the network quality detection value in the network detection data reported in time, and the fluctuation in the path can be quickly and accurately identified.

In step 370, users on the scheduled fluctuation path avoid accessing the fluctuation path.

After the fluctuation path is identified through the steps, users on the fluctuation path can be mobilized in the user group. Specifically, the users on the fluctuation path are scheduled to intervene in the fluctuation path, so that the users access to the paths without fluctuation in the multiple paths of the user group.

In the process, the identified fluctuation paths are separated, and paths which do not fluctuate in the user group still adopt the original fine granularity, such as optimal scheduling of gateway address granularity, so that users on the fluctuation paths are subjected to scheduling intervention in time, for example, the users are accessed to other paths of the user group, and then the real-time fluctuation intervention and the optimal scheduling are separated, thereby solving the problem of untimely fluctuation response.

Through the process, as the fluctuation path identification is realized by adopting the network detection data obtained by continuous receiving, seasonal influences can be counteracted, and because a large amount of network detection data are reported every moment, the network detection data can quickly and accurately identify which paths have network fluctuation, and automatically intervene users affected by the fluctuation, namely users on the fluctuation paths, on the relatively optimal paths, and avoid the fluctuation paths by automatically implementing intervention on scheduling.

Fig. 4 is a flowchart illustrating details of step 330 according to the corresponding embodiment of fig. 3. With fine granularity being the gateway address, this step 330, as shown in fig. 4, may include the following steps.

In step 331, the gateway addresses corresponding to the users are extracted from the network connection information of the users, respectively.

In the present exemplary embodiment, the information extraction for implementing user aggregation is performed in the network connection information of the user according to a preset user aggregation granularity, and this information extraction is extraction of a gateway address. The gateway address is a network address corresponding to a gateway accessed by a user to realize network access.

Through step 310 in the embodiment corresponding to fig. 3, after the network probe data from each user is continuously received, the network connection information of the user is obtained from the network probe data, and then the gateway address corresponding to the user is extracted from the network connection information.

In step 333, users corresponding to the same gateway address are grouped together to form a user group, and paths accessed by the grouped users form a plurality of paths of the user group.

The users are aggregated based on the fine granularity of the gateway address, and for the dispatch server 130, all the users reporting data to the dispatch server have the gateway addresses corresponding to the users, so that the users can be aggregated based on the gateway addresses to obtain a user group formed by users having the same gateway address, and the access paths of the users having the same gateway address form multiple paths of the user group.

Through the process, a specific implementation is provided for the aggregation of fine granularity of the user, and the identification of the fluctuation path can be identified under the fine granularity.

In an exemplary embodiment, the access scheduling method, as shown in fig. 5, further includes the following steps before step 330.

And controlling to shorten the time interval for executing the user aggregation according to the conversion from the preset fine granularity to the coarse granularity of the user aggregation granularity.

In the embodiments corresponding to fig. 3 and 4, aggregation of users is implemented by presetting fine granularity, that is, the users are divided into the finest parts, and then the paths corresponding to the users are correspondingly implemented, so as to obtain multiple groups of paths for respectively performing fluctuation path recognition, that is, paths corresponding to multiple user groups.

It can be understood that for the user aggregation with the preset fine granularity, a large amount of network probe data, specifically, a user group obtained by aggregation, must be based on, a large amount of network probe data corresponding to the user group, that is, a large amount of samples are collected for the user group in a short time, and network fluctuation is determined in time.

However, in the case that the network fluctuation cannot be recognized in time, for example, in the case that the collected network probe data is insufficient for the user group obtained by fine-granularity aggregation, the user aggregation granularity is changed from the preset fine granularity to the coarse granularity.

The user aggregation process of coarse granularity is performed by converting fine granularity into coarse granularity, so that more samples are brought by sacrificing granularity, namely, the network detection data corresponding to the user group is abundant enough.

According to the granularity conversion, the user aggregation and the identification of the fluctuation path are carried out in a shorter time, and the scheduling is intervened after the fluctuation path is identified.

Fig. 5 is a flowchart illustrating details of step 350 according to a corresponding embodiment of fig. 3. Process 350, as shown in FIG. 5, may include the following steps.

In step 351, among the multiple paths of the user group, the path with the network quality detection value indicating the optimal network quality is obtained as a reference path for the user group to perform path fluctuation identification.

After the user aggregation is completed through step 330 in the embodiment of fig. 3 to obtain the user group and the multiple paths of the user group, a reference path is selected from the multiple paths of the user group.

The reference path is a path with the best network quality among a plurality of paths of the user group. In an exemplary embodiment, the network quality detection value may be a packet loss rate, and a path with the lowest packet loss rate among the multiple paths of the user group is used as a reference path of the user group.

In step 353, the difference between the network quality detection value of the other paths in the user group and the reference path is calculated to obtain the difference corresponding to the other paths in the user group.

In the method, the user group comprises a plurality of paths, wherein the paths comprise other paths besides the reference path, so that the difference between each path in the network quality detection value and the reference path is calculated in the other paths of the user group, and the difference corresponding to each path is obtained.

The calculated difference value indicates the distance of the corresponding path from the reference path in the network quality detection value, and the greater the difference value, the higher the possibility that the corresponding path fluctuates.

It can be understood that the reference path is a path with the best network quality, and the other paths have network quality lower than that of the reference path under normal conditions, but the network quality detection value has only a slight deviation from the reference path and does not have a large deviation, and only when the path fluctuates will have a very large deviation from the reference path, so that whether the path fluctuates or not can be identified by the difference corresponding to each path to become a fluctuating path.

In step 355, it is determined whether the difference exceeds a preset threshold, if so, the path corresponding to the difference is identified as a fluctuation path, and if not, the process is ended.

After calculating the difference value corresponding to each path of the user group except the reference path, judging whether the difference value exceeds a preset threshold value, if so, indicating that the path corresponding to the difference value fluctuates and is a fluctuation path.

Through the process, specific realization is provided for the identification of the fluctuation path in the multiple paths of the user group, so that the scheduling server 130 can timely and accurately identify the fluctuation path along with the continuous network detection data reception.

Fig. 6 is a diagram illustrating a multi-path topology and network quality for a group of users, according to an example embodiment. As described in the embodiment shown in fig. 1, one path of the user group connects the user side and the server requested by the user group, so the path access performed by the users in the user group is substantially the server connection performed by the users, and further, for the user group, the user group is connected to a plurality of servers, as shown in fig. 6, the user group is connected to server a, server B and server C.

In the scenario shown in fig. 6, the network quality of the path is indicated by the packet loss rate, i.e., the smaller the packet loss rate, the better the network quality. As described in the embodiment corresponding to fig. 5, the path having the minimum packet loss rate, which is the path with the optimal network quality, may be used as the reference path, that is, the path where the server B is located.

And if the preset threshold is 5%, the packet loss rate of each path and the reference path, that is, the difference between the minimum packet loss rate of the paths in the user group being 3%, is respectively: the difference value corresponding to the server a is 2%, and the difference value corresponding to the server C is 17% and is greater than the threshold, so that the path from the user group to the server C is the fluctuation path.

Through the above identification process of the fluctuation path, when the network quality of all paths is deteriorated, for example, the packet loss rate reaches 90%, although it indicates that all paths fluctuate, through this process, it is not determined to fluctuate, so as to avoid the execution of subsequent invalid intervention.

That is to say, the identification of the fluctuation path aims to intervene in the fluctuation path, and then, in this scenario, all paths fluctuate, and subsequent effective intervention cannot be realized, so that through the identification process of the fluctuation path, scheduling resources are also saved, and the effectiveness of subsequent intervention is ensured.

Fig. 7 is a flow chart illustrating a method of access scheduling in accordance with another example embodiment. The fluctuating path includes an optimal path when the user group performs path access, as shown in fig. 7, before step 330, the access scheduling method may further include the following steps.

In step 410, an optimal path and a suboptimal path are calculated for the user group according to the network quality detection value.

The network quality detection value of each path in the plurality of paths of the user group identifies the network quality of the path, so that the optimal path and the suboptimal path can be calculated in the plurality of paths of the user group according to the network quality detection values.

The optimal path is necessarily the path with the optimal network quality in a plurality of paths of the user group; and the sub-optimal path, the network quality is inferior to the other paths of the optimal path only.

In step 430, when a user in the user group initiates a path access request, the user is scheduled to access the optimal path, and the optimal path fluctuates after the user accesses.

When a user requests to access a path, the user periodically probes all connectable paths and reports corresponding network quality probe data to the scheduling server 130, as described above, even though the user does not access the path to realize the network access of the user. Therefore, the scheduling server 130 performs user aggregation according to the network quality detection data reported by the scheduling server, and the user aggregation performed necessarily includes aggregation of users requesting an access path.

Correspondingly, the user initiating the path access request has a user group to which the user initiates the path access request through the user aggregation performed by the scheduling server 130 before initiating the path access request. The scheduling server 130 initiates access scheduling for the user initiating the path access request, so that the user initiating the path access request accesses an optimal path in multiple paths of the user group where the user is located, and the optimal path accessed by the user is a path with the optimal network quality when the user accesses the optimal path, but there is a possibility that the user fluctuates after accessing the optimal path.

The fluctuation of the optimal path after the user access is identified accurately and quickly by step 350 in the corresponding embodiment of fig. 3, and is separated by step 370.

Accordingly, step 370 may include the following steps.

And scheduling the user accessing the optimal path to access the suboptimal path again through the optimal path.

It can be understood that, according to the network quality detection values from multiple users corresponding to the paths, an optimal path and a suboptimal path can be obtained through calculation on multiple paths of the user group, and the user is scheduled to access the optimal path, and when the optimal path fluctuates, the user is scheduled to access the suboptimal path, so as to avoid the fluctuating optimal path and ensure that the user can obtain better network quality.

Fig. 8 is a flowchart illustrating a method of scheduling access according to another example embodiment. The fluctuating path also includes a sub-optimal path when the user group performs path access, and before step 370, as shown in fig. 8, the method further includes the following steps.

In step 510, network probing data reported by the users in the current user group is obtained from the received network probing data, and a network quality probing value currently corresponding to the access path of the user group is extracted from the obtained network probing data.

In the embodiment corresponding to fig. 7, the user initiating the path access request accesses the optimal path of the user group, but the optimal path fluctuates after the user accesses, and the suboptimal path of the user group also fluctuates, that is, the optimal path and the suboptimal path both become fluctuating paths.

At this point, the dispatch server 130 needs to intervene. The scheduling server 130 first needs to determine, according to the real-time data, a path with the best network quality among multiple paths of the user group, where the path is the best path re-determined by the scheduling server 130, and then schedules the users on the best path and the suboptimal path to the re-determined best path.

Specifically, as described above, the network probing data received by the scheduling server 130 is reported by each user, so that the network probing data reported by the users in the current user group, that is, the real-time data of the current time node in the network probing data, can be obtained from the received network probing data, where the real-time data is the network quality probing data currently corresponding to the user group access path.

In step 530, a path with the best network quality is selected from the multiple paths of the user group according to the network quality detection value currently corresponding to the user group access path.

As mentioned above, the paths respectively accessed by all users in the user group form multiple paths of the user group, and therefore, the network quality detection value currently corresponding to the user group access path is used to characterize the network quality of the path corresponding to the user group access path among the multiple paths of the user group, and the selection of the path with the best current network quality can be realized based on the network quality detection value.

Accordingly, step 370 may include the following steps.

And scheduling the user accessing the optimal path to re-access the path with the optimal network quality from the optimal path.

After the path with the optimal current network quality in the multiple paths of the user group is selected through the steps, the users on the fluctuation path are dispatched to the path with the optimal current network quality, so that the network fluctuation is responded, and the rapid response to the generated network fluctuation is realized.

In summary, the present disclosure implements path fluctuation identification and user scheduling on a fluctuating path, that is, the scheduling server 130 is implemented to quickly identify and respond to a network fluctuation, for example, for implementing a service, a user may access the service through any path, and the scheduling server 130 provides a guarantee for reliable implementation of the service, that is, once the network fluctuation occurs on a path accessed by the user, the user on the path is quickly scheduled after the fluctuating path is quickly identified, so as to ensure that the service implementation on the user side is not affected by the network fluctuation.

Fig. 9 is a flow chart illustrating a method of access scheduling on a user side according to an example embodiment. The access scheduling method, as shown in fig. 9, may include the following steps.

In step 610, network probing is performed on all connectable paths to obtain network probing data of the user.

At the user side, each user performs network detection on all connectable paths to obtain corresponding network detection data. One side of the path is a certain user or a certain user set, for example, a set with granularity of network address, network address segment (IPC), provincial operator, and the like; the other side of the path may be a network address + port of a certain server, or a set of an IP address + port of the server, that is, a certain machine room, or an address where the server is located.

Based on this, the network probing performed by the user on all connectable paths refers to probing the network quality on all connectable servers. In an exemplary embodiment, the network probing performed by the users is performed at preset time intervals, so that all users probe each server that may be connected at regular intervals.

In step 630, the network probe data of the user is reported to the server, where the network probe data is used to identify the fluctuation path of the user group where the user is located and the path access scheduling on the fluctuation path.

After completing the detection of all connectable servers, the user reports the network detection data to the server, that is, the scheduling server 130, and the process is implemented on the user side, for the scheduling server 130, the network detection data is reported at all times macroscopically, and network detection is implemented microscopically at intervals of a period of time by each user, so as to avoid excessive traffic consumption, but since a large amount of network detection data is reported at all times, the scheduling server 130 can identify the occurrence of fluctuation on the path based on the network detection data and automatically intervene the user affected by the fluctuation on the relatively optimal path.

Through the process, the method for rapidly sensing the network fluctuation is realized, so that the method can automatically act on an access scheduling mechanism to avoid a fluctuation path, the realization difficulty and effect are optimized, and the realization is better than the realization of the conventional access scheduling.

The following is an embodiment of the apparatus of the present disclosure, which may be used to execute an embodiment of an access scheduling method executed by the scheduling server 130 of the present disclosure. For details not disclosed in the embodiments of the apparatus of the present disclosure, please refer to the embodiments of the access scheduling method of the present disclosure.

Fig. 10 is a block diagram illustrating an access scheduling apparatus according to an example embodiment. The block diagram of the access scheduling apparatus can be used in the implementation environment shown in fig. 1 to execute all the steps of the access scheduling method shown in fig. 10. As shown in fig. 10, the access scheduling device includes but is not limited to: a receiving module 710, an aggregation module 730, an identification module 750, and a scheduling module 770.

The receiving module 710 is configured to obtain network probe data from each user by receiving user reported data, where the network probe data includes network connection information of the user and a network quality probe value of a path.

The aggregation module 730 is configured to aggregate users according to the network connection information of the users and a preset user aggregation granularity to obtain a user group, where paths accessed by the aggregated users form multiple paths of the user group, and the user aggregation granularity is a fine granularity.

And an identifying module 750, configured to obtain a reference path in the user group according to the network quality detection value, and identify a fluctuation path according to a difference between other paths in the user group and the reference path in the network quality detection value.

And a scheduling module 770, configured to schedule users on the fluctuation path to avoid accessing the fluctuation path.

By the access scheduling device, a scheduling server is realized for network access of a user, and the user can not be interfered by network fluctuation when realizing services provided by a service party.

Fig. 11 is a block diagram illustrating details of an aggregation module according to the corresponding embodiment of fig. 10. The fine granularity is the gateway address, and the aggregation module 730, as shown in fig. 11, may include but is not limited to: an address extraction unit 731, and an aggregation execution unit 733.

An address extracting unit 731, configured to extract gateway addresses corresponding to the users from the network connection information of the users, respectively.

The aggregation performing unit 733 is configured to aggregate users corresponding to the same gateway address to form a user group, and paths accessed by the aggregated users form a plurality of paths of the user group.

By the aggregation module 730, all users are aggregated based on the gateway address, which is a fine-grained aggregation, so that the fine-grained fluctuation path identification can be realized at the scheduling server 130.

In another exemplary embodiment, the access scheduling apparatus further includes an aggregation control module, and the aggregation control module is configured to control to shorten a time interval for performing the user aggregation according to a conversion of the user aggregation granularity from a preset fine granularity to a coarse granularity.

Fig. 12 is a block diagram illustrating details of an identification module according to the corresponding embodiment of fig. 10. The identification module 750, as shown in fig. 12, may include, but is not limited to: a reference path selecting unit 751, a difference value calculating unit 753, and a judging unit 755.

The reference path selecting unit 751 is configured to obtain, among multiple paths of the user group, a path with the network quality detection value indicating that the network quality is optimal as a reference path for performing path fluctuation identification for the user group.

A difference calculation unit 753, configured to calculate differences between the network quality detection values of other paths in the user group and the reference path, so as to obtain differences corresponding to the other paths in the user group.

The determining unit 755 is configured to determine whether the difference exceeds a preset threshold, and if so, identify a path corresponding to the difference as a fluctuation path.

The configuration of the identifying module 750 enables the identification of the fluctuating path to be implemented based on the path with the best network quality among the multiple paths of the user group, that is, the fluctuating path is identified according to the difference between the network quality detection values of other paths and the path by taking the network quality detection value of the path as a reference.

The method can offset seasonal influence, does not have the defects of the existing method, and ensures that the carried out fluctuation path identification is adaptive to the current network quality of the user group.

Fig. 13 is a block diagram illustrating an access scheduling apparatus according to another exemplary embodiment. The fluctuating path includes an optimal path when the user performs path access, and as shown in fig. 13, the access scheduling apparatus may further include, but is not limited to: a path computation module 810 and an optimal path scheduling module 830.

And a path calculating module 810, configured to calculate an optimal path and a suboptimal path for the user group according to the network quality detection value.

The optimal path scheduling module 830 is configured to schedule a user to access an optimal path when the user in the user group initiates a path access request, where the optimal path fluctuates after the user accesses.

Correspondingly, the scheduling module 770 is further configured to schedule the user accessing the optimal path to re-access the sub-optimal path from the optimal path.

Fig. 14 is a block diagram illustrating an access scheduling apparatus according to another exemplary embodiment. The fluctuating path further includes a sub-optimal path when the user group performs path access, and the access scheduling apparatus, as shown in fig. 14, further includes but is not limited to: a real-time data acquisition module 910 and a real-time selection module 930.

The real-time data obtaining module 910 is configured to obtain network probe data reported by users in a current user group from the received network probe data, and extract a network quality detection value currently corresponding to an access path of the user group from the obtained network probe data.

The real-time selecting module 930 is configured to select a path with the best current network quality from the multiple paths of the user group according to the network quality detection value currently corresponding to the user group access path.

The scheduling module 770 is further configured to schedule the user accessing the optimal path to re-access the path with the best network quality from the optimal path.

Fig. 15 is a block diagram of an access scheduling apparatus shown on a user side according to an example embodiment. As shown in fig. 15, the access scheduling apparatus includes, but is not limited to: a network detection module 1010 and a data reporting module 1030.

The network probing module 1010 is configured to perform network probing on all connectable paths to obtain network probing data of the user.

A data reporting module 1030, configured to report network probe data of a user to a server, where the network probe data is used to identify a fluctuation path of a user group where the user is located and a path access schedule on the fluctuation path.

In another exemplary embodiment, the access scheduling apparatus further includes a scheduling response module. The scheduling response module is used for responding to the user scheduling on the fluctuation path of the server in the network access of the user through the access path, and the user avoids the path when the accessed path fluctuates.

Optionally, the present disclosure further provides a scheduling server, which may be used in the implementation environment shown in fig. 1 to execute all or part of the steps of the access scheduling method shown in any one of fig. 3, fig. 4, fig. 5, fig. 7, and fig. 8. The device comprises:

a processor;

a memory for storing processor-executable instructions;

wherein the processor is configured to perform:

The specific manner in which the processor of the apparatus in this embodiment performs the operation has been described in detail in the embodiment related to the access scheduling method, and will not be elaborated here.

Optionally, the present disclosure further provides a terminal, where the terminal may be used in the implementation environment shown in fig. 1 to execute all or part of the steps of the access scheduling method shown in any one of fig. 9. The device comprises:

a processor;

a memory for storing processor-executable instructions;

wherein the processor is configured to perform:

It will be understood that the invention 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 invention is limited only by the appended claims.

Claims

1. An access scheduling method, the method comprising:

acquiring a reference path in the user group according to the network quality detection value, and identifying a fluctuation path according to the difference of other paths in the user group relative to the reference path in the network quality detection value, wherein the reference path is a path with the optimal network quality indicated by the network quality detection value in a plurality of paths of the user group;

2. The method of claim 1, wherein the fine granularity is a gateway address, and wherein aggregating the users according to the network connection information of the users and a preset user aggregation granularity to obtain a user group, and paths accessed by the users aggregated together form multiple paths of the user group, includes:

respectively extracting gateway addresses corresponding to the users from the network connection information of the users;

and aggregating the users corresponding to the same gateway address to form a user group, wherein the paths accessed by the users aggregated together form a plurality of paths of the user group.

3. The method according to claim 1, wherein before the aggregating the users according to the network connection information of the users according to a preset user aggregation granularity to obtain a user group, and paths accessed by the users aggregated together form a plurality of paths of the user group, the method further comprises:

and controlling to shorten the time interval for executing the user aggregation according to the conversion of the user aggregation granularity from the fine granularity to the coarse granularity.

4. The method according to claim 1, wherein the obtaining a reference path in the user group according to the network quality detection value and identifying a fluctuation path according to a difference between other paths in the user group and the reference path in the network quality detection value comprises:

obtaining a path with the network quality detection value indicating the optimal network quality from a plurality of paths of the user group as a reference path for performing path fluctuation identification on the user group;

calculating the difference between the network quality detection value and the reference path of other paths in the user group to obtain the difference corresponding to the other paths in the user group;

and judging whether the difference value exceeds a preset threshold value, and if so, identifying the path corresponding to the difference value as a fluctuation path.

5. The method according to claim 1, wherein the fluctuating path includes an optimal path when the user group performs path access, the user group is obtained by aggregating the users according to the network connection information of the users and according to a preset user aggregation granularity, and before the aggregated paths accessed by the users form a plurality of paths of the user group, the method further includes:

calculating an optimal path and a suboptimal path for the user group according to the network quality detection value;

and when a user in the user group initiates a path access request, scheduling the user to access the optimal path, wherein the optimal path fluctuates after the user accesses.

6. The method of claim 5, wherein the scheduling users on the fluctuating path to avoid accessing the fluctuating path comprises:

and scheduling the users accessing the optimal path to access a suboptimal path again through the optimal path.

7. The method of claim 5, wherein the fluctuating path further comprises a suboptimal path when the user group performs path access, and before the user on the fluctuating path is scheduled to avoid accessing the fluctuating path, the method further comprises:

acquiring network detection data reported by users in the user group from the received network detection data, and extracting a network quality detection value corresponding to the user group access path from the acquired network detection data;

selecting a path with the optimal current network quality from a plurality of paths of the user group according to the network quality detection value currently corresponding to the user group access path;

the scheduling the users on the fluctuation path to avoid accessing the fluctuation path comprises:

8. An access scheduling method, the method comprising:

performing network detection on all connectable paths to obtain network detection data of a user, wherein the network detection data comprises network connection information of the user and a network quality detection value of the path;

and reporting network detection data of the user to a server, wherein the network detection data is used for identifying a fluctuation path of a user group where the user is located and path access scheduling on the fluctuation path, so that the server aggregates the user according to network connection information of the user and preset user aggregation granularity to obtain the user group, the aggregated paths accessed by the user form a plurality of paths of the user group, the user aggregation granularity is fine granularity, the server obtains a path with the optimal network quality indicated by a network quality detection value in the plurality of paths in the user group according to the network quality detection value, and identifies the fluctuation path according to a difference value of other paths in the user group relative to the network quality detection value in the plurality of paths indicated by the network quality detection value to indicate the optimal network quality on the network quality detection value.

9. The method of claim 8, further comprising:

in the network access of the user through the access path, responding to the user scheduling on the fluctuation path of the server, the user avoids the path when the accessed path fluctuates.

10. An access scheduling apparatus, the apparatus comprising:

the receiving module is used for receiving the reported data of the users to obtain network detection data from each user, wherein the network detection data comprises network connection information of the users and a network quality detection value of a path;

the aggregation module is used for aggregating users according to the network connection information of the users and preset user aggregation granularity to obtain a user group, the aggregated paths accessed by the users form a plurality of paths of the user group, and the user aggregation granularity is fine granularity;

the identification module is used for obtaining a reference path in the user group according to the network quality detection value and identifying a fluctuation path according to the difference value of other paths in the user group relative to the reference path in the network quality detection value, wherein the reference path is a path with the optimal network quality indicated by the network quality detection value in a plurality of paths of the user group;

and the scheduling module is used for scheduling the users on the fluctuation path to avoid accessing the fluctuation path.

11. The apparatus of claim 10, wherein the fine granularity is a gateway address, and wherein the aggregation module comprises:

an address extracting unit, configured to extract gateway addresses corresponding to the users from the network connection information of the users, respectively;

and the aggregation execution unit is used for aggregating the users corresponding to the same gateway address to form a user group, and the paths accessed by the users which are aggregated form a plurality of paths of the user group.

12. The apparatus of claim 10, further comprising:

and the aggregation control module is used for controlling and shortening the time interval for executing the user aggregation according to the conversion from the fine granularity to the coarse granularity of the user aggregation granularity.

13. The apparatus of claim 10, wherein the identification module comprises:

a reference path selecting unit, configured to obtain, among multiple paths of the user group, a path with a network quality detection value indicating that the network quality is optimal as a reference path for performing path fluctuation identification on the user group;

a difference value calculating unit, configured to calculate difference values between other paths in the user group and a reference path in the network quality detection, so as to obtain difference values corresponding to the other paths in the user group;

and the judging unit is used for judging whether the difference value exceeds a preset threshold value, and if so, identifying the path corresponding to the difference value as a fluctuation path.

14. The apparatus of claim 10, wherein the fluctuation path comprises an optimal path for the user group to access, the apparatus further comprising:

a path calculation module for calculating an optimal path and a suboptimal path for the user group according to the network quality detection value

Diameter;

and the optimal path scheduling module is used for scheduling the user to access the optimal path when the user in the user group initiates a path access request, wherein the optimal path fluctuates after the user accesses.

15. The apparatus of claim 14, wherein the scheduling module is further configured to schedule users accessing the optimal path to re-access a sub-optimal path from the optimal path.

16. The apparatus of claim 14, wherein the fluctuating path further comprises a sub-optimal path for path access by a user group, the apparatus further comprising:

the real-time data acquisition module is used for acquiring the network detection data reported by the users in the current user group from the received network detection data and extracting the network quality detection value corresponding to the access path of the user group from the acquired network detection data;

a real-time selection module, configured to select a path with the best current network quality from multiple paths of the user group according to a network quality detection value currently corresponding to the user group access path;

the scheduling module is further used for scheduling the user accessing the optimal path to re-access the path with the optimal network quality from the optimal path.

17. An access scheduling apparatus, the apparatus comprising:

the network detection module is used for carrying out network detection on all connectable paths to obtain network detection data of a user, wherein the network detection data comprises network connection information of the user and a network quality detection value of the path;

a data reporting module, configured to report network probe data of the user to a server, where the network probe data is used to identify a fluctuation path of a user group where the user is located and a path access schedule on the fluctuation path, so that the server aggregates the users according to the network connection information of the users and preset user aggregation granularity to obtain a user group, the aggregated paths accessed by the users form a plurality of paths of the user group, the user aggregation granularity is fine granularity, and the server obtains a path with the network quality detection value indicating the optimal network quality from the plurality of paths in the user group according to the network quality detection value, and identifying a fluctuation path according to the difference of the network quality detection values of other paths in the user group relative to the path with the optimal network quality indicated by the network quality detection values in the multiple paths.

18. The apparatus of claim 17, further comprising:

and the scheduling response module is used for responding to the user scheduling on the fluctuation path of the server in the network access of the user through the access path, and the user avoids the path when the accessed path fluctuates.

19. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the access scheduling method according to any one of claims 1 to 9.

20. An electronic device, comprising:

one or more processors;

storage means for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to implement the access scheduling method of any of claims 1 to 9.