Edge node scheduling method and device, storage medium and electronic equipment

文档序号：1966035 发布日期：2021-12-14 浏览：25次中文

阅读说明：本技术 边缘节点调度方法及装置、存储介质、电子设备 (Edge node scheduling method and device, storage medium and electronic equipment ) 是由邱昊鄢智勇唐莉莉徐教强高丽华吴林江陈文华李林倩颜雪于 2021-09-15 设计创作，主要内容包括：本公开属于通信技术领域,涉及一种边缘节点调度方法及装置、存储介质、电子设备。该方法包括：获取用户终端发送的域名请求,并对域名请求进行域名解析处理得到请求地址信息；获取路由调度策略,并对请求地址信息和路由调度策略进行策略匹配处理得到策略匹配结果；根据策略匹配结果调度与用户终端对应的缓存服务器,以使用户终端与缓存服务器进行数据交互。本公开无需借助提前导入的IP库或外围设备嗅探即可实时筛选更新,为精准调度提供了数据基础,并且,路由调度策略细粒度划分,使得调度过程不会受制于频次和物理位置,以此为最优调度提供数据支持,根据策略匹配结果进行调度,实现节点调度全面覆盖、实时准确和精准最优且最短的效果。(The disclosure belongs to the technical field of communication, and relates to a method and a device for scheduling edge nodes, a storage medium and electronic equipment. The method comprises the following steps: acquiring a domain name request sent by a user terminal, and performing domain name resolution processing on the domain name request to obtain request address information; obtaining a routing scheduling strategy, and performing strategy matching processing on the request address information and the routing scheduling strategy to obtain a strategy matching result; and scheduling the cache server corresponding to the user terminal according to the strategy matching result so as to enable the user terminal and the cache server to carry out data interaction. According to the method, the updating can be screened in real time without sniffing by means of an IP library or peripheral equipment which is imported in advance, a data basis is provided for accurate scheduling, the fine granularity of the routing scheduling strategy is divided, so that the scheduling process is not limited by frequency and physical positions, data support is provided for optimal scheduling, scheduling is carried out according to strategy matching results, and the effects of comprehensive coverage, real-time accuracy, optimization and shortest node scheduling are achieved.)

1. An edge node scheduling method, the method comprising:

acquiring a domain name request sent by a user terminal, and performing domain name resolution processing on the domain name request to obtain request address information;

obtaining a routing scheduling strategy, and performing strategy matching processing on the request address information and the routing scheduling strategy to obtain a strategy matching result;

and scheduling a cache server corresponding to the user terminal according to the strategy matching result so as to enable the user terminal to perform data interaction with the cache server.

2. The edge node scheduling method of claim 1, wherein the obtaining the routing scheduling policy comprises:

obtaining route detail information corresponding to the user terminal through a link directly connected with a routing device, and performing information elimination processing on the route detail information to obtain target route information;

and acquiring a server label of a cache server corresponding to the user terminal, and performing policy generation processing on the target routing information and the server label to obtain a routing scheduling policy.

3. The edge node scheduling method according to claim 2, wherein the performing information elimination processing on the routing detail information to obtain target routing information includes:

performing invalid elimination processing on the routing detail information to obtain effective routing information;

and performing detail combination processing on the effective routing information to obtain target routing information.

4. The edge node scheduling method according to claim 3, wherein the performing invalid elimination processing on the route detail information to obtain valid route information includes:

performing invalid elimination processing on the directly-connected routing information in the routing detail information to obtain first routing information;

and carrying out invalid elimination processing on the private network routing information in the first routing information to obtain valid routing information.

5. The edge node scheduling method according to claim 3, wherein the performing detail merging processing on the effective routing information to obtain target routing information includes:

obtaining mask information in the effective routing information, and obtaining mask conditions corresponding to the mask information;

and performing mask matching processing on the mask information and the mask condition to obtain a matching processing result, and performing detail merging processing according to the matching processing result to obtain target routing information.

6. The edge node scheduling method of claim 2, wherein the routing detail information comprises full routing information and modified routing information,

the obtaining of the route detail information corresponding to the user terminal through the link directly connected to the routing device includes:

acquiring the total routing information corresponding to the user terminal through a link directly connected with the routing equipment; or

And acquiring modified routing information corresponding to the user terminal through a link directly connected with the routing equipment.

7. The edge node scheduling method of claim 1, wherein the method further comprises:

and when the strategy matching result is not matched with the cache server, scheduling a default server corresponding to the user terminal according to a default strategy corresponding to the strategy matching result so as to enable the user terminal to perform data interaction with the default server.

8. An edge node scheduling apparatus, comprising:

the domain name resolution module is configured to acquire a domain name request sent by a user terminal, and perform domain name resolution processing on the domain name request to obtain request address information;

the strategy matching module is configured to acquire a routing scheduling strategy and perform strategy matching processing on the request address information and the routing scheduling strategy to obtain a strategy matching result;

and the node scheduling module is configured to schedule a cache server corresponding to the user terminal according to the strategy matching result so as to enable the user terminal to perform data interaction with the cache server.

9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the edge node scheduling method according to any one of claims 1 to 7.

10. An electronic device, comprising:

a processor;

a memory for storing executable instructions of the processor;

wherein the processor is configured to perform the edge node scheduling method of any of claims 1-7 via execution of the executable instructions.

Technical Field

The present disclosure relates to the field of communications technologies, and in particular, to an edge node scheduling method, an edge node scheduling apparatus, a computer-readable storage medium, and an electronic device.

Background

The edge computing supports various access modes such as broadband, wifi, (the 4th Generation Mobile Communication Technology, fourth Generation Mobile Communication Technology) or 5G (5th Generation Mobile Communication Technology, fifth Generation Mobile Communication Technology), and aims to provide a low-latency, large-connection, high-bandwidth nearby service, improve the service experience of a user, reduce the bandwidth cost of a backbone network, and provide possibility for service convergence of a centralized cloud and high-computing power service migration of a client. And a Content Delivery Network (CDN) is a traditional service, and a large number of nodes reach the edge of the Network directly, and has the characteristics of large connection and large bandwidth. However, the low-latency and high-computation-power services carried by the edge computing nodes have high requirements for end-to-end latency, and if traditional CDN scheduling based on a Domain name resolution (DNS) Domain name is adopted, shortest path optimization is not achieved in combination with a network topology of each edge computing node, and routing changes of a client cannot be sensed in real time, so that accurate scheduling is achieved.

Currently, the scheduling of the edge node is mostly implemented based on DNS domain name scheduling of CDN services, and mainly combines a preset IP (Internet Protocol Address) Address library or real-time dial-test two directions. Although the preset IP address library is accurate, the routing of the metro network client frequently changes, and accurate scheduling cannot be performed on the client IP with added and deleted changes; the real-time dial testing requires a certain number of terminals to be deployed for simulation, and is limited by frequency and physical position, so that optimal scheduling of all client requests can not be guaranteed.

In view of this, there is a need in the art to develop a new edge node scheduling method and apparatus.

It is to be noted that the information disclosed in the above background section is only for enhancement of understanding of the background of the present disclosure, and thus may include information that does not constitute prior art known to those of ordinary skill in the art.

Disclosure of Invention

The present disclosure is directed to providing an edge node scheduling method, an edge node scheduling apparatus, a computer-readable storage medium, and an electronic device, so as to overcome the technical problems of low scheduling accuracy and poor real-time performance due to the limitations of the related art at least to some extent.

Additional features and advantages of the disclosure will be set forth in the detailed description which follows, or in part will be obvious from the description, or may be learned by practice of the disclosure.

According to a first aspect of the embodiments of the present invention, there is provided a method for scheduling an edge node, the method including: acquiring a domain name request sent by a user terminal, and performing domain name resolution processing on the domain name request to obtain request address information;

obtaining a routing scheduling strategy, and performing strategy matching processing on the request address information and the routing scheduling strategy to obtain a strategy matching result;

and scheduling a cache server corresponding to the user terminal according to the strategy matching result so as to enable the user terminal to perform data interaction with the cache server.

In an exemplary embodiment of the present invention, the obtaining the routing scheduling policy includes:

In an exemplary embodiment of the present invention, the performing information elimination processing on the route detail information to obtain target route information includes:

performing invalid elimination processing on the routing detail information to obtain effective routing information;

and performing detail combination processing on the effective routing information to obtain target routing information.

In an exemplary embodiment of the present invention, the performing invalid elimination processing on the route detail information to obtain valid route information includes:

performing invalid elimination processing on the directly-connected routing information in the routing detail information to obtain first routing information;

and carrying out invalid elimination processing on the private network routing information in the first routing information to obtain valid routing information.

In an exemplary embodiment of the present invention, the performing detail merging processing on the effective routing information to obtain target routing information includes:

obtaining mask information in the effective routing information, and obtaining mask conditions corresponding to the mask information;

In an exemplary embodiment of the invention, the route detail information includes a full amount of route information and modified route information,

the obtaining of the route detail information corresponding to the user terminal through the link directly connected to the routing device includes:

acquiring the total routing information corresponding to the user terminal through a link directly connected with the routing equipment; or

And acquiring modified routing information corresponding to the user terminal through a link directly connected with the routing equipment.

In an exemplary embodiment of the invention, the method further comprises:

According to a second aspect of the embodiments of the present invention, there is provided an edge node scheduling apparatus, including: the domain name resolution module is configured to acquire a domain name request sent by a user terminal, and perform domain name resolution processing on the domain name request to obtain request address information;

According to a third aspect of embodiments of the present invention, there is provided an electronic apparatus including: a processor and a memory; wherein the memory has stored thereon computer readable instructions, which when executed by the processor, implement the edge node scheduling method in any of the above exemplary embodiments.

According to a fourth aspect of embodiments of the present invention, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the edge node scheduling method in any of the above-described exemplary embodiments.

As can be seen from the foregoing technical solutions, the edge node scheduling method, the edge node scheduling apparatus, the computer storage medium and the electronic device in the exemplary embodiments of the present disclosure have at least the following advantages and positive effects:

in the method and the device provided by the exemplary embodiment of the disclosure, the obtained routing scheduling policy can be screened and updated in real time without sniffing by an IP library or peripheral equipment which is imported in advance, so that a data basis is provided for accurate scheduling, and the routing scheduling policy is divided in a fine granularity manner, so that the scheduling process is not limited by frequency and a physical position, and data support is provided for optimal scheduling. Furthermore, the strategy matching result obtained by strategy matching processing is scheduled according to the request address information and the routing strategy scheduling, so that the effects of comprehensive coverage, real-time accuracy, optimality and shortest node scheduling are achieved.

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

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and together with the description, serve to explain the principles of the disclosure. It is to be understood that the drawings in the following description are merely exemplary of the disclosure, and that other drawings may be derived from those drawings by one of ordinary skill in the art without the exercise of inventive faculty.

Fig. 1 schematically illustrates a flowchart of an edge node scheduling method in an exemplary embodiment of the present disclosure;

fig. 2 schematically illustrates a flowchart of a method for obtaining a routing scheduling policy in an exemplary embodiment of the present disclosure;

FIG. 3 is a flow diagram schematically illustrating a method of obtaining full-scale routing information and modifying routing information in an exemplary embodiment of the disclosure;

FIG. 4 schematically illustrates a flow diagram of a method of information culling processing in an exemplary embodiment of the disclosure;

FIG. 5 schematically illustrates a flow diagram of a method of invalidation culling processing in an exemplary embodiment of the disclosure;

FIG. 6 schematically illustrates a flow diagram of a method of detail merge processing in an exemplary embodiment of the disclosure;

FIG. 7 is a system block diagram schematically illustrating edge node scheduling under an application scenario in an exemplary embodiment of the present disclosure;

fig. 8 is a schematic flow chart illustrating a method for generating a routing scheduling policy in an application scenario in an exemplary embodiment of the present disclosure;

FIG. 9 is a flow chart diagram schematically illustrating a method for implementing data access in an application scenario in an exemplary embodiment of the present disclosure;

FIG. 10 is a diagram schematically illustrating a system architecture for edge node scheduling in an application scenario in an exemplary embodiment of the present disclosure;

fig. 11 is a schematic structural diagram of an edge node scheduling apparatus in an exemplary embodiment of the present disclosure;

fig. 12 schematically illustrates an electronic device for implementing an edge node scheduling method in an exemplary embodiment of the present disclosure;

fig. 13 schematically illustrates a computer-readable storage medium for implementing an edge node scheduling method in an exemplary embodiment of the present disclosure.

Detailed Description

Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in many different forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the disclosure. One skilled in the relevant art will recognize, however, that the subject matter of the present disclosure can be practiced without one or more of the specific details, or with other methods, components, devices, steps, and the like. In other instances, well-known technical solutions have not been shown or described in detail to avoid obscuring aspects of the present disclosure.

The terms "a," "an," "the," and "said" are used in this specification to denote the presence of one or more elements/components/parts/etc.; the terms "comprising" and "having" are intended to be inclusive and mean that there may be additional elements/components/etc. other than the listed elements/components/etc.; the terms "first" and "second", etc. are used merely as labels, and are not limiting on the number of their objects.

Furthermore, the drawings are merely schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus their repetitive description will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities.

The edge computing supports various access modes such as broadband, wifi, 4G/5G and the like, aims to provide a low-time-delay, large-connection and high-bandwidth nearby service, improves the service experience of a user, reduces the bandwidth cost of a backbone network, and simultaneously provides possibility for the service convergence of a centralized cloud and the high-computation-power service upward movement of a client. The CDN is a traditional service, mass nodes directly reach the edge of a network, and the CDN has the characteristics of large connection and large bandwidth. However, the low-latency and high-computation-power services carried by the edge computing nodes have high requirements for end-to-end latency, and if traditional CDN scheduling based on the DNS domain name is adopted, shortest path optimization is not achieved in combination with a network topology of each edge computing node, and routing changes of the client cannot be sensed in real time, so that accurate scheduling is achieved.

The current edge node scheduling is mostly realized based on the Domain Name System (DNS) scheduling of CDN services, and is mainly combined with a preset IP address library or real-time dial-test in two directions. Although the preset IP address library is accurate, the routing of the metro network client frequently changes, and accurate scheduling cannot be performed on the client IP with added and deleted changes; the real-time dial testing requires a certain number of terminals to be deployed for simulation, and is limited by frequency and physical position, so that optimal scheduling of all client requests can not be guaranteed.

In order to improve the accuracy, the traditional scheduling system determines the nodes served nearby according to the pre-stored IP section information or dial test results, the granularity is coarse, and the nodes are generally provincial or city dimensions.

In the 5G era, the low-delay services such as live broadcast, AR, VR, etc. require that the network delay from the client to the server is controlled within 10ms, and the end-to-end delay is within 50ms, so that the smooth experience after the high-computing power service moves up to the edge node can be ensured. The traditional scheduling can only ensure low time delay and short path in local and quasi-real time after optimization, cannot ensure wide coverage and provide scheduling capability of low time delay in real time, and is represented by increased hop count and unstable time delay. The services such as live broadcast and the like carried by the edge computing node are often presented in the form of APP (Application, mobile phone software), and have different access modes such as wifi, 4G/5G and the like. In order to ensure the consistency of service experience, the edge node is required to meet the requirement of low delay under various access modes.

For the problems in the related art, the present disclosure provides an edge node scheduling method, and fig. 1 shows a flowchart of the edge node scheduling method, and as shown in fig. 1, the edge node scheduling method at least includes the following steps:

step S110, a domain name request sent by the user terminal is obtained, and domain name resolution processing is carried out on the domain name request to obtain request address information.

And S120, acquiring a routing scheduling strategy, and performing strategy matching processing on the request address information and the routing scheduling strategy to obtain a strategy matching result.

And S130, scheduling a cache server corresponding to the user terminal according to the strategy matching result so as to enable the user terminal and the cache server to perform data interaction.

In the exemplary embodiment of the disclosure, the obtained routing scheduling policy can be screened and updated in real time without sniffing by an IP library or peripheral equipment which is imported in advance, so that a data basis is provided for accurate scheduling, and the routing scheduling policy is divided in a fine granularity manner, so that the scheduling process is not limited by frequency and physical positions, thereby providing data support for optimal scheduling. Furthermore, the strategy matching result obtained by strategy matching processing is scheduled according to the request address information and the routing strategy scheduling, so that the effects of comprehensive coverage, real-time accuracy, optimality and shortest node scheduling are achieved.

The following describes each step of the edge node scheduling method in detail.

In step S110, a domain name request sent by the user terminal is obtained, and domain name resolution processing is performed on the domain name request to obtain request address information.

In an exemplary embodiment of the present disclosure, a user terminal may initiate a domain name request through a client. In addition, the client does not need to be modified, as long as the client can normally initiate the domain name request.

After receiving the domain name request, the domain name resolution process may be performed on the domain name request.

Domain name resolution is a service that directs domain names to a website space IP, allowing people to conveniently access the website through the registered domain names. The IP address is a digital address for identifying a station on the network, and a domain name is adopted to replace the IP address to identify the station address for the convenience of memory. Domain name resolution is the process of converting a domain name to an IP address. The resolution work of the domain name is completed by the DNS server.

Domain name resolution is also called domain name pointing, server setting, domain name configuration, reverse IP registration, etc. Briefly, a well-documented domain name is resolved to an IP, and the service is performed by a DNS server, which resolves the domain name to an IP address and then binds a subdirectory to the domain name on the host of the IP address.

The addresses in the internet are digital IP addresses, and the function of domain name resolution is mainly to facilitate memory.

Therefore, the request address information corresponding to the domain name request, i.e. the client IP address, can be obtained through the domain name resolution process.

In step S120, a routing scheduling policy is obtained, and policy matching processing is performed on the request address information and the routing scheduling policy to obtain a policy matching result.

In an exemplary embodiment of the present disclosure, further, a routing scheduling policy may also be obtained.

In an alternative embodiment, fig. 2 is a flowchart illustrating a method for obtaining a routing scheduling policy, where as shown in fig. 2, the method at least includes the following steps: in step S210, route detail information corresponding to the user terminal is obtained through a link directly connected to the routing device, and information elimination processing is performed on the route detail information to obtain target route information.

The routing detail information comprises full routing information and modified routing information. In an alternative embodiment, fig. 3 is a flowchart illustrating a method for obtaining full-volume routing information and modifying the routing information, where as shown in fig. 3, the method at least includes the following steps: in step S310, the total amount of routing information corresponding to the user terminal is acquired through the link directly connected to the routing device.

The full amount of routing information may be synchronized through links directly connected to the routing devices. The routing device may be a BRAS (Broadband Access Server) or other devices, which is not limited in this exemplary embodiment.

And, the full amount of routing information may be BGP (Border Gateway Protocol) routing information.

It should be noted that the full amount of routing information is only needed when the detailed routing information is first obtained.

In step S320, modified routing information corresponding to the user terminal is obtained through a link directly connected to the routing device.

The routing information can be synchronously modified through a link directly connected with the routing equipment. The routing device may also be a BRAS, or may also be another device, which is not particularly limited in this exemplary embodiment.

When the modified routing information is BGP routing information, the modified routing information is triggered to be sent and received only after the BGP routing of the routing client is changed. Wherein, the modified route information may include added, deleted and modified route entry information.

In the exemplary embodiment, the full amount of routing information or the modified routing information corresponding to the user terminal can be acquired through a link directly connected with the routing device, the change of the client routing on the network is monitored in real time and synchronized to the scheduling system, and a data base and support are provided for really realizing full coverage and real-time accurate node scheduling without sniffing by an IP library or peripheral devices which are imported in advance.

After the route detail information is acquired, information elimination processing may be performed on the route detail information.

In an alternative embodiment, fig. 4 shows a flowchart of a method for information culling processing, and as shown in fig. 4, the method at least includes the following steps: in step S410, invalid culling processing is performed on the route detail information to obtain valid route information.

In an alternative embodiment, fig. 5 shows a flow diagram of a method for invalid culling processing, and as shown in fig. 5, the method at least includes the following steps: in step S510, the direct-connection routing information in the routing detail information is subjected to an invalid elimination process to obtain first routing information.

Because the direct connection routes of a single router are more and information which cannot be scheduled exists, the direct connection route information in the route detail information can be subjected to invalid elimination processing to obtain the first route information.

In step S520, the private network routing information in the first routing information is subjected to an invalid elimination process to obtain valid routing information.

In addition, there are a lot of private network routes beginning with 10 or 192, for example, and this part of private network route information is also invalid for service scheduling, so that the private network route information included in the first route information can be subjected to invalid elimination processing to obtain corresponding valid route information.

In the exemplary embodiment, by performing invalid elimination processing on the direct-connection routing information and the private network routing information in the routing detail information, a website which is invalid for service scheduling can be discarded, effective routing information is provided for node scheduling, and pressure brought by screening and converting routing information is relieved.

In step S420, the valid routing information is subjected to detail merging processing to obtain target routing information.

After the effective routing information is obtained, the effective routing information can be further subjected to detail combination processing.

In an alternative embodiment, fig. 6 shows a flowchart of a method of detail merge processing, and as shown in fig. 6, the method at least includes the following steps: in step S610, mask information in the valid routing information is acquired, and a mask condition corresponding to the mask information is acquired.

The mask information in the valid routing information may be a subnet mask.

The subnet mask, also called the netmask, address mask or subnet mask, may be used to indicate which bits of an IP address identify the subnet where the host is located and which bits identify the bitmask of the host.

The subnet mask cannot exist alone and it must be used in conjunction with the IP address. The subnet mask has only one role of dividing an IP address into two parts, a network address and a host address.

The subnet mask is a 32-bit address that masks a portion of the IP address to distinguish between the network identification and the host identification and indicates whether the IP address is on a local area network or a wide area network.

Further, the mask condition may be a condition that is determined to be set for mask information in valid route information to be merged. For example, the mask condition may be routing information below segment C.

Since the IP address is allocated by the nic (network information center) according to the network scale of the user unit and the recent development plan. Conceptually, each IP address can consist of two parts, namely a network identification and a netid and a host identification hostid. In fact, IP addresses can be classified into A, B, C categories.

The class C is distributed to a small-scale network, each network with a C address only has a small number of hosts, the first three 8 bits in a 32-bit address domain are specified as network identifiers, the first three bits are 110 and represent class C addresses, and the rest 8 bits are host identifiers and are distributed by a manager of the network.

In step S620, mask matching processing is performed on the mask information and the mask condition to obtain a matching processing result, and detail merging processing is performed according to the matching processing result to obtain target routing information.

After the mask information and the mask condition are obtained, mask matching processing may be performed on the mask information and the mask condition to obtain a corresponding matching processing result.

When the mask condition is under the C-segment condition, it may be determined by the mask matching process that the mask information is under/24 masks as a matching process result. Furthermore, the effective routing information ending in the effective routing information of/28,/30,/32 corresponding to the matching processing result is processed by detail combination to obtain the target routing information.

In the exemplary embodiment, the routing information which does not satisfy the mask condition is subjected to detail merging processing from the effective routing information, so that the problem of excessive entries can be solved, the effective routing information is provided for node scheduling, and the pressure brought by screening and converting the routing information is relieved.

In step S220, a server tag of the cache server corresponding to the user terminal is obtained, and policy generation processing is performed on the target routing information and the server tag to obtain a routing scheduling policy.

After the target routing information is determined, the server label of the cache corresponding to the user terminal can also be acquired. The server tag may be information that uniquely characterizes the corresponding cache server.

In the 5G era, the low-delay services such as live broadcast, AR, VR, etc. require that the network delay from the client to the server is controlled within 10ms, and the end-to-end delay is within 50ms, so that the smooth experience after the high-computing power service moves up to the edge node can be ensured. The traditional scheduling can only ensure low time delay and short path in local and quasi-real time after optimization, cannot ensure wide coverage and provide scheduling capability of low time delay in real time, and is represented by increased hop count and unstable time delay. The services such as live broadcast and the like carried by the edge computing node are often presented in the form of APP and have different access modes such as wifi, 4G/5G and the like. In order to ensure the consistency of service experience, the edge node is required to meet the requirement of low delay under various access modes.

It should be noted that the cache server may be divided into dimensions such as a district or a county at a fine granularity, so as to ensure that the time delay of the nearby call service is shortest and the access hop count is smallest.

Further, the server label is used for performing label printing processing on the target routing information to obtain a mapping relation between the user terminal and the cache service, so as to generate a corresponding routing scheduling strategy.

In the exemplary embodiment, the corresponding routing scheduling policy can be obtained through policy generation processing of the target routing information and the server tag, support can be provided for node scheduling, the real-time performance and accuracy of scheduling are improved, and the requirement of immediate matching of the local service of the corresponding cache server when a client requests is met.

After the routing scheduling policy is obtained, policy matching processing may be performed on the request address information and the routing scheduling policy.

Specifically, the request address information may be used as a primary key, and a routing scheduling policy matching the request address information may be queried as a policy matching result.

In step S130, a cache server corresponding to the user terminal is scheduled according to the policy matching result, so that the user terminal and the cache server perform data interaction.

In the exemplary embodiment of the present disclosure, since the route scheduling policy is obtained by performing policy generation processing on the target route information and the server tag, the server tag in the policy matching result may be obtained to determine the cache server according to the server tag, and schedule the cache server in the near future. Specifically, the user terminal may be linked with the cache server to implement data access.

In addition, when policy matching processing is performed on the request address information and the routing scheduling policy, a situation that the cache server in the routing scheduling policy cannot be matched occurs, so that a default policy can be set for scheduling.

In an optional embodiment, when the policy matching result is that the user terminal is not matched with the cache server, a default server corresponding to the user terminal is scheduled according to a default policy corresponding to the policy matching result, so that the user terminal and the default server perform data interaction.

The default strategy can be that a provincial-city-level server corresponding to the user terminal is determined nearby to serve as a default server, and a link is established between the user terminal and the default server to achieve data access.

In addition, the method and the system for scheduling the edge node can also provide an edge node scheduling system, wherein the edge node scheduling system comprises a client, an access network, a router, a scheduling server and a cache server, and the client can communicate with the scheduling server and the cache server through broadband, wifi and 4G/5G.

The scheduling server comprises a scheduling module, an address library module and a receiving module.

And the scheduling module judges when the client requests to be scheduled to the local city node side through domain name resolution, preferentially schedules the IP section information stored in the address base module to a corresponding cache server, and schedules the IP section information which is not in the address base module according to a default strategy.

And the address library module labels the route according to the dimension of the cache server and stores the label as a scheduling strategy.

Specifically, the target routing information and the server tag of the cache corresponding to the user terminal are obtained. The server tag may be information that uniquely characterizes the corresponding cache server.

The receiving module is responsible for receiving the routing information sent by the plurality of cache servers in real time.

And the cache server comprises a synchronization module, a screening and conversion module and a sending module.

And the synchronization module receives the IBGP route sent by the BRAS and other devices through the BGP route module of the server.

Specifically, route detail information corresponding to the user terminal is obtained through a link directly connected with the routing device, and information elimination processing is performed on the route detail information to obtain target route information.

The routing detail information comprises full routing information and modified routing information.

And, the full amount of routing information may be BGP (Border Gateway Protocol) routing information.

It should be noted that the full amount of routing information is only needed when the detailed routing information is first obtained.

The screening and translation module discards invalid address segments in the received BGP routes and merges routes below the/24 mask.

And carrying out invalid elimination processing on the routing detail information to obtain effective routing information.

And performing invalid elimination processing on the directly-connected routing information in the routing detail information to obtain first routing information.

Because the number of the direct routes of a single router is large, but the direct route pair cannot be scheduled, the direct route information in the route detail information can be subjected to invalid elimination processing to obtain the first route information.

And carrying out invalid elimination processing on the private network routing information in the first routing information to obtain valid routing information.

In addition, a large number of private network routes beginning with 10 or 192 exist, and the private network route information is also invalid for service scheduling, so that the private network route information included in the first route information can be subjected to invalid elimination processing to obtain corresponding valid route information.

After the effective routing information is obtained, the effective routing information can be further subjected to detail combination processing.

And acquiring mask information in the effective routing information and acquiring a mask condition corresponding to the mask information.

The mask information in the valid routing information may be a subnet mask.

Since the IP address is allocated by the NIC according to the network size of the user unit and the recent development plan. Conceptually, each IP address can consist of two parts, namely a network identification and a netid and a host identification hostid. In fact, IP addresses can be classified into A, B, C categories.

And performing mask matching processing on the mask information and the mask condition to obtain a matching processing result, and performing detail merging processing according to the matching processing result to obtain target routing information.

And the sending module sends the locally received and processed routing information to the scheduling server in real time.

The routing scheduling strategy acquired by the edge node scheduling system can be screened and updated in real time without sniffing by an IP library or peripheral equipment which is imported in advance, a data basis is provided for accurate scheduling, and the routing scheduling strategy is divided in a fine granularity mode, so that the scheduling process is not limited by frequency and a physical position, and data support is provided for optimal scheduling. Furthermore, the strategy matching result obtained by strategy matching processing is scheduled according to the request address information and the routing strategy scheduling, so that the effects of comprehensive coverage, real-time accuracy, optimality and shortest node scheduling are achieved.

The following describes the edge node scheduling method in the embodiment of the present disclosure in detail with reference to an application scenario.

Fig. 7 shows a system framework diagram of edge node scheduling in an application scenario, and as shown in fig. 7, in an edge computing scenario, various clients such as a mobile phone, a notebook, a tablet computer, and a desktop access nodes in different manners such as a base station, Wi-Fi, fixed broadband, and dedicated line.

The single node is generally formed in a form of "router or switch + cache server", and client detailed routes, i.e. route detailed information, of various access modes are stored in the router or switch. The cache server of the node can be a single server or a cluster formed by a plurality of servers. The client, the scheduling server and the cache server can communicate with each other in a 4G or 5G mode, a fixed broadband and other access modes.

In the edge computing scenario, the system architecture shown in fig. 7 is used for scheduling, so that the client requests can be accurately served at the nearest node, and the low-delay service requirements of the clients are ensured.

The edge node scheduling system can capture the change of the route at the first time, dynamically updates the scheduling strategy, and has the characteristics of full coverage, real-time accuracy and shortest path.

Fig. 8 is a flowchart illustrating a method for generating a routing scheduling policy in an application scenario, where as shown in fig. 8, in step S810, a router sends a locally stored client BGP route to a directly connected cache server.

The cache server installs BGP routing software and receives IBGP detail routing, namely routing detail information, according to links directly connected with routing equipment such as BRAS and the like. The routing detail information comprises full routing information and modified routing information.

And, the full amount of routing information may be BGP (Border Gateway Protocol) routing information.

It should be noted that the full amount of routing information is only needed when the detailed routing information is first obtained.

And acquiring modified routing information corresponding to the user terminal through a link directly connected with the routing equipment.

The method can acquire the full amount of routing information or the modified routing information corresponding to the user terminal through a link directly connected with the routing equipment, monitor the change of the client routing on the network in real time, and synchronize to a scheduling system without sniffing by means of an IP library or peripheral equipment which is imported in advance, thereby providing a data base and support for really realizing full coverage and real-time and accurate node scheduling.

In step S820, the cache server, after receiving the route, first rejects the invalid private network and duplicate routes.

After the route detail information is acquired, information elimination processing may be performed on the route detail information.

And carrying out invalid elimination processing on the routing detail information to obtain effective routing information.

And performing invalid elimination processing on the directly-connected routing information in the routing detail information to obtain first routing information.

And carrying out invalid elimination processing on the private network routing information in the first routing information to obtain valid routing information.

By performing invalid elimination processing on the direct-connection routing information and the private network routing information in the routing detail information, websites invalid for service scheduling can be discarded, effective routing information is provided for node scheduling, and pressure brought by screening and routing information conversion is relieved.

In step S830, the cache server then merges the detailed routes and sends them to the dispatch server.

After the effective routing information is obtained, the effective routing information can be further subjected to detail combination processing.

And acquiring mask information in the effective routing information and acquiring a mask condition corresponding to the mask information.

The mask information in the valid routing information may be a subnet mask.

The routing information which does not meet the mask condition is subjected to detail combination processing from the effective routing information, the problem of excessive entries can be solved, effective routing information is provided for node scheduling, and pressure brought by screening and routing information conversion is relieved.

And at this moment, the server synchronizes BGP detailed routes through links directly connected with routing equipment such as BRAS and the like, screens invalid routes and regular scattered routes, and finally sends the processed routes to the scheduling server.

In step S840, the dispatch server saves the route locally and tags the cache server.

In order to improve the scheduling accuracy and real-time performance, the scheduling server is required to tag the received client route of each cache server, so that the client can be matched with the corresponding cache server for nearby service immediately when requesting.

Specifically, after the target routing information is determined, the server tag of the cache corresponding to the user terminal may also be obtained. The server tag may be information that uniquely characterizes the corresponding cache server.

Based on this, the original BGP route has the problems of invalid entries, excessive number of entries, and the like, and the cache server needs to perform screening and conversion after receiving the BGP route, so that the BGP route can become information usable by the scheduling server.

In addition, the BGP route synchronization and conversion method does not need to use an IP library which is imported in advance or peripheral equipment for sniffing, and BGP routes are synchronized with routing equipment such as BRAS (broadband remote access server) in real time and are screened and converted to be used by a scheduling system. Besides, the method is not limited to BRAS equipment, and three-layer digital communication equipment such as UP of a novel metropolitan area network, CE of a 4G/5G core network, ER and the like for storing the BGP route of the user are suitable for the method.

The corresponding routing scheduling strategy can be obtained through strategy generation processing of the target routing information and the server label, support can be provided for node scheduling, scheduling instantaneity and accuracy are improved, and the requirement of immediate matching of the local service of the corresponding cache server when a client requests is met.

Fig. 9 is a flowchart illustrating a method for implementing data access in an application scenario, as shown in fig. 9, in step S910, a client initiates a normal domain name request.

The user terminal may initiate a domain name request through the client. In addition, the client does not need to be modified, as long as the client can normally initiate the domain name request.

After receiving the domain name request, the domain name resolution process may be performed on the domain name request.

Domain name resolution is also called domain name pointing, server setting, domain name configuration, reverse IP registration, etc. The simple point is to resolve a well-documented domain name into an IP, and the service is done by a DNS server, which resolves the domain name into an IP address and then binds a subdirectory to the domain name on the host of this IP address.

The addresses in the internet are digital IP addresses, and the function of domain name resolution is mainly to facilitate memory.

Therefore, the request address information corresponding to the domain name request, i.e. the client IP address, can be obtained through the domain name resolution process.

In step S920, after receiving the request, the scheduling server matches a scheduling policy according to the client IP.

After the routing scheduling policy is obtained, policy matching processing may be performed on the request address information and the routing scheduling policy.

Specifically, the request address information may be used as a primary key, and a routing scheduling policy matching the request address information may be queried as a policy matching result.

In step S930, if the client IP is transmitted for a certain cache server, the scheduling server schedules the client request to the cache server nearby.

Since the routing scheduling policy is obtained by performing policy generation processing on the target routing information and the server tag, the server tag in the policy matching result may be obtained, so as to determine the cache server according to the server tag.

In step S940, the client establishes a link with the nearby cache server to implement data access according to the result returned by the scheduling server request.

In the 5G era, the low-delay services such as live broadcast, AR, VR, etc. require that the network delay from the client to the server is controlled within 10ms, and the end-to-end delay is within 50ms, so that the smooth experience after the high-computing power service moves up to the edge node can be ensured. The traditional scheduling can only ensure low time delay and short path in local and quasi-real time after optimization, cannot ensure wide coverage and provide scheduling capability of low time delay in real time, and is represented by increased hop count and unstable time delay. The services such as live broadcast and the like carried by the edge computing node are often presented in the form of APP and have different access modes such as wifi, 4G/5G and the like. In order to ensure the consistency of service experience, the edge node is required to meet the requirement of low delay under various access modes.

When the user terminal is dispatched to the cache server nearby, the user terminal can be linked with the cache server to realize data access. It is worth noting that the cache server is fine grained in a district or county.

And when the cache server is not scheduled nearby, the default strategy can determine the provincial-city-level server corresponding to the user terminal nearby as the default server, and establish a link between the user terminal and the default server to realize data access.

Based on the method, the scheduling server is matched with the scheduling strategy according to the client IP and schedules the client IP to the cache server service receiving the client IP address route. And the requests of the client are scheduled according to the real-time optimal path, so that the shortest access hop count and shortest time delay of the client can be ensured. Moreover, the flow of the edge node accurate scheduling does not need the client to prestore or support extra capacity, and the method is suitable for rapid replication and popularization.

Fig. 10 shows a system structure diagram of edge node scheduling in an application scenario, as shown in fig. 10, the scheduling system includes a client 1010, an access network 1020, a router 1030, a scheduling server 1040 and a cache server 1050, and the client 1010 may communicate with the scheduling server 1040 and the cache server 1050 through broadband, wifi, 4G/5G.

The scheduling server 1040 includes a scheduling module 1041, an address library module 1042 and a receiving module 1043.

And, the scheduling module 1041 performs a judgment when the client requests to be scheduled to the local city node side through domain name resolution, preferentially schedules the IP segment information stored in the address base module to the corresponding cache server, and schedules the IP segment not in the address base module according to a default policy.

The address library module 1042 tags the route according to cache server dimensions and saves it as a scheduling policy.

The receiving module 1043 is responsible for receiving the routing information sent by the plurality of cache servers in real time.

Also, the cache server 1050 includes a synchronization module 1051, a screening and translation module 1052, and a sending module 1053.

And, the synchronization module 1051 receives the IBGP route sent by the device such as the BRAS through the server BGP route module.

The routing detail information comprises full routing information and modified routing information.

And, the full amount of routing information may be BGP (Border Gateway Protocol) routing information.

It should be noted that the full amount of routing information is only needed when the detailed routing information is first obtained.

The screening and translation module 1052 discards invalid address segments in the received BGP routes and merges routes below the/24 mask.

And carrying out invalid elimination processing on the routing detail information to obtain effective routing information.

And performing invalid elimination processing on the directly-connected routing information in the routing detail information to obtain first routing information.

And carrying out invalid elimination processing on the private network routing information in the first routing information to obtain valid routing information.

After the effective routing information is obtained, the effective routing information can be further subjected to detail combination processing.

And acquiring mask information in the effective routing information and acquiring a mask condition corresponding to the mask information.

The mask information in the valid routing information may be a subnet mask.

The sending module 1053 sends the locally received and processed routing information to the scheduling server in real time.

In the edge node scheduling method in the application scene, the obtained routing scheduling strategy can be screened and updated in real time without sniffing by an IP library or peripheral equipment which is imported in advance, a data basis is provided for accurate scheduling, and the routing scheduling strategy is divided in a fine granularity mode, so that the scheduling process is not limited by frequency and physical positions, and data support is provided for optimal scheduling. Furthermore, the strategy matching result obtained by strategy matching processing is scheduled according to the request address information and the routing strategy scheduling, so that the effects of comprehensive coverage, real-time accuracy, optimality and shortest node scheduling are achieved.

In addition, in the exemplary embodiment of the present disclosure, an edge node scheduling apparatus is also provided. Fig. 11 shows a schematic structural diagram of an edge node scheduling apparatus, and as shown in fig. 11, the edge node scheduling apparatus 1100 may include: domain name resolution module 1110, policy matching module 1120, and node scheduling module 1130. Wherein:

a domain name resolution module 1110, configured to obtain a domain name request sent by a user terminal, and perform domain name resolution processing on the domain name request to obtain request address information;

the policy matching module 1120 is configured to obtain a routing scheduling policy, and perform policy matching processing on the request address information and the routing scheduling policy to obtain a policy matching result;

and the node scheduling module 1130 is configured to schedule the cache server corresponding to the user terminal according to the policy matching result, so that the user terminal and the cache server perform data interaction.