CN110417676B

CN110417676B - Scheduling method, device and scheduling network

Info

Publication number: CN110417676B
Application number: CN201910722119.2A
Authority: CN
Inventors: 吴刚; 查毅勇; 韩云博; 黄巍
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd; Tencent Cloud Computing Beijing Co Ltd
Priority date: 2019-08-06
Filing date: 2019-08-06
Publication date: 2022-10-04
Anticipated expiration: 2039-08-06
Also published as: CN110417676A

Abstract

The invention discloses a scheduling method, a device and a scheduling network, wherein the method is applied to an MEC scheduling server in the scheduling network, and comprises the following steps: receiving a first DNS request sent by a user, analyzing domain name information from the first DNS request, under the condition that the MEC dispatching server does not store the domain name information, regenerating a second DNS request containing an IP address of the MEC dispatching server, and sending the second DNS request to the GSLB dispatching server through a core network so that the GSLB dispatching server distributes CDN nodes for the user according to the IP address of the MEC dispatching server. Because the MEC has the characteristic of being closer to the user, the IP address of the MEC scheduling server is closer to the user, and the CDN node distributed by the MEC scheduling server is closer to the user, so that the network delay is reduced, and the user experience is improved.

Description

Scheduling method, device and scheduling network

Technical Field

The present invention relates to the field of communications, and in particular, to a scheduling method, apparatus, and scheduling network.

Background

With the development of communication technology, the progress of industry and the improvement of people's life and entertainment demands, more and more internet services, such as high-definition video, live broadcast, AR/VR, car networking services and the like, emerge. Typically, the resources of the internet service are stored in a server, and the user uses the resources of the internet service by accessing the server.

However, the proliferation of internet services also brings huge user access amount, and further generates huge network traffic, which easily causes network congestion, and the network congestion may cause a situation that a user is stuck when accessing services, for example, when the user watches high definition video, if the access amount is too large, the user cannot watch the video smoothly, and a situation that the video is played and stuck occurs, which seriously affects the user experience.

Therefore, in the prior art, in order to improve the user experience, network congestion caused by huge network traffic becomes an urgent problem to be solved.

Disclosure of Invention

In view of this, the embodiment of the present invention discloses a scheduling method, an apparatus and a scheduling network, where an MEC scheduling server is deployed in the scheduling network, and a CDN node is allocated to a user according to an IP address of the MEC scheduling server by using a characteristic that the MEC is closer to the user, so that the CDN node allocated to the user is closer to the user, a distance between the CDN node allocated to the user and the user is greatly shortened, network delay is shortened, and user experience is further improved.

The embodiment of the invention discloses a scheduling method, which is applied to an MEC scheduling server in a scheduling network, wherein the scheduling network comprises the following steps:

the system comprises a base station, an MEC scheduling server, a core network and a GSLB scheduling server;

the method comprises the following steps:

receiving a first DNS request of a user sent by a base station, and analyzing domain name information from the first DNS request;

if the domain name information is not stored in the MEC dispatching server, generating a second DNS request containing the IP address of the MEC dispatching server;

sending the second DNS request to the GSLB dispatching server through the core network; receiving an IP address of a CDN node distributed for a user by a GSLB scheduling server sent by a core network; the CDN node is distributed for the user by the GSLB dispatching server according to the IP address of the MEC dispatching server;

and sending the IP address of the CDN node allocated for the user to the user through the base station.

Optionally, the method further includes:

if the MEC scheduling server stores the domain name information, acquiring at least one MEC content server which has a mapping relation with the domain name information; business resources are cached in the MEC content server;

selecting an MEC content server which accords with a preset rule from at least one MEC content server which has a mapping relation with the domain name information and distributing the MEC content server to a user;

and sending the IP address of the MEC content server which accords with the preset rule to the user through the base station.

Optionally, selecting an MEC content server meeting a preset rule from at least one MEC content server having a mapping relationship with the domain name information includes:

if only one MEC content server which has a mapping relation with the domain name information is contained, the only MEC content server which has the mapping relation with the domain name information is distributed to the user;

and if the MEC content server comprises a plurality of MEC content servers which have mapping relations with the domain name information, selecting an MEC content server with the optimal performance from the MEC content servers which have the mapping relations with the domain name information according to the load information and the time delay information of the MEC content servers, and distributing the MEC content server with the optimal performance to the user.

Optionally, the selecting, according to the load information and the delay information of the MEC content server, an MEC content server with an optimal performance from a plurality of MEC content servers having a mapping relationship with the domain name information includes:

judging whether the load service condition of each MEC content server meets a preset load condition or not according to the load information of each MEC content server;

if a plurality of MEC content servers meet the preset load condition, selecting an MEC content server with the lowest time delay from the MEC content servers meeting the preset load condition as an MEC content server with the optimal performance;

if only one MEC content server meets the preset load condition, taking the MEC content server which only meets the preset load condition as the MEC content server with the optimal performance;

and if the MEC content server meeting the preset load condition does not exist, selecting the MEC content server with the lowest time delay as the MEC content server with the optimal performance.

Optionally, the method further includes:

and acquiring the time delay information and the load information of each MEC content server connected with the MEC scheduling server, and updating the time delay information and the load information of the MEC content server stored in the MEC content server.

Optionally, the obtaining of the delay information and the load information of each MEC content server connected to the MEC scheduling server includes:

sending a message to the MEC content server by using a ping command; receiving time delay information fed back by the MEC content server based on the message;

receiving load information sent by the MEC content server; the load information of the MEC content server is collected by the MEC content server according to a preset instruction.

Optionally, the load information includes:

any one or more of CPU utilization rate, memory utilization rate, disk read-write rate and network card rate.

The embodiment of the invention also discloses a scheduling device, which comprises:

a first receiving unit, configured to receive a first DNS request of a user sent by a base station, and parse domain name information from the first DNS request;

a second DNS request generating unit, configured to generate a second DNS request including an IP address of the MEC scheduling server if the domain name information is not stored in the MEC scheduling server;

a first sending unit, configured to send the second DNS request to the GSLB scheduling server through the core network;

the second receiving unit is used for receiving the IP address of the CDN node distributed for the user by the GSLB scheduling server sent by the core network; the CDN node is distributed for the user by the GSLB dispatching server according to the IP address of the MEC dispatching server;

and the second sending unit is used for sending the IP address of the CDN node allocated to the user through the base station.

Optionally, the method further includes:

an obtaining unit, configured to obtain at least one MEC content server having a mapping relationship with the domain name information if the domain name information is stored in the MEC scheduling server; business resources are cached in the MEC content server;

an MEC content server selecting unit, configured to select an MEC content server meeting a preset rule from at least one MEC content server having a mapping relationship with the domain name information and allocate the server to a user;

and the third sending unit is used for sending the IP address of the MEC content server which accords with the preset rule to the user through the base station.

The embodiment of the invention also discloses a scheduling method, which is applied to the GSLB scheduling server in the scheduling network, and the scheduling network comprises the following steps:

the method comprises the following steps:

receiving a second DNS request sent by the core network, and resolving an IP address of the MEC scheduling server contained in the second DNS request; the second DNS request is generated by an MEC dispatch server;

distributing CDN nodes for users according to the IP address of the MEC scheduling server;

and sending the IP address of the CDN node distributed for the user to the MEC scheduling server through a core network.

Optionally, the allocating CDN nodes to the user according to the IP address of the MEC scheduling server includes:

determining a first CDN node closest to the IP address of the MEC scheduling server according to the IP address of the MEC scheduling server;

judging whether the access amount of the first CDN node exceeds a preset threshold value or not;

if the access amount of the first CDN node exceeds a preset threshold value, acquiring at least one second CDN node of which the distance to the IP address of the MEC scheduling server is within a preset distance range;

and selecting the CDN node with the lowest access amount from the at least one second CDN node, and distributing the CDN node with the lowest access amount to the user.

a third receiving unit, configured to receive a second DNS request sent by a core network, and resolve an IP address of the MEC scheduling server in the second DNS request; the second DNS request is generated by an MEC dispatch server;

the allocation unit is used for allocating CDN nodes for the users according to the IP address of the MEC scheduling server;

a fourth sending unit, configured to send the IP address of the CDN node allocated to the user to the MEC scheduling server through the core network.

The embodiment of the invention also discloses a scheduling network, which comprises:

the base station is used for sending a first DNS request of a user to the MEC scheduling server;

the MEC scheduling server is used for analyzing domain name information contained in the first DNS request, regenerating a second DNS request containing an IP address of the MEC scheduling server under the condition that the MEC scheduling server does not store the domain name information, sending the second DNS request to the GSLB scheduling server through a core network, and sending the IP address of the CDN node to a user through a base station after receiving the IP address of the CDN node distributed for the user;

the core network is used for sending the second DNS request generated by the MEC scheduling server to the GSLB scheduling server and sending the IP address of the CDN node distributed by the GSLB scheduling server to the MEC scheduling server;

and the GSLB scheduling server is used for allocating CDN nodes according to the IP addresses of the MEC scheduling servers and sending the allocated IP addresses of the CDN nodes to the MEC scheduling servers through a core network.

Optionally, the MEC-based scheduling network further includes:

the MEC content servers are connected with the MEC scheduling server and store business resources;

the MEC scheduling server comprises: domain name information and a mapping relation between each domain name and an MEC content server; the one domain name may be mapped to at least one MEC content server.

Optionally, the MEC scheduling server is further configured to:

and sending the IP address of the MEC content server which meets the preset rule to a user through the base station.

The invention discloses a scheduling method, which is applied to a MEC scheduling server in a scheduling network, wherein the network comprises: the system comprises a base station, an MEC scheduling server, a core network and a GSLB scheduling server. The method comprises the following steps: receiving a first DNS request sent by a user, analyzing domain name information from the first DNS request, regenerating the DNS request under the condition that the MEC dispatching server does not store the domain name information to obtain a second DNS request containing the IP address of the MEC dispatching server, and sending the second DNS request to the GSLB dispatching server through a core network so that the GSLB dispatching server distributes CDN nodes for the user according to the IP address of the MEC dispatching server.

Therefore, when network scheduling is performed by the method, the MEC scheduling server is deployed in the scheduling network, and the MEC scheduling server regenerates the DNS request containing the IP address of the MEC scheduling server under the condition that the MEC scheduling server does not contain domain name information, so that the GSLB scheduling server can allocate CDN nodes to users according to the IP address of the MEC scheduling server. In this way, since the MEC has a characteristic of being closer to the user, the IP address of the MEC scheduling server is also closer to the user, and then the CDN node allocated by the MEC scheduling server is also closer to the user. Therefore, the problem that in the prior art, a CDN node which is allocated to a user through an IP address of a Local DNS server is far away from the user, so that network delay is caused is solved.

And when the MEC scheduling server contains domain name information, an MEC content server with the optimal performance can be selected from a plurality of MEC content servers connected with the MEC scheduling server according to the mapping relation between the domain name and the MEC content server and the performance of the MEC content server, so that the MEC content server with the excellent performance is provided for a user, and the experience degree of the user for accessing the internet service is further improved.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the prior art descriptions will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.

Fig. 1 is a schematic structural diagram illustrating a scheduling network according to an embodiment of the present invention;

fig. 2 is a schematic structural diagram of another scheduling network provided in an embodiment of the present invention;

fig. 3 is a schematic flowchart illustrating a scheduling method according to an embodiment of the present invention;

FIG. 4 is a schematic flowchart illustrating a scheduling method according to an embodiment of the present invention;

fig. 5 is a schematic structural diagram illustrating a scheduling apparatus according to an embodiment of the present invention;

fig. 6 is another schematic flow chart of a scheduling method according to an embodiment of the present invention;

fig. 7 is a schematic structural diagram illustrating a scheduling apparatus according to an embodiment of the present invention;

fig. 8 is an interaction diagram illustrating a scheduling method according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

In order to alleviate network congestion caused by huge network traffic, a CDN (full name in english: content delivery network, full name in chinese: content delivery network) should be created.

The basic principle of the CDN is to widely adopt various cache servers in a network, distribute the cache servers to a region where user access is relatively concentrated, and when a user accesses a website, point the user access to a cache server that works normally and that is closest to the user by using a global load technology, so as to achieve offloading of user access volumes and solve network congestion.

However, the inventors have found that even though the CDN can solve the network congestion problem, there is still a problem of latency when the user accesses the internet traffic.

The inventor finds that, starting from a CDN scheduling method, in a conventional CDN scheduling method, when allocating a CDN node to a user, an IP address of a Local DNS server is used as a basis, but because Local DNS has wide coverage, for example, a coverage area of the Local DNS may be one province, the Local DNS may not be in a region with the user or may be far away from the user, a CDN node allocated to the user according to an exit IP address of the Local DNS is also far away from the user, and a physical distance is a factor causing delay, and in a case where the physical distance is far, a high network delay may be generated.

Therefore, based on the above research, it is found that the physical distance is a key factor that causes delay, and thus the key to solve the delay problem of accessing internet services is to shorten the physical distance between the CDN node allocated to the user and the user.

In order to solve the above problems, the inventors have found through research that, the MEC (Mobile Edge Computing) defined by ETSI (European Telecommunications Standards Institute) has a maximum feature point close to the customer, and it can also be understood that the coverage of the MEC is small, for example, a plurality of MECs may be deployed in a city, and the coverage of each MEC is a plurality of cells.

Therefore, the discoverer can solve the network delay caused by the distance problem by deploying the MEC by utilizing the characteristic that the MEC is close to the user.

In view of this, the present invention discloses a scheduling method, which is applied to a MEC scheduling server in a scheduling network, wherein the network includes: the system comprises a base station, an MEC scheduling server, a core network and a GSLB scheduling server. The method comprises the following steps: receiving a first DNS request sent by a user, resolving domain name information from the first DNS request, under the condition that the MEC dispatching server does not store the domain name information, regenerating the DNS request to obtain a second DNS request containing an IP address of the MEC dispatching server, and sending the second DNS request to the GSLB dispatching server through a core network so that the GSLB dispatching server allocates CDN nodes for the user according to the IP address of the MEC dispatching server.

Therefore, when network scheduling is performed by the method, the MEC scheduling server is deployed in the scheduling network, and the MEC scheduling server regenerates the DNS request including the IP address of the MEC scheduling server under the condition that the MEC scheduling server does not include domain name information, so that the GSLB scheduling server can allocate CDN nodes to users according to the IP address of the MEC scheduling server. In this way, since the MEC has a characteristic of being closer to the user, the IP address of the MEC scheduling server is also closer to the user, and then the CDN node allocated by the MEC scheduling server is also closer to the user. Therefore, the problem that in the prior art, a CDN node which is allocated to a user through an IP address of a Local DNS server is far away from the user, so that network delay is caused is solved.

Furthermore, the MEC content servers are deployed in the network, when the MEC scheduling server contains domain name information, an MEC content server with the optimal performance can be selected from the MEC content servers connected with the MEC scheduling server according to the mapping relation between the domain name and the MEC content server and the performance of the MEC content server, the MEC content server with the excellent performance can be provided for users, and the experience degree of the users for accessing internet services is further improved.

The scheduling method and scheduling network disclosed in the present application are described in detail as follows: referring to fig. 1, a schematic structural diagram of a scheduling network according to an embodiment of the present invention is shown, where in this embodiment, the network includes:

a base station 100, an MEC scheduling server 200, a core network 300, a GSLB scheduling server 400;

the base station 100 is configured to send a first DNS request of a user to an MEC scheduling server;

the MEC scheduling server 200 is configured to parse domain name information included in the first DNS request, regenerate a second DNS request including an IP address of the MEC scheduling server when the MEC scheduling server does not store the domain name information, send the second DNS request to the GSLB scheduling server through the core network, and send the IP address of the CDN node to the user through the base station after receiving the IP address of the CDN node allocated to the user;

the core network 300 is configured to send the second DNS request generated by the MEC scheduling server to the GSLB scheduling server, and send the IP address of the CDN node allocated by the GSLB scheduling server to the MEC scheduling server;

the GSLB scheduling server 400 is configured to allocate CDN nodes according to IP addresses of the MEC scheduling servers, and send the allocated IP addresses of the CDN nodes to the MEC scheduling server through a core network;

as can be seen from the above description, the MEC scheduling server is deployed in the scheduling network, and the coverage area of the MEC scheduling server is small, and is closer to the user, for example, one MEC scheduling service may cover the range of several cells. When the MEC scheduling server schedules the network resources of the user, a DNS request containing the IP address of the MEC scheduling server is generated in the MEC scheduling server, so that the GSLB scheduling server distributes CDN nodes for the user according to the IP address of the MEC scheduling server. In this way, the user may be assigned to a CDN node closer to the user.

The GSLB (Global Server Load Balance, chinese full name: load Balance) has the functions of realizing flow allocation among servers in different regions on a wide area network (including the Internet), ensuring that the client nearest to the Server is served by the best Server, and further ensuring the access quality.

When distributing CDN nodes for a user through the GSLB scheduling service, in principle, a CDN node closest to the user is distributed to the user, and in this embodiment, a CDN node closest to an IP address of the MEC scheduling server is distributed to the user.

However, in some special cases, for example, the CDN node closest to the IP address of the MEC scheduling server has a large user access amount, and even if the CDN node closest to the IP address of the MEC scheduling server is allocated to a user, a stuck situation may occur.

In order to solve the above problem, the GSLB scheduling server may also allocate CDN nodes to users according to a performance optimization principle, and ensure that CDN nodes with better performance are allocated to users. A specific method for allocating CDN nodes with better performance to a user will be described in the following embodiments (S601-S603), which are not described again in this embodiment.

In the scheduling network disclosed in this embodiment, an MEC scheduling server is deployed, and when receiving a first DNS request sent by a base station, if domain name information in the first DNS request is not stored in the MEC scheduling server, the MEC scheduling server generates a second DNS request including an IP address of the MEC scheduling server, and sends the second DNS request to a GSLB scheduling server; and the GSLB dispatching server distributes CDN nodes for the users according to the IP addresses of the MEC dispatching servers. In this case, since the MEC is characterized by being closer to the user, the IP address of the MEC scheduling server is closer to the user, and the GSLB scheduling server allocates CDN nodes to the user according to the IP address of the MEC scheduling server, and is also closer to the user. Therefore, the CDN nodes distributed in the mode shorten the distance between the user and the CDN nodes, reduce the network delay and further improve the experience of the user on Internet services.

In this embodiment, in order to provide a smoother network service for the user, a server may be locally set, and a service resource may be stored in the local server, so that the purpose of providing an internet service for the user may be achieved without calling a CDN node through an external network. Specifically, referring to fig. 2, a schematic structural diagram of another scheduling network provided in an embodiment of the present invention is shown, where in this embodiment, the network includes:

a base station 100, an MEC scheduling server 200, a plurality of MEC content servers 500 connected to the MEC scheduling server, a core network 300, a GSLB scheduling server 400;

the MEC content server 500 stores therein service resources;

the MEC scheduling server 200 stores therein: domain name information and a mapping relation between each domain name and an MEC content server; wherein a domain name may be mapped to at least one MEC content server.

In this embodiment, under the network architecture, another scheduling method includes:

a base station sends a first DNS request of a user to an MEC scheduling server;

the MEC scheduling server analyzes domain name information contained in the first DNS request, and if the MEC scheduling server stores the domain name information, at least one MEC content server which has a mapping relation with the domain name information is obtained; selecting an MEC content server which accords with a preset rule from at least one MEC content server which has a mapping relation with the domain name information; and sending the IP address of the MEC content server which meets the preset rule to a user through the base station.

The method for selecting an MEC content server meeting a preset rule from at least one MEC content server having a mapping relationship with domain name information may include multiple selection manners, which is not limited in this embodiment and may include:

the method comprises the steps that in the first mode, one MEC content server is randomly selected from at least one MEC content server which has a mapping relation with domain name information;

in a second manner, under the network architecture of this embodiment, in order to enable a user to experience smoother internet services, the MEC scheduling server may select an MEC content server with the best performance from the related MEC content servers according to the principle of the best performance when selecting the MEC content server through the domain name information.

The process of the second mode may include:

and if the MEC content servers with the mapping relation with the domain name information are included, selecting an MEC content server with the optimal performance from the MEC content servers with the mapping relation with the domain name information according to the load information and the time delay information of the MEC content servers, and distributing the MEC content server with the optimal performance to the user.

In this embodiment, the load and the time delay of the server may reflect the operation condition of the server, and the lower the load occupancy rate and the lower the time delay of the server are, the better the operation condition of the server is, that is, the better the performance of the server is; conversely, the higher the load rate and the higher the delay, the worse the performance of the server is.

In this embodiment, a specific implementation manner of selecting an optimal MEC content server from a plurality of MEC content servers having a mapping relationship with domain name information according to the load information and the time delay information is not limited in this embodiment, and may include:

if a plurality of MEC content servers meet a preset load condition, selecting an MEC content server with the lowest time delay from the MEC content servers meeting the preset load condition as an MEC content server with the optimal performance;

if only one MEC content server meets the preset load condition, taking the only MEC content server meeting the preset load condition as the MEC content server with the optimal performance;

In this embodiment, the load information of the MEC content server includes at least one load index, where the load index may include: CPU utilization rate, memory utilization rate, disk read-write rate, network card rate and the like. In addition, other indexes capable of representing the load occupation condition of the MEC content server may be included, which is not limited in this embodiment.

In this embodiment, the manner for the MEC scheduling server to obtain the delay information and the load information may include multiple manners, which are not limited in this embodiment, and may include:

sending a message to the MEC content server by adopting a ping command;

receiving delay information fed back by the MEC content server based on the message;

receiving load information sent by an MEC content server; the load information of the MEC content server is collected by the MEC content server according to a preset instruction.

The MEC content server collects various load indexes of the MEC content server regularly or irregularly and sends the collected load indexes to the MEC scheduling server; for the acquisition of the delay information, the MEC scheduling server may send a message to the MEC content server by using a ping command periodically or aperiodically, and the MEC content server may feed back the time length of the delay to the MEC scheduling server after receiving the message.

In this embodiment, a plurality of MEC content servers are deployed in the scheduling network, and the MEC content servers include service resources. When scheduling is carried out through the MEC scheduling server, if domain name information is stored in the MEC scheduling server, an MEC content server which has a mapping relation with the domain name information is obtained, and an MEC content server with the optimal performance is selected and distributed to users. Therefore, when the MEC content server is deployed, the CDN node does not need to be distributed to the user through the external network, and the access to the Internet service can be realized. In addition, according to the introduction, the MEC scheduling server is closer to the user, so that the MEC content servers connected with the MEC scheduling server are closer to the user, and performance factors are considered when the MEC content servers are selected, so that the network delay is reduced, and the smoothness degree of the user in accessing the Internet is improved.

As can be known from the above description of the scheduling network, in the scheduling network, the base station is used to forward information between the user and the MEC scheduling server, and the core network is used to forward information between the MEC scheduling server and the GSLB scheduling server. Therefore, the MEC scheduling server and the GSLB scheduling server play a core scheduling role in the scheduling network.

The following describes the role of MEC scheduling server and GSLB scheduling server in the network scheduling process in detail:

for MEC scheduling server, refer to fig. 3 and 4: referring to fig. 3, a schematic flowchart of a scheduling method provided by an embodiment of the present invention is shown, where the method is applied to an MEC scheduling server in the scheduling network, where the scheduling network, referring to fig. 1, includes: the system comprises a base station, an MEC scheduling server, a core network and a GSLB scheduling server;

the method comprises the following steps:

s301: receiving a first DNS request of a user sent by a base station, and resolving domain name information from the first DNS request;

the DNS Domain Name System (abbreviated DNS) is a service of the Internet. It acts as a distributed database that maps domain names and IP addresses to each other, enabling people to more conveniently access the internet.

In this embodiment, the IP address where the internet service resource corresponding to the DNS request is located can be found for the user through the DNS request.

S302: if the domain name information is not stored in the MEC dispatching server, generating a second DNS request containing the IP address of the MEC dispatching server;

in this embodiment, for some popular internet services, the service resources may be stored locally, where the IP addresses and domain name information stored locally have a mapping relationship, and a domain name may be mapped to one or more IP addresses. The MEC scheduling server stores domain name information and the mapping relation between each domain name information and the IP address. Specifically, the description will be given in the following fourth embodiment (S401 to S405), and will not be described herein again.

However, for some cold services, resources that do not store the service in the MEC scheduling server need to call CDN nodes through an external network to access some cold services. In this embodiment, when a CDN node is called by an external network, the CDN node needs to be located according to an IP address, and an IP address of the MEC scheduling server is closer to a user, and the CDN node may be allocated to the user by the IP address of the MEC scheduling server. However, since the first CDN node does not include an IP address, it is necessary to regenerate a second DNS request including an IP address of the MEC scheduling server.

S304: sending the second DNS request to a GSLB scheduling server through a core network;

in this embodiment, the MEC scheduling server sends the second DNS request to the core network, and sends the second DNS request to the GSLB scheduling server through the core network.

S305: receiving an IP address of a CDN node distributed for a user by a GSLB dispatching server sent by a core network;

and the CDN node is distributed for the user by the GSLB scheduling server according to the IP address of the MEC scheduling server.

In this embodiment, the CDN node allocated by the GSLB scheduling server for the user may be a CDN node closest to the IP address of the MEC scheduling server, or may also be a CDN node with the optimal performance selected according to a performance optimization principle, where the CDN node with the optimal performance may be the CDN node closest to the IP address of the MEC scheduling server, or may not be the CDN node closest to the MEC scheduling server.

The process of the GSLB scheduling server allocating CDN nodes for the user is explained in the following embodiments (S501 to S503).

S306: and sending the IP address of the CDN node allocated for the user to the user through the base station.

In this embodiment, the IP address of the CDN node is sent to the user, and the user can access the CDN node through the IP address, thereby accessing the service resource in the CDN node.

In this embodiment, when the MEC scheduling server performs scheduling, if the MEC scheduling server does not store domain name information of the first DNS request, a second DINS request including an IP address of the MEC scheduling server is generated again, and the second DNS request is sent to the GSLB scheduling server through the core network, so that the GSLB scheduling server allocates CDN nodes to users according to the IP address of the MEC scheduling server. Therefore, as the MEC has the characteristic of being closer to the user, the IP address of the MEC scheduling server is closer to the user, and therefore the GSLB scheduling server allocates CDN nodes to the user according to the IP address of the MEC scheduling server, and is also closer to the user. Therefore, the CDN nodes distributed for the users in the mode greatly shorten the distance between the users and the CDN nodes, reduce the network delay and further improve the experience of the users on Internet services.

In addition, when the MEC scheduling server stores the first DNS domain name information, the local server may be further allocated to the user by locally querying the domain name information, specifically, refer to S401 to S405 as follows:

referring to fig. 4, a schematic flowchart of a scheduling method disclosed in an embodiment of the present invention is shown, where the method includes:

s401: receiving a first DNS request of a user sent by a base station, and analyzing domain name information from the first DIS request;

s401 is the same as S301 described above, and is not limited in this embodiment.

S402: if the MEC scheduling server stores the domain name information, acquiring at least one MEC content server which has a mapping relation with the domain name information;

in this embodiment, as can be seen from the above description, a plurality of MEC content servers are also deployed in the scheduling network, and the plurality of MEC content servers are connected to the MEC scheduling server. And, the MEC content server stores therein the service resources. The MEC scheduling server stores: domain name information and a mapping relation between each domain name and an MEC content server; the one domain name may be mapped to at least one MEC content server.

S403: selecting an MEC content server which accords with a preset rule from at least one content server which has a mapping relation with the domain name information and distributing the MEC content server to the user;

in this embodiment, as to a method for selecting an MEC content server meeting a preset rule from at least one MEC content server having a mapping relationship with domain name information, the selection manner may include multiple manners, and the method in this embodiment is not limited, and may include:

secondly, distributing the MEC content server for the user according to the performance optimization principle;

in order to enable a user to experience smoother internet services, when the MEC scheduling server selects the MEC content server through the domain name information, the MEC content server with the optimal performance can be selected from the related MEC content servers according to the optimal performance principle.

For the second mode, S403 may include:

if only one MEC content server with the mapping relation with the domain name information is contained, the only MEC content server with the mapping relation with the domain name information is distributed to the user;

In this embodiment, the load and the time delay of the server may represent the operation condition of the server, where the lower the load occupancy rate and the lower the time delay of the server are, the better the operation condition of the server is, that is, the better the performance of the server is; conversely, the higher the load rate and the higher the delay, the worse the performance of the server.

In this embodiment, a specific implementation manner of selecting an optimal MEC content server from multiple MEC content servers having a mapping relationship with domain name information according to the load information and the time delay information is not limited in this embodiment, and may include, for example:

In this embodiment, the load information of the MEC content server includes at least one load index, where the load index may include: any one or more of CPU utilization rate, memory utilization rate, disk read-write rate and network card rate. In addition, other indexes capable of representing the load occupation condition of the MEC content server may be included, which is not limited in this embodiment.

The load condition may be represented as whether the load index is smaller than a preset threshold, for example, the threshold may be 80%.

For example, the following steps are carried out: if the load indexes of a plurality of MEC content servers are smaller than 80%, selecting the MEC content server with the smallest time delay from the MEC content servers with the load indexes smaller than 80%; if the load index of one MEC content server is smaller than 80%, distributing the only MEC content server with the load index smaller than 80% to the user; and if no MEC content server with the load index smaller than 80% exists, selecting the MEC content server with the minimum time delay.

The method includes the steps that the load indexes of the MEC content server are possibly multiple, and for the load condition that whether the load indexes are smaller than a preset threshold value or not, the MEC content server can be considered to meet the preset load condition when at least N load indexes are smaller than the preset threshold value, and the MEC content server cannot meet the preset load condition when the load indexes are smaller than the N load indexes and the preset threshold value; if the MEC content server comprises N load indexes, N is more than or equal to 1 and less than or equal to N.

For example, the following steps are carried out: the load indexes include: the method comprises the following steps that a CPU utilization rate, a memory utilization rate, a disk read-write rate and a network card rate are determined, namely N =4, N =2 is assumed, if the CPU utilization rate and the memory utilization rate of an MEC content server are smaller than 80%, namely two load indexes are smaller than 80%, the MEC content server meets a load condition, and if only the CPU utilization rate is smaller than 80%, the memory utilization rate, the disk read-write rate and the network card rate are all larger than 80%, namely only one load index is smaller than 80%, the MEC content server does not meet the load condition.

Furthermore, the MEC scheduling server may store the delay information and the load information of the MEC content server, and when the performance of the MEC content server is determined, the stored delay information and the stored load information of the MEC content server may be directly retrieved from the scheduling server.

Specifically, still include:

the MEC scheduling server may obtain delay information and load information of each MEC content server connected to the MEC scheduling server, and update the delay information and load information of the MEC content servers stored in the MEC scheduling server.

The MEC scheduling server can periodically or aperiodically acquire the load information and the time delay information of the MEC content server.

In addition, when the performance of the MEC content server is judged, the load information and the time delay information can be directly obtained from the related MEC content server.

The manner of acquiring the delay information and the load information by the MEC scheduling server may include multiple manners, which are not limited in this embodiment, and may include:

sending a message to the MEC content server by adopting a ping command;

receiving load information sent by an MEC content server; the load information of the MEC content server is collected by the MEC content server according to preset instructions.

The MEC content server collects each load index of the MEC content server regularly or irregularly and sends the collected load indexes to the MEC scheduling server; for the acquisition of the delay information, the MEC scheduling server may send a message to the MEC content server by using a ping command periodically or aperiodically, and the MEC content server may feed back the time length of the delay to the MEC scheduling server.

As can be seen from the above description, in this embodiment, the performance of the server is represented by the load and the time delay of the MEC content server, but the present invention is not limited to using the load and the time delay, and other indexes that can be used to represent the performance of the MEC content server may also be used.

S404: and sending the IP address of the MEC content server which meets the preset rule to a user through the base station.

In this embodiment, after receiving the IP address of the MEC content server sent by the base station, the user may access the MEC content server according to the IP address, and further access the service resource required by the user in the MEC content server.

In this embodiment, the MEC content server is deployed in the scheduling network, so that a CDN node does not need to be allocated to a user through an external network, and access to an internet service can also be achieved. Moreover, the introduction shows that the MEC scheduling server is closer to the user, so that a plurality of MEC content servers connected with the MEC scheduling server are closer to the user, and performance factors are considered when the MEC content servers are selected, so that the network delay is reduced, and the fluency degree of the user when accessing the Internet is improved.

Referring to fig. 5, a schematic structural diagram of a scheduling apparatus provided in an embodiment of the present invention is shown, where the apparatus is applied to a MEC scheduling server in a scheduling network, and the apparatus includes:

a first receiving unit 501, configured to receive a first DNS request of a user sent by a base station, and resolve domain name information from the first DNS request;

a second DNS request generating unit 502, configured to generate a second DNS request including an IP address of the MEC scheduling server if the domain name information is not stored in the MEC scheduling server;

a first sending unit 503, configured to send the second DNS request to the GSLB scheduling server through the core network;

a second receiving unit 504, configured to receive an IP address of a CDN node allocated by a GSLB scheduling server for a user, where the IP address is sent by a core network; the CDN node is distributed for the user by the GSLB dispatching server according to the IP address of the MEC dispatching server;

a second sending unit 505, configured to send the IP address of the CDN node allocated to the user through the base station.

Optionally, the method further includes:

Optionally, the MEC content server selecting unit is configured to:

and if the MEC content server comprises a plurality of MEC content servers which have mapping relations with the domain name information, selecting an MEC content server with the optimal performance from the MEC content servers which have mapping relations with the domain name information according to the load information and the time delay information of the MEC content servers, and distributing the MEC content server with the optimal performance to the user.

Optionally, the scheduling apparatus further includes:

Optionally, the obtaining of the time delay information and the load information of each MEC content server connected to the MEC scheduling server includes:

Optionally, the load information includes:

By the device of the embodiment, when scheduling is performed by the MEC scheduling server, if the MEC scheduling server does not store domain name information of the first DNS request, a second DINS request including an IP address of the MEC scheduling server is regenerated, and the second DNS request is sent to the GSLB scheduling server through the core network, so that the GSLB scheduling server allocates CDN nodes to users according to the IP address of the MEC scheduling server. Therefore, as the MEC has the characteristic of being closer to the user, the IP address of the MEC scheduling server is closer to the user, and therefore the GSLB scheduling server allocates CDN nodes to the user according to the IP address of the MEC scheduling server, and is also closer to the user. Therefore, the CDN nodes distributed for the users in the mode greatly shorten the distance between the users and the CDN nodes, reduce the network delay and further improve the experience of the users on Internet services.

Scheduling servers for GSLB:

referring to fig. 6, another flow chart of a scheduling method provided by an embodiment of the present invention is shown, and the method is applied to the GSLB scheduling server in the scheduling network described above, where the scheduling network is shown in fig. 1 and includes: the system comprises a base station, an MEC scheduling server, a core network and a GSLB scheduling server;

the method comprises the following steps:

s601: receiving a second DNS request sent by a core network, and resolving an IP address of an MEC scheduling server contained in the second DNS request;

the second DNS request is generated on an MEC dispatching server, and after receiving a first DNS request of a user sent by a base station, the MEC dispatching server regenerates the second DNS request containing the IP address of the MEC dispatching server on the MEC dispatching service if the MEC dispatching server does not store the domain name information of the first DNS request.

S602: according to the IP address of the MEC scheduling server, distributing CDN nodes for users;

when a CDN node is allocated to a user through the GSLB scheduling service, a CDN node closest to an IP address of the MEC scheduling server is allocated to the user in principle.

However, in some special cases, for example, the user access amount of the CDN node closest to the IP address of the MEC scheduling server is large, and even if the CDN node closest to the IP address of the MEC scheduling server is allocated to the user, a stuck situation may occur. In order to solve the above problem, the GSLB scheduling server may allocate CDN nodes to users according to a performance optimization principle, and ensure that CDN nodes with better performance are allocated to users.

Based on the performance optimization principle, S602 may include multiple ways to allocate CDN nodes to users, for example, the following implementation ways may be included:

the first method,

if the access amount of the first CDN node exceeds a preset threshold value, acquiring at least one second CDN node of which the distance from the IP address of the MEC scheduling server is within a preset distance range;

In the present embodiment, the first and second electrodes are,

the second method,

Determining at least one third CDN node with a distance between the third CDN node and the IP address of the MEC scheduling server within a preset second distance range according to the IP address of the MEC scheduling server;

and selecting the CDN node with the lowest access amount from the at least one third CDN node, and distributing the selected CDN node with the lowest access amount to the user.

The preset distance range and the second distance range may be specific distance lengths, or interval values of the distance lengths. For example, the preset distance range or the second distance range is 10km, or the second distance range is [10km,20km ].

The access amount of the CDN node mentioned in this embodiment may be represented by multiple indexes, for example, the access amount may be represented by a use condition of a load, for example, a CPU use rate, a memory use rate, a disk read-write rate, a network card rate, and the like.

The GSLB scheduling server calls load information of the CDN nodes, and when determining the access amount, the GSLB scheduling server may include the following two methods:

the method comprises the steps that pre-stored load information of CDN nodes is directly called;

the GSLB scheduling server can obtain the load use condition of the CDN node in advance, and when the access amount of the CDN node needs to be judged according to the load use condition, the stored load use condition of the CDN node is directly called. In this case, the GSLB scheduling server may periodically obtain load information of the CDN nodes.

Acquiring current load information of the CDN node from the CDN node;

the GSLB scheduling server does not pre-store the load use condition of the CDN node, and when the load use condition of the CDN node needs to be judged, the load use condition of the CDN node is obtained.

S603: and sending the IP address of the CDN node distributed for the user to an MEC scheduling server through a core network.

The method comprises the steps that the IP address of a CDN node distributed for a user is sent to an MEC scheduling server through a core network, the MEC scheduling server sends the IP address of the CDN node to the user through a base station, the user can access the CDN node according to the IP address, and therefore the purpose of accessing service resources to be accessed by the user in the CDN node is achieved.

In this embodiment, when a CDN node is allocated to a user by the GSLB scheduling server, the CDN node allocated to the user according to the IP address of the MEC scheduling server is closer to the user because the MEC scheduling server is closer to the user. And moreover, CDN nodes are distributed for the user according to the performance optimization principle, so that the user can experience smoother internet services, and the experience degree of the user on the internet services is further improved.

Referring to fig. 7, a schematic diagram of another structure of a scheduling apparatus provided in an embodiment of the present invention is shown, where the apparatus is applied to a GSLB scheduling server in a scheduling network, and in this embodiment, the apparatus includes:

a third receiving unit 701, configured to receive a second DNS request sent by a core network, and resolve an IP address of the MEC scheduling server in the second DNS request; the second DNS request is generated by an MEC dispatch server;

an allocating unit 702, configured to allocate CDN nodes for a user according to the IP address of the MEC scheduling server;

a fourth sending unit 703 is configured to send the IP address of the CDN node allocated to the user to the MEC scheduling server through the core network.

Optionally, the allocating unit is configured to:

Based on the device of the embodiment, when the GSLB scheduling server allocates CDN nodes to the user, the MEC scheduling server is closer to the user, and the CDN nodes allocated to the user according to the IP address of the MEC scheduling server are closer to the user. And moreover, CDN nodes are distributed for the user according to the performance optimization principle, so that the user can experience smoother internet services, and the experience degree of the user on the internet services is further improved.

Referring to fig. 8, an interaction diagram of a scheduling method provided in an embodiment of the present invention is shown, where the method is based on the scheduling network introduced in fig. 2, and in this embodiment, the method includes:

s801: a user sends a first DNS request to an MEC scheduling server through a base station;

s802: after receiving the first DNS request, the MEC scheduling server analyzes domain name information in the first DNS request;

s803: judging whether the MEC scheduling server stores domain name information or not;

s804: if the MEC scheduling server stores domain name information, acquiring an MEC content server which has a mapping relation with the domain name information;

s805: selecting at least one MEC content server which accords with a preset rule from at least one content server which has a mapping relation with the domain name information and distributing the content server to a user;

s806: the IP address of the MEC content server which accords with the preset rule is sent to a user through a base station;

s807: if the domain name information is not stored in the MEC dispatching server, generating a second DNS request containing the IP address of the MEC dispatching server;

s808: sending the second DNS request to a GSLB scheduling server through a core network;

s809: after receiving the second NDS request, the GSLB scheduling server analyzes the IP address of the MEC scheduling server contained in the second DNS request, and allocates CDN nodes for the user according to the IP address of the MEC scheduling server;

s810: the GSLB dispatching server sends the IP address of the CDN node distributed for the user to the MEC dispatching server through the core network;

s811: and the MEC scheduling server sends the IP address of the CDN node distributed for the user to the user through the base station.

Further, it should be noted that the service resources mentioned above may include any service that is accessed through the internet, such as video resources, software resources, game resources, and the like.

Based on the implementation of S801 to S811, the present solution is described by specific implementation scenarios:

assuming that the service resources are video resources:

a terminal user sends a DNS request for accessing a certain video to an MEC scheduling server through a base station;

the MEC scheduling server judges according to the DNS request, and if the MEC scheduling server contains domain name information in the DNS request (for example, the video is a hot video), the MEC scheduling server determines an IP address of an MEC content server which has a mapping relation with the DNS request;

the MEC scheduling server selects an MEC content server which accords with a preset rule from at least one MEC content server which has a mapping relation with the domain name information, or can be understood as selecting an MEC content server with the optimal performance;

the MEC scheduling server sends the IP address of the MEC content server which accords with the preset rule to a terminal user through a base station;

if the MEC scheduling server does not contain the domain name information (for example, the video is a cold video), regenerating a DNS request containing the IP address of the MEC scheduling server, and sending the regenerated DNS request to the GSLB server;

and the GSLB dispatching server analyzes the received DNS request, analyzes the IP address of the MEC dispatching server, and distributes CDN nodes for the user according to the IP address of the MEC dispatching server.

For example, a CDN node closest to the IP address of the MEC scheduling server may be allocated, or a CDN node with better performance may be allocated in consideration of both the distance and the performance.

The GSLB dispatching server sends CDN nodes distributed for users to the MEC dispatching server through a core network;

and the MEC scheduling server sends the IP address of the CDN node allocated to the user to the terminal user through the base station.

In this embodiment, when performing network scheduling by the above method, the MEC scheduling server regenerates a DNS request including an IP address of the MEC scheduling server when the MEC scheduling server does not include domain name information, so that the GSLB scheduling server may allocate a CDN node to a user according to the IP address of the MEC scheduling server. In this way, since the MEC has a characteristic of being closer to the user, the IP address of the MEC scheduling server is also closer to the user, and then the CDN node allocated by the MEC scheduling server is also closer to the user. Therefore, the problem that in the prior art, a CDN node which is allocated to a user through an IP address of a Local DNS server is far away from the user, so that network delay is caused is solved.

It should be noted that, in the present specification, the embodiments are all described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments may be referred to each other.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A scheduling method, applied to a MEC scheduling server in a scheduling network, the scheduling network comprising:

the method comprises the following steps:

receiving a first DNS request of a user sent by a base station, and resolving domain name information from the first DNS request;

if the domain name information is not stored in the MEC dispatching server, generating a second DNS request containing the IP address of the MEC dispatching server, wherein the second DNS request is generated on the MEC dispatching server;

sending the second DNS request to the GSLB dispatching server through the core network;

receiving an IP address of a CDN node distributed for a user by a GSLB scheduling server sent by a core network; the CDN node is a CDN node that is allocated by the GSLB scheduling server to the user according to the IP address of the MEC scheduling server and is closest to the IP address of the MEC scheduling server, where the GSLB scheduling server allocates to the user according to the IP address of the MEC scheduling server a CDN node that is closest to the IP address of the MEC scheduling server, and includes: determining at least one third CDN node with a distance between the third CDN node and the IP address of the MEC scheduling server within a preset second distance range according to the IP address of the MEC scheduling server; selecting a CDN node with the lowest access amount from at least one third CDN node, and distributing the selected CDN node with the lowest access amount to a user, wherein the access amount is characterized by a CPU (Central processing Unit) utilization rate, a memory utilization rate, a disk read-write rate and a network card rate;

the method comprises the steps that IP addresses of CDN nodes distributed for users are sent to the users through a base station;

2. The method according to claim 1, wherein selecting an MEC content server meeting a preset rule from at least one MEC content server having a mapping relationship with the domain name information comprises:

3. The method according to claim 2, wherein the selecting an MEC content server with the best performance from a plurality of MEC content servers mapped with the domain name information according to the load information and the time delay information of the MEC content server comprises:

4. The method of claim 1, further comprising:

and acquiring the time delay information and the load information of each MEC content server connected with the MEC scheduling server, and updating the time delay information and the load information of the MEC content servers stored in the MEC scheduling server.

5. The method of claim 4, wherein the obtaining latency information and load information of each MEC content server connected to the MEC scheduling server comprises:

sending a message to the MEC content server by using a ping command;

receiving time delay information fed back by the MEC content server based on the message;

6. A scheduling apparatus, wherein the scheduling apparatus is applied to a MEC scheduling server in a scheduling network, and the scheduling network includes: base station, MEC dispatch server, core network and GSLB dispatch server, the said scheduling device includes:

a second DNS request generating unit, configured to generate a second DNS request including an IP address of the MEC scheduling server if the domain name information is not stored in the MEC scheduling server, where the second DNS request is generated on the MEC scheduling server;

the second receiving unit is used for receiving the IP address of the CDN node distributed by the GSLB dispatching server for the user and sent by the core network; the CDN node is a CDN node that is allocated by the GSLB scheduling server to the user according to the IP address of the MEC scheduling server and is closest to the IP address of the MEC scheduling server, where the GSLB scheduling server allocates to the user according to the IP address of the MEC scheduling server a CDN node that is closest to the IP address of the MEC scheduling server, and includes: determining at least one third CDN node with a distance between the third CDN node and the IP address of the MEC scheduling server within a preset second distance range according to the IP address of the MEC scheduling server; selecting a CDN node with the lowest access amount from at least one third CDN node, and distributing the selected CDN node with the lowest access amount to a user, wherein the access amount is characterized by a CPU (central processing unit) utilization rate, a memory utilization rate, a disk read-write rate and a network card rate;

the second sending unit is used for sending the IP address of the CDN node distributed for the user to the user through the base station;

the MEC content server selecting unit is used for selecting an MEC content server which accords with a preset rule from at least one MEC content server which has a mapping relation with the domain name information and distributing the server to the user;

7. A scheduling method, wherein the method is applied to a GSLB scheduling server in a scheduling network, and the scheduling network comprises:

the method comprises the following steps:

receiving a second DNS request sent by the core network, and analyzing an IP address of the MEC scheduling server contained in the second DNS request; the second DNS request is generated by an MEC dispatch server;

according to the fact that the user is allocated with the IP address of the MEC scheduling server, the CDN node which is closest to the IP address of the MEC scheduling server comprises: determining at least one third CDN node with a distance between the third CDN node and the IP address of the MEC scheduling server within a preset second distance range according to the IP address of the MEC scheduling server; selecting a CDN node with the lowest access amount from at least one third CDN node, and distributing the selected CDN node with the lowest access amount to a user, wherein the access amount is characterized by a CPU (central processing unit) utilization rate, a memory utilization rate, a disk read-write rate and a network card rate;

8. The method of claim 7, wherein the allocating CDN nodes for users according to the IP addresses of the MEC scheduling servers comprises:

if the access amount of the first CDN node exceeds a preset threshold value, acquiring at least one second CDN node of which the distance to the IP address of the MEC scheduling server is within a preset distance range, wherein the preset distance range and the second distance range can be the same or different;

9. A scheduling apparatus, wherein the scheduling apparatus is applied to a GSLB scheduling server in a scheduling network, and the scheduling network comprises: base station, MEC dispatch server, core network and GSLB dispatch server, the said scheduling device includes:

an allocation unit, configured to allocate, according to the IP address of the MEC scheduling server, a CDN node closest to the IP address of the MEC scheduling server to a user, where allocating, according to the IP address of the MEC scheduling server, a CDN node closest to the IP address of the MEC scheduling server to the user includes: determining at least one third CDN node of which the distance from the IP address of the MEC scheduling server is within a preset second distance range according to the IP address of the MEC scheduling server; selecting a CDN node with the lowest access amount from at least one third CDN node, and distributing the selected CDN node with the lowest access amount to a user, wherein the access amount is characterized by a CPU (central processing unit) utilization rate, a memory utilization rate, a disk read-write rate and a network card rate;

and the fourth sending unit is used for sending the IP address of the CDN node allocated to the user to the MEC scheduling server through the core network.

10. A dispatch network, the network comprising:

the MEC scheduling server is used for analyzing domain name information contained in the first DNS request, regenerating a second DNS request containing an IP address of the MEC scheduling server under the condition that the MEC scheduling server does not store the domain name information, sending the second DNS request to the GSLB scheduling server through a core network, sending the IP address of a CDN node to a user through a base station after receiving the IP address of the CDN node distributed for the user, and acquiring at least one MEC content server which has a mapping relation with the domain name information if the MEC scheduling server stores the domain name information; business resources are cached in the MEC content server; selecting an MEC content server which accords with a preset rule from at least one MEC content server which has a mapping relation with the domain name information and distributing the MEC content server to a user; sending the IP address of the MEC content server which meets the preset rule to a user through a base station, wherein the second DNS request is generated on the MEC scheduling server;

the GSLB scheduling server is configured to allocate, according to the IP address of the MEC scheduling server, a CDN node closest to the IP address of the MEC scheduling server, and send the allocated CDN node IP address to the MEC scheduling server through a core network, where the allocating CDN node closest to the IP address of the MEC scheduling server for a user according to the IP address of the MEC scheduling server includes: determining at least one third CDN node of which the distance from the IP address of the MEC scheduling server is within a preset second distance range according to the IP address of the MEC scheduling server; and selecting a CDN node with the lowest access amount from the at least one third CDN node, and distributing the selected CDN node with the lowest access amount to the user, wherein the access amount is characterized by a CPU (Central processing Unit) utilization rate, a memory utilization rate, a disk read-write rate and a network card rate.

11. The network of claim 10, wherein the MEC-based scheduling network further comprises:

12. The network of claim 11, wherein the MEC scheduling server is further configured to: