A network bandwidth-based edge computing computing power scheduling method, device and system

Abstract

The application relates to the technical field of computing power network scheduling, in particular to an edge computing computing power scheduling method, device and system based on network bandwidth, which comprises the following steps: a computing power scheduling system binds application services and domain names; a client accesses an application to perform EDNS request; the computing power scheduling system performs DPI analysis on the EDNS request, and confirms the IP address of the client; the computing power scheduling system combines the computing power resource distribution situation of the system itself and the IP address of the client; beneficial effects are as follows: the edge computing computing power scheduling method, device and system based on network bandwidth mainly face the computing power scheduling of edge computing, utilize the EDNS protocol characteristics and network bandwidth probe technology, calculate the nearest computing power node of the network bandwidth satisfying the demand of the client through an algorithm, deploy relevant services, and automatically guide the access flow of the client to the new computing power node, so that the real-time performance of the service is improved, and the response time of the service is shortened.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computing power network scheduling, in particular to an edge computing computing power scheduling method, device and system based on network bandwidth. BACKGROUND

[0002] The rise of edge computing is due to the development of technologies such as the Internet of Things and artificial intelligence, which makes more and more devices need to have computing and storage capabilities.

[0003] In the prior art, edge devices have many advantages, such as fast response speed, low network delay, and can work offline, etc.

[0004] However, edge devices have limited resources and different computing capabilities, so the computing power scheduling of edge computing is very important. A reasonable computing power scheduling strategy can maximize the utilization of edge device resources, thereby improving the response speed and computing efficiency of computing tasks, meeting user network bandwidth requirements, and reducing network delay and other problems. SUMMARY

[0005] The purpose of the present application is to provide an edge computing computing power scheduling method, device and system based on network bandwidth to solve the problems raised in the background art.

[0006] To achieve the above purpose, the present application provides the following technical scheme: an edge computing computing power scheduling method based on network bandwidth, the scheduling method comprising the following steps:

[0007] The computing power scheduling system binds the application service and the domain name;

[0008] The client accesses the application to make an EDNS request;

[0009] The computing power scheduling system performs DPI analysis on the EDNS request and confirms the IP address of the client;

[0010] The computing power scheduling system combines its own computing power resource distribution and the IP address of the client, and uses an algorithm to calculate the nearest computing power node that meets the network bandwidth requirement of the client;

[0011] The computing power scheduling system deploys the application at the calculated node;

[0012] The client makes an EDNS resolution request again.

[0013] Preferably, when the computing power scheduling system deploys the application at the calculated node, it modifies the EDNS server configuration.

[0014] Preferably, when the client accesses the application to make an EDNS request, the EDNS server returns the IP address of the deployed service;

[0015] When the client performs EDNS resolution request again, the EDNS server returns the IP address of the nearest computing power service meeting the network bandwidth requirement of the client.

[0016] The edge computing computing power scheduling device based on network bandwidth comprises a binding module, an access module, an analysis module, a calculation module, a deployment module and a resolution module.

[0017] The binding module is used for binding the application service and the domain name by the computing power scheduling system.

[0018] The access module is used for the client to access the application to perform EDNS request.

[0019] The analysis module is used for the computing power scheduling system to perform DPI analysis on the EDNS request and confirm the IP address of the client.

[0020] The calculation module is used for the computing power scheduling system to calculate the nearest computing power node meeting the network bandwidth requirement of the client by combining the computing power resource distribution condition of the system and the IP address of the client.

[0021] The deployment module is used for the computing power scheduling system to deploy the application at the calculated node.

[0022] The resolution module is used for the client to perform EDNS resolution request again.

[0023] Preferably, the deployment module is further used for modifying the EDNS server configuration.

[0024] The edge computing computing power scheduling system based on network bandwidth comprises a client and an EDNS server. The client starts EDNS resolution, the EDNS resolution address points to the EDNS server, and the IP address configuration of the application instance accessed by the client is the bound domain name. When the client accesses the application instance for the first time, the client first performs EDNS domain name resolution, that is, sends the domain name resolution to the EDNS server. The request carries the ECS information of the client. The EDNS server returns the IP address of the deployed application instance.

[0025] Preferably, after the first EDNS domain name request of the client is sent to the EDNS server, the computing power scheduling system performs DPI analysis on the EDNS protocol, analyzes the source address, the destination address, the source port, the destination port and the ECS information of the client.

[0026] Preferably, the computing power scheduling system determines the address of the client, i.e. the source address of the EDNS request packet, through analysis of the domain name and the client ECS information and the source address, and then uses its own computing power resources to perform network bandwidth detection on the client using an algorithm to quickly determine the nearest computing power node that meets the network bandwidth requirement of the client.

[0027] Preferably, the computing power scheduling system deploys a service related to the domain name to the calculated computing power node to form a second application instance of the service, modifies the configuration of the EDNS server end to configure the IP of the second application instance of the service to the server end, and the client performs domain name resolution to the EDNS server end for the second time carrying the ECS information, and the EDNS server end returns the IP information of the second application instance.

[0028] Preferably, before the client starts EDNS resolution, the edge computing computing power scheduling system optionally deploys an application instance to a computing power node, binds a certain application and domain name, and points the IP in the configuration of the EDNS server end to the deployed application instance.

[0029] Compared with the prior art, the present application has the following beneficial effects:

[0030] The edge computing computing power scheduling method, device and system based on network bandwidth provided by the present application mainly face the computing power scheduling of edge computing, use the EDNS protocol characteristics and network bandwidth probe technology, calculate the nearest computing power node that meets the network bandwidth requirement of the client through an algorithm, deploy related services, and automatically guide the access traffic of the client to the new computing power node, thereby improving the real-time performance of the service and shortening the response time of the service. BRIEF DESCRIPTION OF DRAWINGS

[0031] Figure 1 The method flowchart of the present application. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical scheme of the present application clear, complete and the advantages more clear and obvious, the embodiments of the present application are further described in detail below in combination with the drawings. It should be understood that the specific embodiments described herein are part of the embodiments of the present application, not all embodiments, and are only used to explain the embodiments of the present application, and do not limit the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0033] Embodiment one

[0034] Please refer to Figure 1The application provides a technical solution: an edge computing computing power scheduling method based on network bandwidth, which comprises the following steps:

[0035] The computing power scheduling system binds application services and domain names;

[0036] The client accesses the application to make an EDNS request, and the EDNS server returns the IP address of the deployed service;

[0037] The computing power scheduling system performs DPI analysis on the EDNS request and confirms the IP address of the client;

[0038] The computing power scheduling system combines the distribution of its own computing power resources and the IP address of the client, and uses an algorithm to calculate the nearest computing power node that meets the network bandwidth demand of the client;

[0039] The computing power scheduling system deploys the application on the calculated node and modifies the EDNS server configuration;

[0040] The client makes an EDNS resolution request again, and the EDNS server returns the IP address of the nearest computing power service that meets the network bandwidth demand of the client.

[0041] Embodiment two

[0042] An edge computing computing power scheduling device based on network bandwidth, the scheduling device comprises a binding module, an access module, an analysis module, a calculation module, a deployment module and a resolution module;

[0043] The binding module is used for the computing power scheduling system to bind application services and domain names;

[0044] The access module is used for the client to access the application to make an EDNS request;

[0045] The analysis module is used for the computing power scheduling system to perform DPI analysis on the EDNS request and confirm the IP address of the client;

[0046] The calculation module is used for the computing power scheduling system to combine the distribution of its own computing power resources and the IP address of the client, and use an algorithm to calculate the nearest computing power node that meets the network bandwidth demand of the client;

[0047] The deployment module is used for the computing power scheduling system to deploy the application on the calculated node and modify the EDNS server configuration;

[0048] The resolution module is used for the client to make an EDNS resolution request again.

[0049] Embodiment three

[0050] The application discloses an edge computing computing power scheduling system based on network bandwidth, and relates to the technical field of edge computing.

[0051] Embodiment four

[0052] The application discloses an edge computing computing power scheduling method based on network bandwidth.

[0053] Step S1, an edge computing computing power scheduling system deploys an application instance on an optional computing power node.

[0054] Step S2, the edge computing computing power scheduling system binds an application and a domain name, and configures an IP of an EDNS server to point to the deployed application instance.

[0055] Step S3, a client starts EDNS resolution, an EDNS resolution address points to an EDNS server, and an IP address of the client accessing an application instance is configured as a bound domain name.

[0056] Step S4, when the client accesses the application instance for the first time, EDNS domain name resolution is first performed, that is, domain name resolution is sent to the EDNS server, and the request carries ECS information of the client.

[0057] Step S5, the EDNS server returns an IP address of the deployed application instance.

[0058] Step S6, after the first EDNS domain name request of the client is sent to the EDNS server, the computing power scheduling system performs DPI analysis on the EDNS protocol, and analyzes a source address, a destination address, a source port, a destination port, ECS information of the client and the like.

[0059] Step S7, the computing power scheduling system determines an address of the client (that is, a source address of an EDNS request packet) through analysis of the domain name and ECS information and the source address of the client, and then uses an algorithm to perform network bandwidth detection on the client by using computing power resources of the computing power scheduling system, so as to quickly determine a nearest computing power node that meets a network bandwidth requirement of the client.

[0060] Step S8, the computing power scheduling system deploys a service related to the domain name to the calculated computing power node, to form a second application instance of the service.

[0061] Step S9, the computing power scheduling system modifies the configuration of the EDNS service end, and configures the IP of the second application instance of the service to the service end.

[0062] Step 10, the client carries the ECS information to the EDNS service end for the second time to resolve the domain name, and the EDNS service end returns the IP information of the second application instance.

[0063] Although the embodiments of the present application have been shown and described, it can be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A network bandwidth-based edge computing computing power scheduling method, characterized in that: The scheduling method comprises the following steps: The computing power scheduling system binds the application service and the domain name; The client accesses the application to make an EDNS request, and the EDNS server returns the IP address of the deployed service; The computing power scheduling system performs DPI analysis on the EDNS request and confirms the IP address of the client; The computing power scheduling system combines the distribution of its own computing power resources and the IP address of the client, and calculates the nearest computing power node that meets the network bandwidth requirement of the client by using an algorithm; The computing power scheduling system deploys the application on the calculated node and modifies the EDNS server configuration; The client makes an EDNS resolution request again, and the EDNS server returns the IP address of the nearest computing power service that meets the network bandwidth requirement of the client.

2. The network bandwidth-based edge computing computing power scheduling method according to claim 1, characterized in that: The computing power scheduling system modifies the EDNS server configuration when deploying the application on the calculated node.

3. The network bandwidth-based edge computing computing power scheduling method of claim 1, wherein: When the client accesses the application to make an EDNS request, the EDNS server returns the IP address of the deployed service; When the client makes an EDNS resolution request again, the EDNS server returns the IP address of the nearest computing power service that meets the network bandwidth requirement of the client.

4. A network bandwidth-based edge computing computing power scheduling device for implementing the network bandwidth-based edge computing computing power scheduling method of any one of claims 1-3. The scheduling device comprises a binding module, an access module, an analysis module, a calculation module, a deployment module and a resolution module; The binding module is configured to bind the application service and the domain name by the computing power scheduling system. The access module is configured to access the application by the client to make an EDNS request. The analysis module is configured to perform DPI analysis on the EDNS request by the computing power scheduling system and confirm the IP address of the client. The calculation module is configured to combine the distribution of the computing power resources of the computing power scheduling system and the IP address of the client, and calculate the nearest computing power node that meets the network bandwidth requirement of the client by using an algorithm. The deployment module is configured to deploy the application on the calculated node by the computing power scheduling system. The resolution module is configured to make an EDNS resolution request again by the client. 5.The network bandwidth-based edge computing computing power scheduling apparatus according to claim 4, characterized in that: The deployment module is further configured to modify the EDNS server configuration.

6. A network bandwidth-based edge computing computing power scheduling system for implementing the network bandwidth-based edge computing computing power scheduling method of any one of claims 1-3. The edge computing computing power scheduling system comprises a client and an EDNS server, the client starts EDNS resolution, the EDNS resolution address points to the EDNS server, and the IP address of the application instance accessed by the client is configured as the bound domain name; The computing power scheduling system binds the application service and the domain name; the client accesses the application to make an EDNS request, and the EDNS server returns the IP address of the deployed service; the computing power scheduling system performs DPI analysis on the EDNS request and confirms the IP address of the client; the computing power scheduling system combines the distribution of its own computing power resources and the IP address of the client, and calculates the nearest computing power node that meets the network bandwidth requirement of the client by using an algorithm; The computing power scheduling system deploys the application on the calculated node and modifies the EDNS server configuration; The client makes an EDNS resolution request again, and the EDNS server returns the IP address of the nearest computing power service that meets the network bandwidth requirement of the client. 7.The network bandwidth-based edge computing computing power scheduling system according to claim 6, characterized in that: After the first EDNS domain name request of the client is sent to the EDNS server, the computing power scheduling system analyzes the EDNS protocol by DPI to analyze the source address, destination address, source port, destination port, and ECS information of the client. 8.The network bandwidth based edge computing computing power scheduling system according to claim 7, characterized in that: The computing power scheduling system determines the address of the client through analysis of the domain name, ECS information, and source address of the client, and then uses its own computing power resources to perform network bandwidth detection on the client using an algorithm to quickly determine the nearest computing power node that meets the network bandwidth requirements of the client. 9.The network bandwidth-based edge computing computing power scheduling system according to claim 8, characterized in that: The computing power scheduling system deploys the application service related to the domain name to the calculated computing power node to form a second application instance of the service, modifies the configuration of the EDNS server, configures the IP of the second application instance of the service to the server, and performs domain name resolution on the EDNS server with the ECS information carried by the client for the second time. 10.The network bandwidth based edge computing computing power scheduling system according to claim 6, characterized in that: Before the client starts EDNS resolution, the edge computing power scheduling system deploys an application instance on a selected computing power node, binds the application to the domain name, and configures the IP of the EDNS server to point to the deployed application instance.

Images