Edge computing power resource scheduling method and system based on terminal triggering

Abstract

The invention discloses an edge computing power resource scheduling method and system based on terminal triggering, and solves the problem of unified management of edge individuals, edge coordinationand the like under an edge computing framework by capturing terminal user data in real time and triggering computing power resource scheduling. Cross-edge node scheduling of terminal user computing power resource application can be rapidly carried out in real time, so that computing power resources can be called more fully and flexibly. The method has the advantages that the request of users for the computing power resources can be captured in real time through terminal equipment, the computing power resources are quickly found on an edge computing power resource pool node and other edge computing power pool nodes close to the edge computing power resource pool node, and the matched computing power resources are allocated.

Description

technical field [0001] The invention belongs to the technical field of edge computing, and in particular relates to a terminal-triggered edge computing power resource scheduling method and system. Background technique [0002] The number of devices and the data generated at the edge of the current network are growing rapidly, and the centralized processing model centered on the cloud computing model will not be able to efficiently process the data generated by edge devices. The centralized processing mode transmits all data to the cloud computing center through the network, and uses the super computing power of the cloud computing center to solve computing and storage problems in a centralized manner, which enables cloud services to create higher economic benefits, but there are the following technologies Problems and defects: (1) Insufficient real-time performance; (2) Insufficient bandwidth; (3) Large energy consumption; (4) Large delay and other issues. In this case, pro...

AI Technical Summary

Problems solved by technology

[0002] The number of devices and the data generated at the edge of the current network are growing rapidly, and the centralized processing model centered on the cloud computing model will not be able to efficiently process the data generated by edge devices

The centralized processing mode transmits all data to the cloud computing center through the network, and uses the super computing power of the cloud computing center to solve computing and storage problems in a centralized manner, which enables cloud services to create higher economic benefits, but there are the following technologies Problems and defects: (1) Insufficient real-time performance; (2) Insufficient bandwidth; (3) Large energy consumption; (4) Large delay, etc.

In actual application scenarios, edge computing is often used for local data management and local computing power resource scheduling. Computing power resource scheduling between cross-edge nodes is generally managed by the cloud center, and edge nodes rarely have computing power resource scheduling capabilities between edge nodes. , in some scenarios where there are computing resource scheduling requests among edge computing nodes, the edge node data is uploaded to the cloud center, and the cloud computing center conducts unified analysis of the data calculation results and handles the collaborative scheduling of computing power resources between edge nodes. It will lead to a large bandwidth consumption between the edge node and the cloud computing center, the real-time performance of the business cannot meet the requirements, and there will be problems such as the edge computing node cannot work normally when the network of the cloud computing center is unreachable.

[0005] The computing power scheduling of existing edge cloud computing platforms is generally realized through two types of algorithms. One type of algorithm is based on two-factor estimation of business model and user scale, and the other type is to bind terminal devices and edge computing devices; The technical disadvantages of the edge computing platform’s computing power scheduling algorithm include: (1) During the project life cycle of the business model, the demand for computing power is relatively stable, and the estimation of the computing power demand can be treated according to the constant k. For different business models k Different values, such as storage, encryption and decryption, image recognition, and VR video rendering, etc., are cumbersome to operate; (2) The user scale is within the project life cycle, and the growth trend fluctuates little, which has a great impact on the long-term estimation of computing power requirements. For example, the estimation cycle is based on months and years, but it cannot meet the estimation requirements for short-term computing power requirements, such as the requirements for estimating cycles such as hours, days, and weeks; (3) Factors that affect computing power requirements in the production environment There are many, such as user behavior preferences, scale of business promotion activities, holidays, hot events, weather, traffic and network failures, etc., the existing computing power demand estimation algorithm can no longer meet the actual computing power scheduling requirements; (4) the terminal equipment and the edge In the case of binding computing devices, computing power cannot be flexibly allocated and scheduled, resulting in insufficient satisfaction of high-level business computing power requirements, and insufficient low-level business computing power requirements resulting in waste of computing power

Images