Edge-buffer management method, personal cloud system, and computer readable storage medium

Abstract

The embodiment of the invention provides an edge buffer management method, a personal cloud system and a computer-readable storage medium. The method comprises the following steps: obtaining user-perceived big data, service-perceived big data and network-perceived big data; predicting the movement trajectories of users with different time scales and different spatial granularities based on the user perceived big data, mining, predicting and associating the user behavior patterns and the potential business requirements based on the service perceived big data; predicting a user's future spatio-temporal two-dimensional service demand based on the movement trajectory of the user with different time scales and different spatial granularities and the user's behavior mode and service potential demand; edge caching is set based on the user's future spatio-temporal two-dimensional service requirements and the network-aware big data.

Description

technical field [0001] The invention relates to the technical field of mobile communication, in particular to an edge cache management method, a personal cloud system and a computer-readable storage medium. Background technique [0002] Traditional content distribution network (Content Distribution Network, CDN) edge nodes are usually deployed in the metropolitan area network. For mobile users, they need to pass through multiple network devices such as base stations, aggregation switches, and gateways. The transmission path is long and the backhaul Bandwidth pressure is high. Therefore, it is necessary to further sink the CDN nodes to the core network, transport network or wireless access network, that is, to introduce the mobile content distribution network (mobile Content Distribution Network, mCDN) or mobile edge computing (Mobile Edge Computing, MEC), to provide users with high Availability and high-performance content services while optimizing content delivery for wire...

AI Technical Summary

Problems solved by technology

However, existing content networks (Cache / CDN / mCDN / IDC) and edge computing nodes (MEC) are often deployed in a centralized manner based on hotspots, making it difficult to reflect differentiated demands and differentiated services, and the introduction of resources is unevenly distributed; Provides normal acceleration services, but is prone to errors in dynamic content, so it has a certain lag in changes in Internet content; cannot track the user's mobile status, and lacks mobility management between wireless cache servers; which content is suitable for delivery to wireless networks The edge is usually determined by third-party service providers. Operators only participate in scheduling management and operation and maintenance. The current situation of supply and demand separation often leads to problems such as low hit rate, poor user experience, and low return on investment.

Images