Edge Computing Caching Method Optimizing Delivery Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In edge computing networks, the limited computing and storage capacity of edge nodes compared to large data centers poses a challenge in minimizing average content delivery latency and maximizing user experience, especially with increasing mobile data traffic and IoT device access.
Innovation Solution
A caching method that ranks information data by popularity and distributes it across edge computing nodes based on available storage space and access weights, adjusting storage space sizes to optimize data distribution and reduce latency, using equations to calculate zone-wide popularity and access weights to determine optimal storage locations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If content is cached at edge computing nodes, then delivery latency is reduced, but storage capacity is limited
Solution Approach 1:
The system segments the network into multiple zones, each containing multiple sub-zones with edge computing nodes. Content is cached at different levels (local sub-zone and zone-shared storage) to distribute the storage burden while maintaining low-latency access. This hierarchical segmentation allows the system to overcome individual node storage limits by collectively utilizing distributed storage across multiple nodes.
Solution Approach 2:
The patent implements local sub-zone storage space in each edge computing node for frequently accessed content, while zone-shared storage space is distributed across multiple nodes. This local quality differentiation ensures that hot content is cached locally at the edge node for immediate access, while less frequently accessed content is stored in zone-shared space, optimizing both latency and storage utilization.
2Productivity
If more services are provided to local users, then user experience is improved, but computing resources are consumed
Solution Approach 1:
The system performs preliminary actions by proactively caching popular content at edge computing nodes before user requests arrive. The caching strategy pre-positiones content in local sub-zone and zone-shared storage spaces based on popularity metrics, so when users access the content, it is already available locally, reducing latency and computation resources during actual service delivery.
Solution Approach 2:
The edge computing nodes autonomously manage their own caching strategies by calculating popularity metrics and determining which content to cache in local versus zone-shared storage. This self-service capability allows nodes to optimize their resource allocation independently, providing services to local users efficiently without requiring centralized control of every decision.
3Quantity of substance
If storage space is increased at edge nodes, then more content can be cached, but device complexity increases
Solution Approach 1:
The patent merges the storage resources of multiple edge computing nodes into a collective zone-shared storage space. Instead of each node independently managing large storage capacity, the system combines storage across nodes, allowing the network to provide increased total storage space while individual nodes maintain simpler, more manageable storage configurations. This merging approach distributes complexity across the network rather than concentrating it at each node.
Data Source
AI summary
The application discloses caching method, system and device of edge computing and computer readable storage medium, wherein the edge computing network is divided into a plurality of zones each including a plurality of sub-zones. Each sub-zone corresponds to an edge computing node, which is divided into local sub-zone storage space and zone-shared storage space. Depending on the popularity for respective information data, they are stored into the local sub-zone storage space and zone-shared storage space in respective edge computing nodes, in order to optimize cooperative processing of user access requests across nodes, reduce the delivery latency, and maximize utilization of the processing capacity and storage space. Through constant adjustments to the size of local sub-zone storage space and changes to the distribution of information data, the storage distribution data at shortest average delivery latency is obtained, so as to achieve the minimum average delivery latency. (FIG. 1)