P2P CDN Scheduling Using Multi-Layer Position Information
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In P2P CDN technology, existing methods for determining the shortest content delivery transmission path between P2P nodes are not accurate, leading to increased transmission delays due to speculative distance calculations based on IP address similarities, which can result in longer routing distances and increased network overhead.
Innovation Solution
A scheduling method that uses position information from different network layers and service content to determine candidate P2P nodes capable of providing service, generating a list with transmission distances, and selecting the node with the shortest distance for content delivery, thereby optimizing the routing path.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If IP address similarity is used to speculate distances between P2P nodes, then the scheduling process is simplified, but the accuracy of distance determination deteriorates
Solution Approach 1:
The patent introduces position information from multiple network layers (IP address, MAC address, BRAS device identifier) as intermediaries to accurately represent the topological distance between P2P nodes. Instead of directly speculating distance from IP similarity, the system uses these layered position indicators to construct a more accurate distance metric, resolving the contradiction between simplification and accuracy.
Solution Approach 2:
The patent changes the parameters used for distance determination from simple IP address similarity to a multi-dimensional parameter set including IP address, MAC address, and BRAS device identifier. This parameter expansion allows for more precise distance calculation while maintaining computational feasibility through structured comparison of these hierarchical network identifiers.
2Speed
If speculative distance calculation based on IP address similarity is used, then the calculation speed is improved, but the transmission delay increases due to longer routing distances
Solution Approach 1:
The patent performs preliminary action by pre-establishing position information for all P2P nodes across multiple network layers before actual content delivery. This advance preparation allows for rapid distance calculation during content delivery without real-time complex computations, thus maintaining high calculation speed while achieving accurate shortest-path routing that reduces transmission delay.
Solution Approach 2:
The patent segments the distance calculation process into hierarchical layers (IP address level, MAC address level, BRAS device level), allowing for progressive refinement of distance metrics. This segmentation enables efficient calculation by first comparing coarse-grained IP addresses and then refining with more specific identifiers only when needed, balancing speed and accuracy.
3Device complexity
If IP address similarity-based distance speculation is used, then the system complexity is reduced, but network performance deteriorates due to increased network overhead
Solution Approach 1:
The patent makes the position information collection mechanism universal by gathering multiple types of network identifiers (IP address, MAC address, BRAS device identifier) through existing network infrastructure components. This multi-functionality allows the same collection mechanism to serve multiple purposes: distance calculation, node identification, and network topology mapping, thereby reducing overall system complexity while improving routing efficiency to decrease network overhead.
Data Source
AI summary
A scheduling method includes: a server receiving a first request of a first peer (P2P) node; the first request containing service content; using position information on different network hierarchies of the first P2P node and the service content to determine at least one candidate P2P node for the first P2P node; the at least one candidate P2P node can provide a service for the first P2P node; the position information being position information generated on the basis of a network topology structure; using the at least one candidate P2P node to generate a candidate P2P node information list; P2P node information in the information list representing a transmission distance between the at least one candidate P2P node and the first P2P node; and returning the information list to the first P2P node. Further disclosed at the same time are a server, a first P2P node and a computer-readable storage medium.