Peer Selection Cost Metric for ISP Traffic Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Peer-to-peer networks, such as BitTorrent, incur increased Internet Service Provider (ISP) expenses and power consumption due to random neighbor selection, leading to inefficient cross-ISP traffic patterns.
Innovation Solution
A method and apparatus that rank and generate neighbor lists based on a cost metric associated with the next ISP hop, prioritizing peers within the local ISP, followed by customer and peering ISPs, and then ranking external peers by the cost of the next ISP hop, to optimize traffic routing and reduce expenses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If random neighbor selection is used in peer-to-peer networks, then peer discovery is simplified and network flexibility is improved, but ISP expenses and power consumption increase due to inefficient cross-ISP traffic patterns
Solution Approach 1:
The patent applies local quality by making neighbor selection non-uniform: peers within the same ISP are prioritized over external peers, and among external peers, selection is based on cost metrics. This creates different selection probabilities for different peer groups, optimizing traffic patterns to reduce cross-ISP traffic and associated power consumption while maintaining network functionality
Solution Approach 2:
The patent changes the selection parameter from uniform random selection to cost-based selection. By introducing cost metrics associated with different ISPs and using these metrics to weight neighbor selection probabilities, the system transforms the neighbor selection process to favor lower-cost paths, thereby reducing overall network energy consumption
2Adaptability or versatility
If random neighbor selection is used in peer-to-peer networks, then peer discovery is simplified and network flexibility is improved, but ISP expenses increase due to inefficient cross-ISP traffic patterns
Solution Approach 1:
The patent applies local quality by making neighbor selection non-uniform: peers within the same ISP are prioritized over external peers, and among external peers, selection is based on cost metrics. This creates different selection probabilities for different peer groups, optimizing traffic patterns to reduce cross-ISP traffic and associated power consumption while maintaining network functionality
Solution Approach 2:
The patent changes the selection parameter from uniform random selection to cost-based selection. By introducing cost metrics associated with different ISPs and using these metrics to weight neighbor selection probabilities, the system transforms the neighbor selection process to favor lower-cost paths, thereby reducing overall network energy consumption
3Loss of energy
If cost-based neighbor selection is implemented, then ISP expenses and power consumption are reduced, but neighbor list generation complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-computing cost metrics for different ISPs and maintaining this cost information in lookup tables. When generating neighbor lists, the system retrieves pre-computed costs rather than calculating them in real-time, significantly reducing the computational complexity of neighbor selection while still achieving cost-optimized traffic patterns
Data Source
AI summary
A method and apparatus for peer-to-peer file sharing is provided. In some embodiments, the method includes receiving a request for a list of neighbor peers, where the request is made by a requesting peer device, and where the requesting peer device has a local internet service provider (ISP). The method may also include employing a server device to rank each neighbor peer in a plurality of neighbor peers based on whether the respective neighbor peer is external to the local ISP, and if the respective neighbor peer is external to the ISP, further based on a cost metric associated with a next ISP hop from the requesting peer device to the respective neighbor peer. The method may also include generating the list of neighbor peers based on the ranking of the neighbor peers, and enabling transmission of the list of neighbor peers to the requesting peer device.


