Hybrid Multicast P2P Video Streaming Server Load Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for delivering high-quality multimedia services, such as Video-On-Demand (VOD) streaming, face challenges including server overload, increased costs for large-scale implementations, and quality deterioration due to start delay and playback interruptions, especially in scenarios with varying user demand.
Innovation Solution
A hybrid data transmission method combining IP multicast and Peer-to-Peer (P2P) technology, where a server sends multicast data and supplemental assistant data through unicast, optimizing server load by adjusting transmission cycles based on user requests and P2P capacity, allowing users to request and receive lost data through unicast or P2P from available seeders.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If IP multicast is used to send content to multiple users simultaneously, then server load is reduced, but start delay time increases for users joining at different times
Solution Approach 1:
The system performs preliminary actions by having users download assistant data (missing video chunks) in advance through unicast before joining the multicast session. This allows users to quickly catch up to the live stream without causing server overload during the catching-up process, thus reducing start delay while maintaining low server load.
Solution Approach 2:
The video content is segmented into multiple chunks, with assistant data (missing chunks) separated from the main multicast stream. Users receive the main stream through multicast while assistant data is delivered through unicast or P2P, allowing independent optimization of each transmission path without mutual interference.
2Power
If P2P technology is used to reduce server load by having users share video content, then server overhead decreases, but service reliability deteriorates when users depart or P2P capacity is insufficient
Solution Approach 1:
The server acts as an intermediary that maintains control over the video content distribution. It provides assistant data directly to users who need it and manages the P2P sharing process, ensuring that content availability is guaranteed even when P2P participants depart, thus maintaining service reliability while still benefiting from reduced server overhead.
Solution Approach 2:
The system prepares assistant data in advance and makes it available through unicast transmission before P2P sharing is needed. This cushioning mechanism ensures that even if P2P capacity becomes insufficient or users depart, the server can immediately provide the required data through unicast, preventing playback interruptions.
3Reliability
If unicast is used to send assistant data to supplement multicast data, then service quality is maintained, but network bandwidth consumption increases
Solution Approach 1:
Users who have already received certain video chunks through multicast or P2P can serve other users by sharing their local copies via P2P communication. This self-service mechanism reduces the need for the server to send unicast data to all users, thereby maintaining service quality while significantly reducing overall network bandwidth consumption.
Solution Approach 2:
Instead of sending complete assistant data sets to all users through unicast, the system sends only the specific missing chunks that each user needs. This partial action approach maintains service quality by delivering exactly what is required while minimizing unnecessary bandwidth consumption from sending redundant data.
4Reliability
If CDN is deployed to deliver content from cache servers close to users, then service quality improves, but installation cost and complexity increase for small-scale operators
Solution Approach 1:
The system uses IP multicast, which is a universally available network feature supported by standard network infrastructure, to deliver video content. This eliminates the need for specialized CDN cache servers while still achieving efficient content distribution to multiple users simultaneously, thereby improving service quality without increasing installation complexity or costs.
Data Source
AI summary
Disclosed herein are a method and apparatus for delivering user-assisted multimedia data using periodic multicast, which provide stable high-capacity Video-On-Demand (VOD) streaming service by proposing a user-assisted video transfer scheme capable of replacing a client-server configuration in providing high-capacity VOD streaming service.