Multicast Overlay Network for Videoconferencing Scalability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing videoconferencing technologies face scalability issues due to the need for separate connections between each pair of users, leading to resource-intensive full-mesh connectivity and relay infrastructure overload, which limits the number of participants and affects call quality.
Innovation Solution
Implementing a multicast overlay network that uses an application layer-over-IP routing solution to distribute video streams, allowing multiple subscribers to receive a single video stream from a publisher through an ingress node, reducing the need for direct peer-to-peer connections and optimizing resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If full-mesh connectivity is implemented between all users, then direct peer-to-peer connection is achieved, but resource requirements grow proportionally to the square of the number of users
Solution Approach 1:
The system segments the full-mesh connectivity requirement by introducing multicast overlay network nodes that divide and distribute video streams to multiple subscribers simultaneously. Instead of requiring direct connections between all user pairs, the network segments the distribution task across multiple relay nodes, reducing individual user connection requirements from O(n²) to O(n).
Solution Approach 2:
Multicast overlay network nodes serve as intermediaries between video publishers and subscribers. These nodes receive video streams from publishers and forward them to multiple subscribers, eliminating the need for direct peer-to-peer connections between all user pairs while maintaining connection reliability.
2Ease of operation
If separate connections are established for each user pair, then direct communication is enabled, but the system does not scale to large numbers of participants
Solution Approach 1:
The system merges multiple individual connection requirements into a single multicast stream distribution. Instead of establishing separate connections for each user pair, the publisher establishes one connection to the multicast network, which then combines and distributes the stream to all subscribers efficiently.
Solution Approach 2:
The multicast overlay network provides universal service to all users in the conference, enabling a single video stream to reach multiple subscribers through the network infrastructure. This multi-functional approach allows the system to scale to large numbers of participants without requiring proportional increases in connection establishment overhead.
3Reliability
If relay infrastructure is used to ensure connectivity across NAT, then connection reliability is improved, but relay servers become overloaded and call quality suffers
Solution Approach 1:
The system creates multiple copies of the video stream at the multicast overlay network nodes, allowing efficient distribution to multiple subscribers. Instead of relaying the same stream through overloaded servers for each user pair, the network creates and distributes copies through its infrastructure, reducing relay server load while maintaining connection reliability.
Data Source
AI summary
A method of multicasting real-time video is described. The method begins by establishing a multicast network of machines capable of ingress, forwarding and broadcasting traffic, together with a mapping infrastructure. The multicast network preferably comprises a portion of an overlay network, such as a content delivery network (CDN). A video stream is published to the multicast network by (a) using the mapping infrastructure to find an ingress node in the multicast network, and then receiving the video stream from a publisher at the ingress node. One or more subscribers then subscribe to the video stream. In particular, and for subscriber, this subscription is carried out by (a) using the mapping infrastructure to find an egress node for the requesting client, and then delivering the video stream to the subscriber from the egress node. Preferably, the publisher and each subscriber use WebRTC to publish or consume the video stream, and video stream is consumed in a videoconference.